\input texinfo @c -*-texinfo-*- @c Copyright 1997-2001 TcX AB, Detron HB and MySQL Finland AB @c @c ********************************************************* @c Note that @node names are used on our Website. @c So do not change node names without checking @c Makefile.am and SitePages first. @c ********************************************************* @c @c %**start of header @c there's a better way to do this.. i just don't know it yet @c sed will remove the "@c ifnusphere " to make this valid @c ifnusphere @set nusphere 1 @setfilename mysql.info @c We want the types in the same index @syncodeindex tp fn @c Get version information. This file is generated by the Makefile!! @include include.texi @ifclear tex-debug @c This removes the black squares in the right margin @finalout @end ifclear @c Set background for HTML @set _body_tags BGCOLOR=silver TEXT=#000000 LINK=#101090 VLINK=#7030B0 @c Set some style elements for the manual in HTML form. 'suggested' @c natural language colors: aqua, black, blue, fuchsia, gray, green, @c lime, maroon, navy, olive, purple, red, silver, teal, white, and @c yellow. From Steeve Buehler <ahr@YogElements.com> @set _extra_head <style> code {color:purple} tt {color:green} samp {color:navy} pre {color:maroon} </style> @settitle MySQL Reference Manual for version @value{mysql_version}. @c We want single-sided heading format, with chapters on new pages. To @c get double-sided format change 'on' below to 'odd' @ifclear nusphere @setchapternewpage on @end ifclear @ifset nusphere @setchapternewpage odd @end ifset @paragraphindent 0 @ifset nusphere @smallbook @end ifset @c %**end of header @ifinfo @format START-INFO-DIR-ENTRY * mysql: (mysql). MySQL documentation. END-INFO-DIR-ENTRY @end format @end ifinfo @titlepage @sp 10 @center @titlefont{MySQL Reference Manual} @sp 10 @center Copyright @copyright{} 1997-2001 MySQL AB @c blank page after title page makes page 1 be a page front. @c also makes the back of the title page blank. @page @end titlepage @c Short contents, blank page, long contents. @c until i can figure out the blank page, no short contents. @c @shortcontents @c @page @c @page @contents @c This should be added. The HTML conversion also needs a MySQL version @c number somewhere. @iftex @c change this to double if you want formatting for double-sided @c printing @headings single @oddheading @thischapter @| @| @thispage @evenheading @thispage @| @| MySQL Technical Reference for Version @value{mysql_version} @end iftex @node Top, Introduction, (dir), (dir) @c @ifhtml @c <IMG SRC="Images/mysql-logo.gif"> @c <!--Image doesn't exist. Can't find suitable replacement. (Matt) --> @c @end ifhtml @ifinfo This is a manual for MySQL. This version is about the @value{mysql_version} version of MySQL. You can find a manual about any older version of MySQL in the binary or source distribution for that version. @end ifinfo @menu * Introduction:: General Information about MySQL * Installing:: Installing MySQL * Tutorial:: MySQL Tutorial * MySQL Database Administration:: MySQL Database Administration * MySQL Optimization:: MySQL Optimization * Reference:: MySQL language reference * Table types:: MySQL table types * Clients:: MySQL client tools and APIs * Extending MySQL:: Extending MySQL * Problems:: Problems * Users:: Some MySQL users * MySQL customer usage:: MySQL customer usage * Contrib:: Contributed programs * Credits:: Contributors to MySQL * News:: MySQL change history * Porting:: Comments on porting to other systems * Environment variables:: MySQL environment variables * Regexp:: Description of MySQL regular expression syntax * Unireg:: What is Unireg? * GPL license:: GNU General Public License * LGPL license:: GNU Library General Public License * Placeholder:: Pieces of the manual in transit * Function Index:: SQL command, type and function index * Concept Index:: Concept Index @end menu @node Introduction, Installing, Top, Top @chapter General Information About MySQL @cindex overview @cindex general information @cindex online location of manual @cindex manual, online location This is the MySQL reference manual; it documents MySQL Version @value{mysql_version}. As MySQL is work in progress, the manual gets updated frequently. There is a very good chance that this version is out of date, unless you are looking at it online. The most recent version of this manual is available at @uref{http://www.mysql.com/documentation/} in many different formats. If you have a hard time finding information in the manual, you can try the searchable PHP version at @uref{http://www.mysql.com/documentation/manual.php}. MySQL is a very fast, multi-threaded, multi-user, and robust SQL (Structured Query Language) database server. MySQL is free software. It is licensed with the @strong{GNU GENERAL PUBLIC LICENSE} @uref{http://www.gnu.org/}. @xref{Licensing and Support}. @uref{http://www.mysql.com/, The MySQL home page} provides the latest information about MySQL. The following list describes some useful sections of the manual: @itemize @bullet @item For information about the company behind MySQL, see @ref{What is MySQL AB}. @item For a discussion of MySQL's capabilities, see @ref{Features}. @item For installation instructions, see @ref{Installing}. @item For tips on porting MySQL to new architectures or operating systems, see @ref{Porting}. @item For information about upgrading from a Version 3.22 release, see @ref{Upgrading-from-3.22}. @item For a tutorial introduction to MySQL, see @ref{Tutorial}. @item For examples of SQL and benchmarking information, see the benchmarking directory (@file{sql-bench} in the distribution). @item For a history of new features and bug fixes, see @ref{News}. @item For a list of currently known bugs and misfeatures, see @ref{Bugs}. @item For future plans, see @ref{TODO}. @item For a list of all the contributors to this project, see @ref{Credits}. @end itemize @strong{IMPORTANT:} Reports of errors (often called bugs), as well as questions and comments, should be sent to the mailing list at @email{mysql@@lists.mysql.com}. @xref{Bug reports}. The @code{mysqlbug} script should be used to generate bug reports. @c FIX! RPMs are also binary For source distributions, the @code{mysqlbug} script can be found in the @file{scripts} directory. For binary distributions, @code{mysqlbug} can be found in the @file{bin} directory. If you have found a sensitive security bug in MySQL, you should send an email to @email{security@@mysql.com}. @cindex errors, reporting @cindex reporting, errors @cindex mailing list address @cindex @code{mysqlbug} script, location If you have any suggestions concerning additions or corrections to this manual, please send them to the manual team at @email{docs@@mysql.com}. This is a reference manual; it does not provide general instruction on SQL or relational database concepts. If you want general information about SQL, see @ref{General-SQL}. For books that focus more specifically on MySQL, see @ref{MySQL-Books}. @menu * MySQL and MySQL AB:: MySQL, MySQL AB, and Open Source * MySQL Information Sources:: MySQL Information Sources * Licensing and Support:: MySQL Licensing and Support * Compatibility:: How Standards-compatible Is MySQL? * Comparisons:: How MySQL Compares to Other Databases * TODO:: MySQL and the future (The TODO) @end menu @node MySQL and MySQL AB, MySQL Information Sources, Introduction, Introduction @section MySQL, MySQL AB, and Open Source @menu * What-is:: What is MySQL? * What is MySQL AB:: What is MySQL AB? * Manual-info:: About this manual * Manual conventions:: Conventions used in this manual * History:: History of MySQL * Features:: The main features of MySQL * Stability:: How stable is MySQL? * Table size:: How Big Can MySQL Tables Be? * Year 2000 compliance:: Year 2000 compliance @end menu @node What-is, What is MySQL AB, MySQL and MySQL AB, MySQL and MySQL AB @subsection What Is MySQL @cindex MySQL, defined @cindex MySQL, introduction MySQL, the most popular Open Source SQL database, is provided by MySQL AB. MySQL AB is a commercial company that builds its business providing services around the MySQL database. @xref{What is MySQL AB}. @table @asis @item MySQL is a database management system. A database is a structured collection of data. It may be anything from a simple shopping list to a picture gallery or the vast amounts of information in a corporate network. To add, access, and process data stored in a computer database, you need a database management system such as MySQL. Since computers are very good at handling large amounts of data, database management plays a central role in computing, as stand-alone utilities, or as parts of other applications. @cindex databases, defined @item MySQL is a relational database management system. A relational database stores data in separate tables rather than putting all the data in one big storeroom. This adds speed and flexibility. The tables are linked by defined relations making it possible to combine data from several tables on request. The SQL part of MySQL stands for "Structured Query Language" - the most common standardized language used to access databases. @cindex relational databases, defined @cindex SQL, defined @item MySQL is Open Source Software. Open Source means that it is possible for anyone to use and modify. Anybody can download MySQL from the Internet and use it without paying anything. Anybody so inclined can study the source code and change it to fit their needs. MySQL uses the GPL (GNU General Public License) @uref{http://www.gnu.org}, to define what you may and may not do with the software in different situations. If you feel uncomfortable with the GPL or need to embed MySQL into a commercial application you can buy a commercially licensed version from us. @cindex open source, defined @cindex General Public License @item Why use MySQL? MySQL is very fast, reliable, and easy to use. If that is what you are looking for, you should give it a try. MySQL also has a very practical set of features developed in very close cooperation with our users. You can find a performance comparison of MySQL to some other database managers on our benchmark page. @xref{MySQL Benchmarks}. MySQL was originally developed to handle very large databases much faster than existing solutions and has been successfully used in highly demanding production environments for several years. Though under constant development, MySQL today offers a rich and very useful set of functions. The connectivity, speed, and security make MySQL highly suited for accessing databases on the Internet. @item The technical features of MySQL For advanced technical information, see @ref{Reference}. MySQL is a client/server system that consists of a multi-threaded SQL server that supports different backends, several different client programs and libraries, administrative tools, and several programming interfaces. We also provide MySQL as a multi-threaded library which you can link into your application to get a smaller, faster, easier to manage product. @item MySQL has a lot of contributed software available. It is very likely that you will find that your favorite application or language already supports MySQL. @end table @cindex pronunciation, MySQL @cindex MySQL, pronunciation @cindex goals of MySQL The official way to pronounce MySQL is ``My Ess Que Ell'' (not MY-SEQUEL). But we try to avoid correcting people who say MY-SEQUEL. @node What is MySQL AB, Manual-info, What-is, MySQL and MySQL AB @subsection What Is MySQL AB @cindex MySQL AB, defined MySQL AB is the Swedish company owned and run by the MySQL founders and main developers. We are dedicated to developing MySQL and spreading our database to new users. MySQL AB owns the copyright to the MySQL server source code and the MySQL trademark. A significant amount of revenues from our services goes to developing MySQL. @xref{What-is}. MySQL AB has been profitable providing MySQL from the start. We don't get any outside funding, but have earned all our money ourselves. We are searching after partners that would like to support our development of MySQL so that we could accelerate the development pace. If you are interested in doing this, you can email @email{partner@@mysql.com} about this! MySQL AB has currently 20+ people (@uref{http://www.mysql.com/development/team.html}) on its payroll and is growing rapidly. Our main sources of income are: @itemize @bullet @item Commercial high quality support for MySQL provided by the MySQL developers themselves. If you are interested in purchasing a support contract, please visit @uref{https://order.mysql.com/} to view our support options or to order support. @item Consulting services. We have developers and consultants in 12 countries and partners in many other countries that can help you with almost any MySQL related issues. If you need consulting services, please email a good description of your needs to @email{info@@mysql.com}! If we can't handle this ourselves we can usually find a partner or a developer that can help you with your problems. @item We sell licenses for using MySQL as an embedded database. @xref{Cost}. If you have a commercial product for which you need a fast, high quality database, but you can't afford to make your product Open Source, you can buy the right to use the MySQL server under a normal commercial copyright. If you are interested in this you can buy MySQL licenses at @uref{https://order.mysql.com/} or contact us at @email{licensing@@mysql.com}. @item Advertising. @uref{http://www.mysql.com/} is a very popular web site with more than 10,000,000 page views per months (January 2001). By putting a banner on this you are guaranteed to reach a lot of potential customers in the Open source, Linux and database community. If you are interested in this email @email{advertising@@mysql.com}. @item We are building a partner program to be able to provide MySQL services in every country. If you are interested in becoming a partner of MySQL AB please visit @uref{http://www.mysql.com/information/partners.html} or email @email{partner@@mysql.com}. @item We provide MySQL training through our partner programs. For more information, please email @email{info@@mysql.com}. @item The MySQL brand has, since 1995, been associated with speed and reliability, and is known to be something you can depend upon. If you are interested in using the MySQL trademark in your marketing, you can email @email{info@@mysql.com} about this. @end itemize The MySQL core values show our dedication to MySQL and Open Source. We want MySQL to be: @itemize @bullet @item The best and the most used database in the world. @item Available and affordable for all. @item Easy to use. @item Continuously improved while remaining fast and safe. @item Fun to use and improve. @item Free from bugs. @end itemize MySQL AB and the people of MySQL AB: @itemize @bullet @item Promote Open Source Philosophy and support the Open Source Community. @item Aim to be good citizens. @item Prefer partners that share our values and mind-set. @item Answer mail and give support. @item Are a virtual company, networking with others. @item Work against software patents. @end itemize @node Manual-info, Manual conventions, What is MySQL AB, MySQL and MySQL AB @subsection About This Manual This manual is currently available in Texinfo, plain text, Info, HTML, PostScript, and PDF versions. The primary document is the Texinfo file. The HTML version is produced automatically using a modified version of @code{texi2html}. The plain text and Info versions are produced with @code{makeinfo}. The Postscript version is produced using @code{texi2dvi} and @code{dvips}. The PDF version is produced with @code{pdftex}. @cindex manual, available formats @cindex Texinfo This manual is written and maintained by David Axmark, Michael (Monty) Widenius, Jeremy Cole, and Paul DuBois. For other contributors, see @ref{Credits}. @node Manual conventions, History, Manual-info, MySQL and MySQL AB @subsection Conventions Used in This Manual This manual uses certain typographical conventions: @cindex manual, typographical conventions @cindex typographical conventions @cindex conventions, typographical @table @asis @item @code{constant} Constant-width font is used for command names and options; SQL statements; database, table and column names; C and Perl code; and environment variables. Example: ``To see how @code{mysqladmin} works, invoke it with the @code{--help} option.'' @item @file{filename} Constant-width font with surrounding quotes is used for filenames and pathnames. Example: ``The distribution is installed under the @file{/usr/local/} directory.'' @item @samp{c} Constant-width font with surrounding quotes is also used to indicate character sequences. Example: ``To specify a wild card, use the @samp{%} character.'' @item @emph{italic} Italic font is used for emphasis, @emph{like this}. @item @strong{boldface} Boldface font is used for access privilege names (for example, ``do not grant the @strong{process} privilege lightly'') and occasionally to convey @strong{especially strong emphasis}. @end table When commands are shown that are meant to be executed by a particular program, the program is indicated by a prompt shown before the command. For example, @code{shell>} indicates a command that you execute from your login shell, and @code{mysql>} indicates a command that you execute from the @code{mysql} client program: @example shell> type a shell command here mysql> type a mysql command here @end example Shell commands are shown using Bourne shell syntax. If you are using a @code{csh}-style shell, you may need to issue commands slightly differently. For example, the sequence to set an environment variable and run a command looks like this in Bourne shell syntax: @example shell> VARNAME=value some_command @end example @cindex shell syntax @cindex command syntax For @code{csh}, you would execute the sequence like this: @example shell> setenv VARNAME value shell> some_command @end example Often, database, table, and column names must be substituted into commands. To indicate that such substitution is necessary, this manual uses @code{db_name}, @code{tbl_name} and @code{col_name}. For example, you might see a statement like this: @example mysql> SELECT col_name FROM db_name.tbl_name; @end example This means that if you were to enter a similar statement, you would supply your own database, table, and column names, perhaps like this: @example mysql> SELECT author_name FROM biblio_db.author_list; @end example SQL statements may be written in uppercase or lowercase. When this manual shows a SQL statement, uppercase is used for particular keywords if those keywords are under discussion (to emphasize them) and lowercase is used for the rest of the statement. For example, you might see the following in a discussion of the @code{SELECT} statement: @example mysql> SELECT count(*) FROM tbl_name; @end example On the other hand, in a discussion of the @code{COUNT()} function, the same statement would be written like this: @example mysql> select COUNT(*) from tbl_name; @end example If no particular emphasis is intended, all keywords are written uniformly in uppercase. In syntax descriptions, square brackets (@samp{[} and @samp{]}) are used to indicate optional words or clauses: @example DROP TABLE [IF EXISTS] tbl_name @end example When a syntax element consists of a number of alternatives, the alternatives are separated by vertical bars (@samp{|}). When one member from a set of choices @strong{may} be chosen, the alternatives are listed within square brackets (@samp{[} and @samp{]}): @example TRIM([[BOTH | LEADING | TRAILING] [remstr] FROM] str) @end example When one member from a set of choices @strong{must} be chosen, the alternatives are listed within braces (@samp{@{} and @samp{@}}): @example @{DESCRIBE | DESC@} tbl_name @{col_name | wild@} @end example @node History, Features, Manual conventions, MySQL and MySQL AB @subsection History of MySQL @cindex MySQL history @cindex history of MySQL @cindex MySQL name @cindex My, derivation We once started out with the intention of using @code{mSQL} to connect to our tables using our own fast low-level (ISAM) routines. However, after some testing we came to the conclusion that @code{mSQL} was not fast enough nor flexible enough for our needs. This resulted in a new SQL interface to our database but with almost the same API interface as @code{mSQL}. This API was chosen to ease porting of third-party code. The derivation of the name MySQL is not perfectly clear. Our base directory and a large number of our libraries and tools have had the prefix ``my'' for well over 10 years. However, Monty's daughter (some years younger) is also named My. Which of the two gave its name to MySQL is still a mystery, even for us. @node Features, Stability, History, MySQL and MySQL AB @subsection The Main Features of MySQL @cindex main features of MySQL @cindex features of MySQL The following list describes some of the important characteristics of MySQL: @c This list is too technical and should be divided into one feature @c list comparable to commercial competition and a very technical on @c with max limits (from crash-me) and so on. @itemize @bullet @item Fully multi-threaded using kernel threads. This means it can easily use multiple CPUs if available. @item C, C++, Eiffel, Java, Perl, PHP, Python and Tcl APIs. @xref{Clients}. @item Works on many different platforms. @xref{Which OS}. @item Many column types: signed/unsigned integers 1, 2, 3, 4, and 8 bytes long, @code{FLOAT}, @code{DOUBLE}, @code{CHAR}, @code{VARCHAR}, @code{TEXT}, @code{BLOB}, @code{DATE}, @code{TIME}, @code{DATETIME}, @code{TIMESTAMP}, @code{YEAR}, @code{SET}, and @code{ENUM} types. @xref{Column types}. @item Very fast joins using an optimized one-sweep multi-join. @item Full operator and function support in the @code{SELECT} and @code{WHERE} parts of queries. For example: @example mysql> SELECT CONCAT(first_name, " ", last_name) FROM tbl_name WHERE income/dependents > 10000 AND age > 30; @end example @item SQL functions are implemented through a highly optimized class library and should be as fast as possible! Usually there isn't any memory allocation at all after query initialization. @item Full support for SQL @code{GROUP BY} and @code{ORDER BY} clauses. Support for group functions (@code{COUNT()}, @code{COUNT(DISTINCT ...)}, @code{AVG()}, @code{STD()}, @code{SUM()}, @code{MAX()} and @code{MIN()}). @item Support for @code{LEFT OUTER JOIN} and @code{RIGHT OUTER JOIN} with ANSI SQL and ODBC syntax. @item You can mix tables from different databases in the same query (as of Version 3.22). @item A privilege and password system that is very flexible and secure, and allows host-based verification. Passwords are secure because all password traffic is encrypted when you connect to a server. @item ODBC (Open-DataBase-Connectivity) support for Win32 (with source). All ODBC 2.5 functions and many others. For example, you can use MS Access to connect to your MySQL server. @xref{ODBC}. @item Very fast B-tree disk tables with index compression. @item Up to 32 indexes per table are allowed. Each index may consist of 1 to 16 columns or parts of columns. The maximum index length is 500 bytes (this may be changed when compiling MySQL). An index may use a prefix of a @code{CHAR} or @code{VARCHAR} field. @item Fixed-length and variable-length records. @item In-memory hash tables which are used as temporary tables. @item Handles large databases. We are using MySQL with some databases that contain 50,000,000 records and we know of users that uses MySQL with 60,000 tables and about 5,000,000,000 rows @item All columns have default values. You can use @code{INSERT} to insert a subset of a table's columns; those columns that are not explicitly given values are set to their default values. @item Uses GNU Automake, Autoconf, and Libtool for portability. @item Written in C and C++. Tested with a broad range of different compilers. @item A very fast thread-based memory allocation system. @item No memory leaks. MySQL has been tested with Purify, a commercial memory leakage detector. @item Includes @code{myisamchk}, a very fast utility for table checking, optimization, and repair. All of the functionality of @code{myisamchk} is also available through the SQL interface as well. @xref{MySQL Database Administration}. @item Full support for several different character sets, including ISO-8859-1 (Latin1), big5, ujis, and more. For example, the Scandinavian characters `@ringaccent{a}', `@"a' and `@"o' are allowed in table and column names. @item All data are saved in the chosen character set. All comparisons for normal string columns are case insensitive. @item Sorting is done according to the chosen character set (the Swedish way by default). It is possible to change this when the MySQL server is started up. To see an example of very advanced sorting, look at the Czech sorting code. MySQL supports many different character sets that can be specified at compile and run time. @item Aliases on tables and columns are allowed as in the SQL92 standard. @item @code{DELETE}, @code{INSERT}, @code{REPLACE}, and @code{UPDATE} return the number of rows that were changed (affected). It is possible to return the number of rows matched instead by setting a flag when connecting to the server. @item Function names do not clash with table or column names. For example, @code{ABS} is a valid column name. The only restriction is that for a function call, no spaces are allowed between the function name and the @samp{(} that follows it. @xref{Reserved words}. @item All MySQL programs can be invoked with the @code{--help} or @code{-?} options to obtain online assistance. @item The server can provide error messages to clients in many languages. @xref{Languages}. @item Clients may connect to the MySQL server using TCP/IP Sockets, Unix Sockets (Unix), or Named Pipes (NT). @item The MySQL-specific @code{SHOW} command can be used to retrieve information about databases, tables, and indexes. The @code{EXPLAIN} command can be used to determine how the optimizer resolves a query. @end itemize @node Stability, Table size, Features, MySQL and MySQL AB @subsection How Stable Is MySQL? @cindex stability This section addresses the questions ``How stable is MySQL?'' and ``Can I depend on MySQL in this project?'' We will try to clarify some issues and to answer some of the more important questions that seem to concern many people. This section has been put together from information gathered from the mailing list (which is very active in reporting bugs). At TcX, MySQL has worked without any problems in our projects since mid-1996. When MySQL was released to a wider public, we noticed that there were some pieces of ``untested code'' that were quickly found by the new users who made queries in a manner different than our own. Each new release has had fewer portability problems than the previous one (even though each has had many new features). Each release of MySQL has been usable, and there have been problems only when users start to use code from the ``gray zones.'' Naturally, outside users don't know what the gray zones are; this section attempts to indicate those that are currently known. The descriptions deal with Version 3.23 of MySQL. All known and reported bugs are fixed in the latest version, with the exception of the bugs listed in the bugs section, which are things that are design-related. @xref{Bugs}. MySQL is written in multiple layers and different independent modules. These modules are listed below with an indication of how well-tested each of them is: @cindex modules, list of @table @strong @item The ISAM table handler --- Stable This manages storage and retrieval of all data in MySQL Version 3.22 and earlier. In all MySQL releases there hasn't been a single (reported) bug in this code. The only known way to get a corrupted table is to kill the server in the middle of an update. Even that is unlikely to destroy any data beyond rescue, because all data are flushed to disk between each query. There hasn't been a single bug report about lost data because of bugs in MySQL. @cindex ISAM table handler @cindex storing, data @cindex retrieving, data @cindex data, ISAM table handler @item The MyISAM table handler --- Stable This is new in MySQL Version 3.23. It's largely based on the ISAM table code but has a lot of new and very useful features. @item The parser and lexical analyser --- Stable There hasn't been a single reported bug in this system for a long time. @item The C client code --- Stable No known problems. In early Version 3.20 releases, there were some limitations in the send/receive buffer size. As of Version 3.21, the buffer size is now dynamic up to a default of 16M. @item Standard client programs --- Stable These include @code{mysql}, @code{mysqladmin}, @code{mysqlshow}, @code{mysqldump}, and @code{mysqlimport}. @item Basic SQL --- Stable The basic SQL function system and string classes and dynamic memory handling. Not a single reported bug in this system. @item Query optimizer --- Stable @item Range optimizer --- Stable @item Join optimizer --- Stable @item Locking --- Gamma This is very system-dependent. On some systems there are big problems using standard OS locking (@code{fcntl()}). In these cases, you should run the MySQL daemon with the @code{--skip-locking} flag. Problems are known to occur on some Linux systems, and on SunOS when using NFS-mounted file systems. @item Linux threads --- Stable The major problem found has been with the @code{fcntl()} call, which is fixed by using the @w{@code{--skip-locking}} option to @code{mysqld}. Some people have reported lockup problems with Version 0.5. LinuxThreads will need to be recompiled if you plan to use 1000+ concurrent connections. Although it is possible to run that many connections with the default LinuxThreads (however, you will never go above 1021), the default stack spacing of 2 MB makes the application unstable, and we have been able to reproduce a coredump after creating 1021 idle connections. @xref{Linux}. @item Solaris 2.5+ pthreads --- Stable We use this for all our production work. @item MIT-pthreads (Other systems) --- Stable There have been no reported bugs since Version 3.20.15 and no known bugs since Version 3.20.16. On some systems, there is a ``misfeature'' where some operations are quite slow (a 1/20 second sleep is done between each query). Of course, MIT-pthreads may slow down everything a bit, but index-based @code{SELECT} statements are usually done in one time frame so there shouldn't be a mutex locking/thread juggling. @item Other thread implementions --- Beta - Gamma The ports to other systems are still very new and may have bugs, possibly in MySQL, but most often in the thread implementation itself. @item @code{LOAD DATA ...}, @code{INSERT ... SELECT} --- Stable Some people thought they had found bugs here, but these usually have turned out to be misunderstandings. Please check the manual before reporting problems! @item @code{ALTER TABLE} --- Stable Small changes in Version 3.22.12. @item DBD --- Stable Now maintained by Jochen Wiedmann (@email{wiedmann@@neckar-alb.de}). Thanks! @item @code{mysqlaccess} --- Stable Written and maintained by Yves Carlier (@email{Yves.Carlier@@rug.ac.be}). Thanks! @item @code{GRANT} --- Stable Big changes made in MySQL Version 3.22.12. @item @strong{MyODBC} (uses ODBC SDK 2.5) --- Gamma It seems to work well with some programs. @item Replication -- Beta / Gamma We are still working on replication, so don't expect this to be rock solid yet. On the other hand, some MySQL users are already using this with good results. @item BDB Tables -- Beta The Berkeley DB code is very stable, but we are still improving the interface between MySQL and BDB tables, so it will take some time before this is as tested as the other table types. @item InnoDB Tables -- Beta This is a recent addition to @code{MySQL}. They appear to work good and can be used after some initial testing. @item Automatic recovery of MyISAM tables - Beta This only affects the new code that checks if the table was closed properly on open and executes an automatic check/repair of the table if it wasn't. @item MERGE tables -- Beta / Gamma The usage of keys on @code{MERGE} tables is still not that tested. The other part of the @code{MERGE} code is quite well tested. @item FULLTEXT -- Beta Text search seems to work, but is still not widely used. @end table MySQL AB provides e-mail support for paying customers, but the MySQL mailing list usually provides answers to common questions. Bugs are usually fixed right away with a patch; for serious bugs, there is almost always a new release. @node Table size, Year 2000 compliance, Stability, MySQL and MySQL AB @subsection How Big Can MySQL Tables Be? @cindex tables, maximum size @cindex size of tables @cindex operating systems, file size limits @cindex limits, file size @cindex files, size limits MySQL Version 3.22 has a 4G limit on table size. With the new @code{MyISAM} in MySQL Version 3.23 the maximum table size is pushed up to 8 million terabytes (2 ^ 63 bytes). Note, however, that operating systems have their own file size limits. Here are some examples: @multitable @columnfractions .5 .5 @item @strong{Operating System} @tab @strong{File Size Limit} @item Linux-Intel 32 bit @tab 2G, 4G or more, depends on Linux version @item Linux-Alpha @tab 8T (?) @item Solaris 2.5.1 @tab 2G (possible 4G with patch) @item Solaris 2.6 @tab 4G @item Solaris 2.7 Intel @tab 4G @item Solaris 2.7 ULTRA-SPARC @tab 8T (?) @end multitable On Linux 2.2 you can get bigger tables than 2G by using the LFS patch for the ext2 file system. On Linux 2.4 there exists also patches for ReiserFS to get support for big files. This means that the table size for MySQL is normally limited by the operating system. By default, MySQL tables have a maximum size of about 4G. You can check the maximum table size for a table with the @code{SHOW TABLE STATUS} command or with the @code{myisamchk -dv table_name}. @xref{SHOW}. If you need bigger tables than 4G (and your operating system supports this), you should set the @code{AVG_ROW_LENGTH} and @code{MAX_ROWS} parameter when you create your table. @xref{CREATE TABLE}. You can also set these later with @code{ALTER TABLE}. @xref{ALTER TABLE}. If your big table is going to be read-only, you could use @code{myisampack} to merge and compress many tables to one. @code{myisampack} usually compresses a table by at least 50%, so you can have, in effect, much bigger tables. @xref{myisampack, , @code{myisampack}}. You can go around the operating system file limit for @code{MyISAM} data files by using the @code{RAID} option. @xref{CREATE TABLE}. Another solution can be the included MERGE library, which allows you to handle a collection of identical tables as one. @xref{MERGE, MERGE tables}. @node Year 2000 compliance, , Table size, MySQL and MySQL AB @subsection Year 2000 Compliance @cindex Year 2000 compliance @cindex compliance, Y2K @cindex date functions, Y2K compliance MySQL itself has no problems with Year 2000 (Y2K) compliance: @itemize @bullet @item MySQL uses Unix time functions and has no problems with dates until @code{2069}; all 2-digit years are regarded to be in the range @code{1970} to @code{2069}, which means that if you store @code{01} in a @code{year} column, MySQL treats it as @code{2001}. @item All MySQL date functions are stored in one file @file{sql/time.cc} and coded very carefully to be year 2000-safe. @item In MySQL Version 3.22 and later, the new @code{YEAR} column type can store years @code{0} and @code{1901} to @code{2155} in 1 byte and display them using 2 or 4 digits. @end itemize You may run into problems with applications that use MySQL in a way that is not Y2K-safe. For example, many old applications store or manipulate years using 2-digit values (which are ambiguous) rather than 4-digit values. This problem may be compounded by applications that use values such as @code{00} or @code{99} as ``missing'' value indicators. Unfortunately, these problems may be difficult to fix, because different applications may be written by different programmers, each of whom may use a different set of conventions and date-handling functions. Here is a simple demonstration illustrating that MySQL doesn't have any problems with dates until the year 2030: @example mysql> DROP TABLE IF EXISTS y2k; Query OK, 0 rows affected (0.01 sec) mysql> CREATE TABLE y2k (date date, date_time datetime, time_stamp timestamp); Query OK, 0 rows affected (0.00 sec) mysql> INSERT INTO y2k VALUES -> ("1998-12-31","1998-12-31 23:59:59",19981231235959), -> ("1999-01-01","1999-01-01 00:00:00",19990101000000), -> ("1999-09-09","1999-09-09 23:59:59",19990909235959), -> ("2000-01-01","2000-01-01 00:00:00",20000101000000), -> ("2000-02-28","2000-02-28 00:00:00",20000228000000), -> ("2000-02-29","2000-02-29 00:00:00",20000229000000), -> ("2000-03-01","2000-03-01 00:00:00",20000301000000), -> ("2000-12-31","2000-12-31 23:59:59",20001231235959), -> ("2001-01-01","2001-01-01 00:00:00",20010101000000), -> ("2004-12-31","2004-12-31 23:59:59",20041231235959), -> ("2005-01-01","2005-01-01 00:00:00",20050101000000), -> ("2030-01-01","2030-01-01 00:00:00",20300101000000), -> ("2050-01-01","2050-01-01 00:00:00",20500101000000); Query OK, 13 rows affected (0.01 sec) Records: 13 Duplicates: 0 Warnings: 0 mysql> SELECT * FROM y2k; +------------+---------------------+----------------+ | date | date_time | time_stamp | +------------+---------------------+----------------+ | 1998-12-31 | 1998-12-31 23:59:59 | 19981231235959 | | 1999-01-01 | 1999-01-01 00:00:00 | 19990101000000 | | 1999-09-09 | 1999-09-09 23:59:59 | 19990909235959 | | 2000-01-01 | 2000-01-01 00:00:00 | 20000101000000 | | 2000-02-28 | 2000-02-28 00:00:00 | 20000228000000 | | 2000-02-29 | 2000-02-29 00:00:00 | 20000229000000 | | 2000-03-01 | 2000-03-01 00:00:00 | 20000301000000 | | 2000-12-31 | 2000-12-31 23:59:59 | 20001231235959 | | 2001-01-01 | 2001-01-01 00:00:00 | 20010101000000 | | 2004-12-31 | 2004-12-31 23:59:59 | 20041231235959 | | 2005-01-01 | 2005-01-01 00:00:00 | 20050101000000 | | 2030-01-01 | 2030-01-01 00:00:00 | 20300101000000 | | 2050-01-01 | 2050-01-01 00:00:00 | 00000000000000 | +------------+---------------------+----------------+ 13 rows in set (0.00 sec) @end example This shows that the @code{DATE} and @code{DATETIME} types will not give any problems with future dates (they handle dates until the year 9999). The @code{TIMESTAMP} type, which is used to store the current time, has a range up to only @code{2030-01-01}. @code{TIMESTAMP} has a range of @code{1970} to @code{2030} on 32-bit machines (signed value). On 64-bit machines it handles times up to @code{2106} (unsigned value). Even though MySQL is Y2K-compliant, it is your responsibility to provide unambiguous input. See @ref{Y2K issues} for MySQL's rules for dealing with ambiguous date input data (data containing 2-digit year values). @node MySQL Information Sources, Licensing and Support, MySQL and MySQL AB, Introduction @section MySQL Information Sources @cindex manuals, about MySQL @cindex books, about MySQL @menu * MySQL-Books:: Books About MySQL * General-SQL:: General SQL Information and Tutorials * Useful Links:: Useful MySQL-related Links * Questions:: MySQL Mailing Lists @end menu @node MySQL-Books, General-SQL, MySQL Information Sources, MySQL Information Sources @subsection Books About MySQL For the latest book information, with user comments, please visit @uref{http://www.mysql.com/portal/books/html/index.html}. While this manual is still the right place for up to date technical information, its primary goal is to contain everything there is to know about MySQL. It is sometimes nice to have a bound book to read in bed or while you travel. Here is a list of books about MySQL and related subjects (in English). By purchasing a book through these hyperlinks provided herein, you are contributing to the development of MySQL. @emph{MySQL} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0735709211&bfmtype=book, Barnes and Noble} @item Publisher @tab New Riders @item Author @tab Paul DuBois @item Pub Date @tab 1st Edition December 1999 @item ISBN @tab 0735709211 @item Pages @tab 800 @item Price @tab $49.99 US @item Downloadable examples @tab @uref{http://www.kitebird.com/mysql-book/, @code{samp_db} distribution} @item Errata @tab @uref{http://www.kitebird.com/mysql-book/errata.html, are available here} @end multitable Foreword by Michael ``Monty'' Widenius, MySQL Moderator. @* In @emph{MySQL}, Paul DuBois provides you with a comprehensive guide to one of the most popular relational database systems. Paul has contributed to the online documentation for MySQL and is an active member of the MySQL community. The principal MySQL developer, Monty Widenius, and a network of his fellow developers reviewed the manuscript, and provided Paul with the kind of insight no one else could supply. @* Instead of merely giving you a general overview of MySQL, Paul teaches you how to make the most of its capabilities. Through two sample database applications that run throughout the book, he gives you solutions to problems you're sure to face. He helps you integrate MySQL efficiently with third-party tools, such as PHP and Perl, enabling you to generate dynamic Web pages through database queries. He teaches you to write programs that access MySQL databases, and also provides a comprehensive set of references to column types, operators, functions, SQL syntax, MySQL programming, C API, Perl @code{DBI}, and PHP API. @emph{MySQL} simply gives you the kind of information you won't find anywhere else. @* If you use MySQL, this book provides you with: @itemize @bullet @item An introduction to MySQL and SQL. @item Coverage of MySQL's data types and how to use them. @item Thorough treatment of how to write client programs in C. @item A guide to using the Perl @code{DBI} and PHP APIs for developing command-line and Web-based applications. @item Tips on administrative issues such as user accounts, backup, crash recovery, and security. @item Help in choosing an ISP for MySQL access. @item A comprehensive reference for MySQL's data types, operators, functions, and SQL statements and utilities. @item Complete reference guides for MySQL's C API, the Perl @code{DBI} API, and PHP's MySQL-related functions. @end itemize @* @emph{MySQL & mSQL} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=1565924347&bfmtype=book, Barnes and Noble} @item Publisher @tab O'Reilly @item Authors @tab Randy Jay Yarger, George Reese & Tim King @item Pub Date @tab 1st Edition July 1999 @item ISBN @tab 1-56592-434-7, Order Number: 4347 @item Pages @tab 506 @item Price @tab $34.95 @end multitable This book teaches you how to use MySQL and @code{mSQL}, two popular and robust database products that support key subsets of SQL on both Linux and Unix systems. Anyone who knows basic C, Java, Perl, or Python can write a program to interact with a database, either as a stand-alone application or through a Web page. This book takes you through the whole process, from installation and configuration to programming interfaces and basic administration. Includes plenty of tutorial material. @* @emph{Sams' Teach Yourself MySQL in 21 Days} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0672319144&bfmtype=book, Barnes and Noble} @item Publisher @tab Sams @item Authors @tab Mark Maslakowski and Tony Butcher @item Pub Date @tab June 2000 @item ISBN @tab 0672319144 @item Pages @tab 650 @item Price @tab $39.99 @end multitable Sams' @emph{Teach Yourself MySQL in 21 Days} is for intermediate Linux users who want to move into databases. A large share of the audience is Web developers who need a database to store large amounts of information that can be retrieved via the Web. Sams' @emph{Teach Yourself MySQL in 21 Days} is a practical, step-by-step tutorial. The reader will learn to design and employ this open source database technology into his or her Web site using practical, hands-on examples to follow. @* @emph{E-Commerce Solutions with MySQL} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0761524452&bfmtype=book, Barnes and Noble} @item Publisher @tab Prima Communications, Inc. @item Authors @tab N/A @item Pub Date @tab January 2000 @item ISBN @tab 0761524452 @item Pages @tab 500 @item Price @tab $39.99 @end multitable No description available. @* @emph{MySQL and PHP from Scratch} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0789724405&bfmtype=book, Barnes and Noble} @item Publisher @tab Que @item Authors @tab N/A @item Pub Date @tab September 2000 @item ISBN @tab 0789724405 @item Pages @tab 550 @item Price @tab $34.99 @end multitable This book puts together information on installing, setting up, and troubleshooting Apache, MySQL, PHP3, and IMP into one complete volume. You also learn how each piece is part of a whole by learning, step-by-step, how to create a web-based e-mail system. Learn to run the equivalent of Active Server Pages (ASP) using PHP3, set up an e-commerce site using a database and the Apache web server, and create a data entry system (such as sales, product quality tracking, customer preferences, etc) that no installation in the PC. @* @emph{Professional MySQL Programming} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://shop.barnesandnoble.com/bookSearch/isbnInquiry.asp?isbn=1861005164, Barnes and Noble} @item Publisher @tab Wrox Press, Inc. @item Authors @tab N/A @item Pub Date @tab Late 2001 @item ISBN @tab 1861005164 @item Pages @tab 1000 @item Price @tab $49.99 @end multitable No description available. @* @emph{Professional Linux Programming} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=1861003013&bfmtype=book, Barnes and Noble} @item Publisher @tab Wrox Press, Inc. @item Authors @tab N/A @item Pub Date @tab September 2000 @item ISBN @tab 1861003013 @item Pages @tab 1155 @item Price @tab $47.99 @end multitable In this follow-up to the best-selling @emph{Beginning Linux Programming}, you will learn from the authors' real-world knowledge and experience of developing software for Linux; you'll be taken through the development of a sample 'DVD Store' application, with 'theme' chapters addressing different aspects of its implementation. Meanwhile, individual ``take-a-break'' chapters cover important topics that go beyond the bounds of the central theme. All focus on the practical aspects of programming, showing how crucial it is to choose the right tools for the job, use them as they should be used, and get things right first time. @* @emph{PHP and MySQL Web Development} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0672317842&bfmtype=book, Barnes and Noble} @item Publisher @tab Sams @item Authors @tab Luke Welling, Laura Thomson @item Pub Date @tab March 2001 @item ISBN @tab 0672317842 @item Pages @tab 700 @item Price @tab $49.99 @end multitable @emph{PHP and MySQL Web Development} introduces you to the advantages of implementing both MySQL and PHP. These advantages are detailed through the provision of both statistics and several case studies. A practical web application is developed throughout the book, providing you with the tools necessary to implement a functional online database. Each function is developed separately, allowing you the choice to incorporate only those parts that you would like to implement. Programming concepts of the PHP language are highlighted, including functions which tie MySQL support into a PHP script and advanced topics regarding table manipulation. @* @strong{Books recommended by the MySQL Developers} @emph{SQL-99 Complete, Really} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0879305681&bfmtype=book, Barnes and Noble} @item Publisher @tab CMP Books @item Authors @tab Peter Gulutzan, Trudy Pelzer @item Pub Date @tab April 1999 @item ISBN @tab 0879305681 @item Pages @tab 1104 @item Price @tab $55.96 @end multitable This book contains complete descriptions of the new standards for syntax, data structures, and retrieval processes of SQL databases. As an example-based reference manual, it includes all of the CLI functions, information, schema tables, and status codes, as well as a working SQL database provided on the companion disk. @* @emph{C, A reference manual} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0133262243&bfmtype=book, Barnes and Noble} @item Publisher @tab Prentice Hall @item Authors @tab Samuel P. Harbison, Guy L. Steele @item Pub Date @tab September 1994 @item ISBN @tab 0133262243 @item Pages @tab 480 @item Price @tab $35.99 @end multitable A new and improved revision of the bestselling C language reference. This manual introduces the notion of "Clean C", writing C code that can be compiled as a C++ program, C programming style that emphasizes correctness, portability, maintainability, and incorporates the ISO C Amendment 1 (1994) which specifies new facilities for writing portable, international programs in C. @* @emph{C++ for Real Programmers} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0120499428&bfmtype=book, Barnes and Noble} @item Publisher @tab Academic Press, Incorporated @item Authors @tab Jeff Alger, Jim Keogh @item Pub Date @tab February 1998 @item ISBN @tab 0120499428 @item Pages @tab 388 @item Price @tab $39.95 @end multitable @emph{C++ For Real Programmers} bridges the gap between C++ as described in beginner and intermediate-level books and C++ as it is practiced by experts. Numerous valuable techniques are described, organized into three simple themes: indirection, class hierarchies, and memory management. It also provides in-depth coverage of template creation, exception handling, pointers and optimization techniques. The focus of the book is on ANSI C++ and, as such, is compiler independent. @emph{C++ For Real Programmers} is a revision of @emph{Secrets of the C++ Masters} and includes a new appendix comparing C++ with Java. The book comes with a 3.5" disk for Windows with source code. @* @emph{Algorithms in C} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0201514257&bfmtype=book, Barnes and Noble} @item Publisher @tab Addison Wesley Longman, Inc. @item Authors @tab Robert Sedgewick @item Pub Date @tab April 1990 @item ISBN @tab 0201514257 @item Pages @tab 648 @item Price @tab $45.75 @end multitable @emph{Algorithms in C} describes a variety of algorithms in a number of areas of interest, including: sorting, searching, string-processing, and geometric, graph and mathematical algorithms. The book emphasizes fundamental techniques, providing readers with the tools to confidently implement, run, and debug useful algorithms. @* @emph{Multithreaded Programming with Pthreads} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=0136807291&bfmtype=book, Barnes and Noble} @item Publisher @tab Prentice Hall @item Authors @tab Bil Lewis, Daniel J. Berg @item Pub Date @tab October 1997 @item ISBN @tab 0136807291 @item Pages @tab 432 @item Price @tab $34.95 @end multitable Based on the best-selling @emph{Threads Primer}, @emph{Multithreaded Programming with Pthreads} gives you a solid understanding of Posix threads: what they are, how they work, when to use them, and how to optimize them. It retains the clarity and humor of @emph{Threads Primer}, but includes expanded comparisons to Win32 and OS/2 implementations. Code examples tested on all of the major UNIX platforms are featured along with detailed explanations of how and why they use threads. @* @emph{Programming the PERL DBI: Database Programming with PERL} @multitable @columnfractions .3 .7 @item Available @tab @uref{http://service.bfast.com/bfast/click?bfmid=2181&sourceid=34233559&bfpid=1565926994&bfmtype=book, Barnes and Noble} @item Publisher @tab O'Reilly & Associates, Incorporated @item Authors @tab Alligator Descartes, Tim Bunce @item Pub Date @tab February 2000 @item ISBN @tab 1565926994 @item Pages @tab 400 @item Price @tab $27.96 @end multitable @emph{Programming the Perl DBI} is coauthored by Alligator Descartes, one of the most active members of the DBI community, and by Tim Bunce, the inventor of DBI. For the uninitiated, the book explains the architecture of DBI and shows you how to write DBI-based programs. For the experienced DBI dabbler, this book explains DBI's nuances and the peculiarities of each individual DBD. The book includes: @itemize @bullet @item An introduction to DBI and its design. @item How to construct queries and bind parameters. @item Working with database, driver, and statement handles. @item Debugging techniques. @item Coverage of each existing DBD. @item A complete reference to DBI. @end itemize @* @node General-SQL, Useful Links, MySQL-Books, MySQL Information Sources @subsection General SQL Information and Tutorials The following book has been recommended by several people on the MySQL mailing list: @example Judith S. Bowman, Sandra L. Emerson and Marcy Darnovsky @emph{The Practical SQL Handbook: Using Structured Query Language} Second Edition Addison-Wesley ISBN 0-201-62623-3 http://www.awl.com @end example The following book has also received some recommendations by MySQL users: @example Martin Gruber @emph{Understanding SQL} ISBN 0-89588-644-8 Publisher Sybex 510 523 8233 Alameda, CA USA @end example A SQL tutorial is available on the net at http://w3.one.net/~jhoffman/sqltut.htm @node Useful Links, Questions, General-SQL, MySQL Information Sources @subsection Useful MySQL-related Links Apart from the following links, you can find and download a lot of MySQL programs, tools and APIs from the @uref{http://www.mysql.com/Downloads/Contrib/, Contrib directory}. @cindex @code{Contrib} directory @cindex URLs to MySQL information @cindex related information URLs MySQL @subheading Tutorials and Manuals @table @asis @item @uref{http://michael.bacarella.com/research/mysqlmyths.html, MySQL Myths Debunked} MySQL used in the real world. @item @uref{http://www.4t2.com/mysql} Information about the German MySQL mailing list. @item @uref{http://www2.rent-a-database.de/mysql/} MySQL handbook in German. @item @uref{http://www.bitmover.com:8888//home/bk/mysql} Web access to the MySQL BitKeeper repository. @item @uref{http://www.analysisandsolutions.com/code/mybasic.htm} Beginners MySQL Tutorial on how to install and set up MySQL on a Windows machine. @item @uref{http://www.devshed.com/Server_Side/MySQL/} A lot of MySQL tutorials. @item @uref{http://mysql.hitstar.com/} MySQL manual in Chinese. @item @uref{http://www.linuxplanet.com/linuxplanet/tutorials/1046/1/} Setting Up a MySQL-based Web site. @item @uref{http://www.hotwired.com/webmonkey/backend/tutorials/tutorial1.html} MySQL-Perl tutorial. @item @uref{http://www.iserver.com/support/contrib/perl5/modules.html} Installing new Perl modules that require locally installed modules. @item @uref{http://www.hotwired.com/webmonkey/databases/tutorials/tutorial4.html} PHP/MySQL Tutorial. @item @uref{http://www.useractive.com/} Hands on tutorial for MySQL. @end table @subheading Porting MySQL/Using MySQL on Different Systems @table @asis @item @uref{http://www.entropy.ch/software/macosx/mysql/} Binary of MySQL for Mac OS X Client. Includes information of how to build and use MySQL on Mac OS X. @item @uref{http://xclave.macnn.com/MySQL/} The Mac OS Xclave. Running MySQL on Mac OS X. @item @uref{http://www.prnet.de/RegEx/mysql.html} MySQL for Mac OS X Server. @item @uref{http://www.latencyzero.com/macosx/mysql.html} Building MySQL for Mac OS X. @item @uref{http://www.essencesw.com/Software/mysqllib.html} New Client libraries for the Mac OS Classic (Macintosh). @item @uref{http://www.lilback.com/macsql/} Client libraries for Mac OS Classic (Macintosh). @item @uref{http://sixk.maniasys.com/index_en.html} MySQL for Amiga @end table @subheading Perl-related Links @table @asis @item @uref{http://dbimysql.photoflux.com/} Perl DBI with MySQL FAQ. @end table @subheading MySQL Discussion Forums @table @asis @item @uref{http://www.weberdev.com/} Examples using MySQL; (check Top 20) @item @uref{http://futurerealm.com/forum/futureforum.htm} FutureForum Web Discussion Software. @end table @c FIX We should get longer descriptions for things in this category! @subheading Commercial Applications that Support MySQL @table @asis @item @uref{http://www.supportwizard.com/} SupportWizard; Interactive helpdesk on the Web (This product includes a licensed copy of MySQL.) @item @uref{http://www.sonork.com/} Sonork, Instant Messenger that is not only Internet oriented. It's focused on private networks and on small to medium companies. Client is free, server is free for up to 5 seats. @item @uref{http://www.stweb.org/} StWeb - Stratos Web and Application server - An easy-to-use, cross platform, Internet/Intranet development and deployment system for development of web-enabled applications. The standard version of StWeb has a native interface to MySQL database. @item @uref{http://www.rightnowtech.com/} Right Now Web; Web automation for customer service. @item @uref{http://www.icaap.org/Bazaar/} Bazaar; Interactive Discussion Forums with Web interface. @cindex PhoneSweep @item @uref{http://www.phonesweep.com/} PhoneSweepT is the world's first commercial Telephone Scanner. Many break-ins in recent years have come not through the Internet, but through unauthorized dial-up modems. PhoneSweep lets you find these modems by repeatedly placing phone calls to every phone number that your organization controls. PhoneSweep has a built-in expert system that can recognize more than 250 different kinds of remote-access programs, including Carbon Copy(TM), pcANYWHERE(TM), and Windows NT RAS. All information is stored in the SQL database. It then generates a comprehensive report detailing which services were discovered on which dial-up numbers in your organization. @end table @subheading SQL Clients and Report Writers @table @asis @item @uref{http://www.urbanresearch.com/software/utils/urbsql/index.html, urSQL} SQL Editor and Query Utility. Custom syntax highlighting, editable results grid, exportable result-sets, basic MySQL admin functions, Etc.. For Windows. @item @uref{http://www.edatanew.com/, MySQL Data Manager} MySQL Data Manager * is platform independent web client (written in perl) for MySQL server over TCP/IP. @item @uref{http://ksql.sourceforge.net/} KDE MySQL client. @item @uref{http://www.ecker-software.de} A Windows GUI client by David Ecker. @item @uref{http://www.icaap.org/software/kiosk/} Kiosk; a MySQL client for database management. Written in Perl. Will be a part of Bazaar. @item @uref{http://www.casestudio.com/} Db design tool that supports MySQL 3.23. @item @uref{http://home.skif.net/~voland/zeos/eng/index.html} Zeos - A client that supports MySQL, Interbase and PostgreSQL. @item @uref{http://www.geocities.com/SiliconValley/Ridge/4280/GenericReportWriter/grwhome.html} A free report writer in Java @item @uref{http://www.javaframework.de} MySQLExport - Export of MySQL create statements and data in a lot of different formats (SQL, HTML, CVS, text, ZIP, GZIP...) @item @uref{http://dlabs.4t2.com} M2D, a MySQL Administration client for Windows. M2D supports administration of MySQL databases, creation of new databases and tables, editing, and more. @item @uref{http://dlabs.4t2.com} Dexter, a small server written in Perl which can be used as a proxy server for MySQL or as a database extender. @item @uref{http://www.scibit.com/Products/Software/Utils/Mascon.asp} Mascon is a powerful Win32 GUI for administering MySQL databases. @item @uref{http://www.rtlabs.com/} MacSQL Monitor. GUI for MySQL, ODBC, and JDBC databases for the Mac OS. @end table @subheading Distributions that Include MySQL @c FIX add the rest (at least a couple more Linuxes) @table @asis @item @uref{http://www.suse.com/} SuSE Linux (6.1 and above) @item @uref{http://www.redhat.com/} RedHat Linux (7.0 and above) @item @uref{http://distro.conectiva.com.br} Conectiva Linux (4.0 and above) @end table @subheading Web Development Tools that Support MySQL @table @asis @item @uref{http://www.php.net/} PHP: A server-side HTML-embedded scripting language. @item @uref{http://www.midgard-project.org} The Midgard Application Server; a powerful Web development environment based on MySQL and PHP. @item @uref{http://www.smartworker.org} SmartWorker is a platform for Web application development. @item @uref{http://xsp.lentus.se/} XSP: e(X)tendible (S)erver (P)ages and is a HTML embedded tag language written in Java (previously known as XTAGS.) @cindex dbServ @item @uref{http://www.dbServ.de/} dbServ is an extension to a web server to integrate database output into your HTML code. You may use any HTML function in your output. Only the client will stop you. It works as standalone server or as Java servlet. @item @uref{http://www.chilisoft.com/} Platform independent ASP from Chili!Soft @item @uref{http://www.voicenet.com/~zellert/tjFM} A JDBC driver for MySQL. @item @uref{http://www.wernhart.priv.at/php/} MySQL + PHP demos. @item @uref{http://www.dbwww.com/} ForwardSQL: HTML interface to manipulate MySQL databases. @item @uref{http://www.daa.com.au/~james/www-sql/} WWW-SQL: Display database information. @item @uref{http://www.minivend.com/minivend/} Minivend: A Web shopping cart. @item @uref{http://www.heitml.com/} HeiTML: A server-side extension of HTML and a 4GL language at the same time. @item @uref{http://www.metahtml.com/} Metahtml: A Dynamic Programming Language for WWW Applications. @item @uref{http://www.binevolve.com/} VelocityGen for Perl and Tcl. @item @uref{http://hawkeye.net/} Hawkeye Internet Server Suite. @item @uref{http://www.fastflow.com/} Network Database Connection For Linux @item @uref{http://www.wdbi.net/} WDBI: Web browser as a universal front end to databases which supports MySQL well. @item @uref{http://www.webgroove.com/} WebGroove Script: HTML compiler and server-side scripting language. @item @uref{http://www.ihtml.com/} A server-side Web site scripting language. @item @uref{ftp://ftp.igc.apc.org/pub/myodbc/README} How to use MySQL with ColdFusion on Solaris. @item @uref{http://calistra.com/MySQL/} Calistra's ODBC MySQL Administrator. @cindex Webmerger @item @uref{http://www.webmerger.com} Webmerger - This CGI tool interprets files and generates dynamic output based on a set of simple tags. Ready-to-run drivers for MySQL and PostgreSQL through ODBC. @item @uref{http://phpclub.net/} PHPclub - Tips and tricks for PHP. @item @uref{http://www.penguinservices.com/scripts} MySQL and Perl Scripts. @item @uref{http://www.widgetchuck.com} The Widgetchuck; Web Site Tools and Gadgets @item @uref{http://www.adcycle.com/} AdCycle - advertising management software. @cindex pwPage @item @uref{http://sourceforge.net/projects/pwpage/} pwPage - provides an extremely fast and simple approach to the creation of database forms. That is, if a database table exists and an HTML page has been constructed using a few simple guidelines, pwPage can be immediately used for table data selections, insertions, updates, deletions and selectable table content reviewing. @item @uref{http://www.omnis-software.com/products/studio/studio.html} OMNIS Studio is a rapid application development (RAD) tool. @cindex Web+ @item @uref{http://www.webplus.com} talentsoft Web+ 4.6 - a powerful and comprehensive development language for use in creating web-based client/server applications without writing complicated, low-level, and time-consuming CGI programs. @end table @subheading Database Design Tools with MySQL Support @table @asis @item @uref{http://www.mysql.com/documentation/dezign/} "DeZign for databases" is a database development tool that uses an entity relationship diagram (ERD). @end table @subheading Web Servers with MySQL Tools @table @asis @item @uref{ftp://ftp.kcilink.com/pub/} mod_auth_mysql, An Apache authentication module. @item @uref{http://www.roxen.com/} The Roxen Challenger Web server. @end table @subheading Extensions for Other Programs @table @asis @item @uref{http://www.seawood.org/msql_bind/} MySQL support for BIND (The Internet Domain Name Server). @item @uref{http://www.inet-interactive.com/sendmail/} MySQL support for Sendmail and Procmail. @end table @subheading Using MySQL with Other Programs @table @asis @item @uref{http://www.iserver.com/support/addonhelp/database/mysql/msaccess.html} Using MySQL with Access. @item @uref{http://www.iserver.com/support/contrib/perl5/modules.html} Installing new Perl modules that require locally installed modules. @end table @subheading ODBC-related Links @table @asis @item @uref{http://www.iodbc.org/} Popular iODBC Driver Manager (libiodbc) now available as Open Source. @item @uref{http://users.ids.net/~bjepson/freeODBC/} The FreeODBC Pages. @item @uref{http://genix.net/unixODBC/} The unixODBC Project goals are to develop and promote unixODBC to be the definitive standard for ODBC on the Linux platform. This is to include GUI support for KDE. @item @uref{http://www.sw-soft.com/products/BtrieveODBC/} A MySQL-based ODBC driver for Btrieve. @end table @subheading @strong{API}-related Links @table @asis @item @uref{http://www.jppp.com/} Partially implemented TDataset-compatible components for MySQL. @item @uref{http://www.riverstyx.net/qpopmysql/} qpopmysql - A patch to allow POP3 authentication from a MySQL database. There's also a link to Paul Khavkine's patch for Procmail to allow any MTA to deliver to users in a MySQL database. @item @uref{http://www.pbc.ottawa.on.ca} Visual Basic class generator for Active X. @item @uref{http://www.essencesw.com/Software/mysqllib.html} New Client libraries for the Mac OS Classic (Macintosh). @item @uref{http://www.lilback.com/macsql/} Client libraries for the Macintosh. @item @uref{http://www.essencesw.com/Plugins/mysqlplug.html} Plugin for REALbasic (for Macintosh) @item @uref{http://www.iis.ee.ethz.ch/~neeri/macintosh/gusi-qa.html} A library that emulates BSD sockets and pthreads on Macintosh. This can be used if you want to compile the MySQL client library on Mac. It could probably even be sued to port MySQL to Macintosh, but we don't know of anyone that has tried that. @cindex SCMDB @item @uref{http://www.dedecker.net/jessie/scmdb/} SCMDB - an add-on for SCM that ports the MySQL C library to scheme (SCM). With this library scheme developers can make connections to a MySQL database and use embedded SQL in their programs. @end table @subheading Other MySQL-related Links @table @asis @item @uref{http://www.satisoft.com/, SAT} The Small Application Toolkit (SAT) is a collection of utilities intended to simplify the development of small, multi-user, GUI based applications in a (Microsoft -or- X) Windows Client / Unix Server environment. @item @uref{http://www.wix.com/mysql-hosting/} Registry of Web providers who support MySQL. @item @uref{http://www.softagency.co.jp/mysql/index.en.html} Links about using MySQL in Japan/Asia. @item @uref{http://abattoir.cc.ndsu.nodak.edu/~nem/mysql/udf/} MySQL UDF Registry. @item @uref{http://www.open.com.au/products.html} Commercial Web defect tracking system. @item @uref{http://www.stonekeep.com/pts/} PTS: Project Tracking System. @item @uref{http://tomato.nvgc.vt.edu/~hroberts/mot} Job and software tracking system. @item @uref{http://www.cynergi.net/exportsql/} ExportSQL: A script to export data from Access95+. @item @uref{http://SAL.KachinaTech.COM/H/1/MYSQL.html} SAL (Scientific Applications on Linux) MySQL entry. @item @uref{http://www.infotech-nj.com/itech/index.shtml} A consulting company which mentions MySQL in the right company. @item @uref{http://www.pmpcs.com/} PMP Computer Solutions. Database developers using MySQL and @code{mSQL}. @item @uref{http://www.aewa.org/} Airborne Early Warning Association. @item @uref{http://www.dedserius.com/y2kmatrix/} Y2K tester. @end table @subheading SQL and Database Interfaces @table @asis @item @uref{http://java.sun.com/products/jdbc/} The JDBC database access API. @item @uref{http://www.gagme.com/mysql} Patch for @code{mSQL} Tcl. @item @uref{http://www.amsoft.ru/easysql/} EasySQL: An ODBC-like driver manager. @item @uref{http://www.lightlink.com/hessling/rexxsql.html} A REXX interface to SQL databases. @item @uref{http://www.mytcl.cx/} Tcl interface based on tcl-sql with many bugfixes. @item @uref{http://www.binevolve.com/~tdarugar/tcl-sql/} Tcl interface. @item @uref{http://www.contrib.andrew.cmu.edu/~shadow/sql.html} SQL Reference Page with a lot of interesting links. @end table @subheading Examples of MySQL Use @table @asis @c Added 990601 @c EMAIL: thuss@little6.com (Todd Huss) @item @uref{http://www.little6.com/about/linux/} Little6 Inc., An online contract and job finding site that is powered by MySQL, PHP3, and Linux. @c Added 990521 @c EMAIL: nh@delec.com (Hillbrecht Nicole) @item @uref{http://www.delec.com/is/products/prep/examples/BookShelf/index.html} DELECis - A tool that makes it very easy to create an automatically generated table documentation. They have used MySQL as an example. @c Added 990521 @c EMAIL: info@worldrecords.com (Jim Rota) @item @uref{http://www.worldrecords.com} World Records - A search engine for information about music that uses MySQL and PHP. @item @uref{http://www.webtechniques.com/archives/1998/01/note/} A Contact Database using MySQL and PHP. @item @uref{http://modems.rosenet.net/mysql/} Web based interface and Community Calendar with PHP. @item @uref{http://www.odbsoft.com/cook/sources.htm} Perl package to generate html from a SQL table structure and for generating SQL statements from an html form. @item @uref{http://www.gusnet.cx/proj/telsql/} Basic telephone database using @code{DBI}/@code{DBD}. @item @uref{http://tecfa.unige.ch/guides/java/staf2x/ex/jdbc/coffee-break} JDBC examples by Daniel K. Schneider. @item @uref{http://www.spade.com/linux/howto/PostgreSQL-HOWTO-41.html} SQL BNF @item @uref{http://www.ooc.com/} Object Oriented Concepts Inc; CORBA applications with examples in source. @item @uref{http://www.pbc.ottawa.on.ca/} DBWiz; Includes an example of how to manage cursors in VB. @cindex Pluribus @item @uref{http://keilor.cs.umass.edu/pluribus/} Pluribus is a free search engine that learns to improve the quality of its results over time. Pluribus works by recording which pages a user prefers among those returned for a query. A user votes for a page by selecting it; Pluribus then uses that knowledge to improve the quality of the results when someone else submits the same (or similar) query. Uses PHP and MySQL. @c EMAIL: paul@sword.damocles.com (Paul Bannister) @item @uref{http://www.stopbit.com/} Stopbit - A technology news site using MySQL and PHP. @item @uref{http://www.linuxsupportline.com/~kalendar/} KDE based calendar manager - The calendar manager has both single user (file based) and multi-user (MySQL database) support. @item @uref{http://tim.desert.net/~tim/imger/} Example of storing/retrieving images with MySQL and CGI. @item @uref{http://www.penguinservices.com/scripts} Online shopping cart system. @c Added 990928 from editor@city-gallery.com @cindex Old Photo Album @item @uref{http://www.city-gallery.com/album/} Old Photo Album - The album is a collaborative popular history of photography project that generates all pages from data stored in a MySQL database. Pages are dynamically generated through a php3 interface to the database content. Users contribute images and descriptions. Contributed images are stored on the web server to avoid storing them in the database as BLOBs. All other information is stored on the shared MySQL server. @end table @subheading General Database Links @table @asis @item @uref{http://www.pcslink.com/~ej/dbweb.html} Database Jump Site @item @uref{http://black.hole-in-the.net/guy/webdb/} Homepage of the webdb-l (Web Databases) mailing list. @item @uref{http://www.symbolstone.org/technology/perl/DBI/index.html} Perl @code{DBI}/@code{DBD} modules homepage. @item @uref{http://www.student.uni-koeln.de/cygwin/} Cygwin tools. Unix on top of Windows. @item @uref{http://dbasecentral.com/} dbasecentral.com; Development and distribution of powerful and easy-to-use database applications and systems. @cindex Tek-Tips forums @cindex forums, Tek-Tips @item @uref{http://www.tek-tips.com/} Tek-Tips Forums are 800+ independent peer-to-peer non-commercial support forums for Computer Professionals. Features include automatic e-mail notification of responses, a links library, and member confidentiality guaranteed. @item @uref{http://www.public.asu.edu/~peterjn/btree/} B-Trees: Balanced Tree Data Structures. @item @uref{http://www.fit.qut.edu.au/~maire/baobab/lecture/sld001.htm} A lecture about B-Trees. @end table There are also many Web pages that use MySQL. @xref{Users}. Send any additions to this list to @email{webmaster@@mysql.com}. We now require that you show a MySQL logo somewhere if you wish your site to be added. It is okay to have it on a ``used tools'' page or something similar. @node Questions, , Useful Links, MySQL Information Sources @subsection MySQL Mailing Lists @cindex reporting, errors @cindex MySQL mailing lists @menu * Mailing-list:: The MySQL mailing lists * Asking questions:: Asking questions or reporting bugs * Bug reports:: How to report bugs or problems * Answering questions:: Guidelines for answering questions on the mailing list @end menu This section introduces you to the MySQL mailing lists, and gives some guidelines as to how to use them. @node Mailing-list, Asking questions, Questions, Questions @subsubsection The MySQL Mailing Lists @cindex mailing lists @cindex email lists To subscribe to the main MySQL mailing list, send a message to the electronic mail address @email{mysql-subscribe@@lists.mysql.com}. To unsubscribe from the main MySQL mailing list, send a message to the electronic mail address @email{mysql-unsubscribe@@lists.mysql.com}. Only the address to which you send your messages is significant. The subject line and the body of the message are ignored. @c the last two addresses in this paragraph are NOT @email because they @c shouldn't be live links. If your reply address is not valid, you can specify your address explicitly. Adding a hyphen to the subscribe or unsubscribe command word, followed by your address with the @samp{@@} character in your address replaced by a @samp{=}. For example, to subscribe @code{your_name@@host.domain}, send a message to @code{mysql-subscribe-your_name=host.domain@@lists.mysql.com}. Mail to @email{mysql-subscribe@@lists.mysql.com} or @email{mysql-unsubscribe@@lists.mysql.com} is handled automatically by the ezmlm mailing list processor. Information about ezmlm is available at @uref{http://www.ezmlm.org, The ezmlm Website}. To post a message to the list itself, send your message to @code{mysql@@lists.mysql.com}. However, please @emph{do not} send mail about subscribing or unsubscribing to @email{mysql@@lists.mysql.com}, because any mail sent to that address is distributed automatically to thousands of other users. Your local site may have many subscribers to @email{mysql@@lists.mysql.com}. If so, it may have a local mailing list, so that messages sent from @code{lists.mysql.com} to your site are propagated to the local list. In such cases, please contact your system administrator to be added to or dropped from the local MySQL list. If you wish to have traffic for a mailing list go to a separate mailbox in your mail program, set up a filter based on the message headers. You can use either the @code{List-ID:} or @code{Delivered-To:} headers to identify list messages. The following MySQL mailing lists exist: @table @code @item @email{announce-subscribe@@lists.mysql.com} announce This is for announcement of new versions of MySQL and related programs. This is a low volume list all MySQL users should subscribe to. @item @email{mysql-subscribe@@lists.mysql.com} mysql The main list for general MySQL discussion. Please note that some topics are better discussed on the more-specialized lists. If you post to the wrong list, you may not get an answer! @item @email{mysql-digest-subscribe@@lists.mysql.com} mysql-digest The @code{mysql} list in digest form. That means you get all individual messages, sent as one large mail message once a day. @item @email{bugs-subscribe@@lists.mysql.com} bugs On this list you should only post a full, repeatable bug report using the @code{mysqlbug} script (if you are running on Windows, you should include a description of the operating system and the MySQL version). Preferably, you should test the problem using the latest stable or development version of MySQL before posting! Anyone should be able to repeat the bug by just using @code{mysql test < script} on the included test case. All bugs posted on this list will be corrected or documented in the next MySQL release! If there are only small code changes involved, we will also post a patch that fixes the problem. @item @email{bugs-digest-subscribe@@lists.mysql.com} bugs-digest The @code{bugs} list in digest form. @item @email{internals-subscribe@@lists.mysql.com} internals A list for people who work on the MySQL code. On this list one can also discuss MySQL development and post patches. @item @email{internals-digest-subscribe@@lists.mysql.com} internals-digest A digest version of the @code{internals} list. @item @email{java-subscribe@@lists.mysql.com} java Discussion about MySQL and Java. Mostly about the JDBC drivers. @item @email{java-digest-subscribe@@lists.mysql.com} java-digest A digest version of the @code{java} list. @item @email{win32-subscribe@@lists.mysql.com} win32 All things concerning MySQL on Microsoft operating systems such as Win95, Win98, NT, and Win2000. @item @email{win32-digest-subscribe@@lists.mysql.com} win32-digest A digest version of the @code{win32} list. @item @email{myodbc-subscribe@@lists.mysql.com} myodbc All things about connecting to MySQL with ODBC. @item @email{myodbc-digest-subscribe@@lists.mysql.com} myodbc-digest A digest version of the @code{myodbc} list. @item @email{plusplus-subscribe@@lists.mysql.com} plusplus All things concerning programming with the C++ API to MySQL. @item @email{plusplus-digest-subscribe@@lists.mysql.com} plusplus-digest A digest version of the @code{plusplus} list. @item @email{msql-mysql-modules-subscribe@@lists.mysql.com} msql-mysql-modules A list about the Perl support in MySQL. msql-mysql-modules @item @email{msql-mysql-modules-digest-subscribe@@lists.mysql.com} msql-mysql-modules-digest A digest version of the @code{msql-mysql-modules} list. @end table You subscribe or unsubscribe to all lists in the same way as described above. In your subscribe or unsubscribe message, just put the appropriate mailing list name rather than @code{mysql}. For example, to subscribe to or unsubscribe from the @code{myodbc} list, send a message to @email{myodbc-subscribe@@lists.mysql.com} or @email{myodbc-unsubscribe@@lists.mysql.com}. If you can't get an answer for your questions from the mailing list, one option is to pay for support from MySQL AB, which will put you in direct contact with MySQL developers. @xref{Support}. The following table shows some MySQL mailing in other languages than English. Note that these are not operated by MySQL AB, so we can't guarantee the quality on these. @table @code @item @email{mysql-france-subscribe@@yahoogroups.com} A French mailing list @item @email{list@@tinc.net} A Korean mailing list Email @code{subscribe mysql your@@email.address} to this list. @item @email{mysql-de-request@@lists.4t2.com} A German mailing list Email @code{subscribe mysql-de your@@email.address} to this list. You can find information about this mailing list at @uref{http://www.4t2.com/mysql}. @item @email{mysql-br-request@@listas.linkway.com.br} A Portugese mailing list Email @code{subscribe mysql-br your@@email.address} to this list. @item @email{mysql-alta@@elistas.net} A Spanish mailing list Email @code{subscribe mysql your@@email.address} to this list. @end table @node Asking questions, Bug reports, Mailing-list, Questions @subsubsection Asking Questions or Reporting Bugs @cindex net etiquette @cindex mailing lists, archive location @cindex searching, MySQL webpages Before posting a bug report or question, please do the following: @itemize @bullet @item Start by searching the MySQL online manual at: @* @uref{http://www.mysql.com/documentation/manual.php} @* We try to keep the manual up to date by updating it frequently with solutions to newly found problems! @item Search the MySQL mailing list archives: @* @uref{http://www.mysql.com/documentation/} @* @item You can also use @uref{http://www.mysql.com/search.html} to search all the Web pages (including the manual) that are located at @uref{http://www.mysql.com/}. @end itemize If you can't find an answer in the manual or the archives, check with your local MySQL expert. If you still can't find an answer to your question, go ahead and read the next section about how to send mail to @email{mysql@@lists.mysql.com}. @node Bug reports, Answering questions, Asking questions, Questions @subsubsection How to Report Bugs or Problems @cindex bugs, reporting @cindex reporting, bugs @cindex problems, reporting @cindex errors, reporting @cindex @code{mysqlbug} script @cindex creating, bug reports @cindex scripts, @code{mysqlbug} Writing a good bug report takes patience, but doing it right the first time saves time for us and for you. A good bug report containing a full test case for the bug will make it very likely that we will fix it in the next release. This section will help you write your report correctly so that you don't waste your time doing things that may not help us much or at all. We encourage everyone to use the @code{mysqlbug} script to generate a bug report (or a report about any problem), if possible. @code{mysqlbug} can be found in the @file{scripts} directory in the source distribution, or, for a binary distribution, in the @file{bin} directory under your MySQL installation directory. If you are unable to use @code{mysqlbug}, you should still include all the necessary information listed in this section. The @code{mysqlbug} script helps you generate a report by determining much of the following information automatically, but if something important is missing, please include it with your message! Please read this section carefully and make sure that all the information described here is included in your report. @cindex bug reports, email address The normal place to report bugs and problems is @email{mysql@@lists.mysql.com}. If you can make a test case that clearly demonstrates the bug, you should post it to the @email{bugs@@lists.mysql.com} list. Note that on this list you should only post a full, repeatable bug report using the @code{mysqlbug} script. If you are running on Windows, you should include a description of the operating system and the MySQL version. Preferably, you should test the problem using the latest stable or development version of MySQL before posting! Anyone should be able to repeat the bug by just using ``@code{mysql test < script}'' on the included test case or run the shell or perl script that is included in the bug report. All bugs posted on the @code{bugs} list will be corrected or documented in the next MySQL release! If there are only small code changes involved to correct this problem, we will also post a patch that fixes the problem. Remember that it is possible to respond to a message containing too much information, but not to one containing too little. Often people omit facts because they think they know the cause of a problem and assume that some details don't matter. A good principle is: if you are in doubt about stating something, state it! It is a thousand times faster and less troublesome to write a couple of lines more in your report than to be forced to ask again and wait for the answer because you didn't include enough information the first time. The most common errors are that people don't indicate the version number of the MySQL distribution they are using, or don't indicate what platform they have MySQL installed on (including the platform version number). This is highly relevant information, and in 99 cases out of 100 the bug report is useless without it! Very often we get questions like, ``Why doesn't this work for me?'' then we find that the feature requested wasn't implemented in that MySQL version, or that a bug described in a report has been fixed already in newer MySQL versions. Sometimes the error is platform dependent; in such cases, it is next to impossible to fix anything without knowing the operating system and the version number of the platform. Remember also to provide information about your compiler, if it is related to the problem. Often people find bugs in compilers and think the problem is MySQL-related. Most compilers are under development all the time and become better version by version. To determine whether or not your problem depends on your compiler, we need to know what compiler is used. Note that every compiling problem should be regarded as a bug report and reported accordingly. It is most helpful when a good description of the problem is included in the bug report. That is, a good example of all the things you did that led to the problem and the problem itself exactly described. The best reports are those that include a full example showing how to reproduce the bug or problem. @xref{Reproduceable test case}. If a program produces an error message, it is very important to include the message in your report! If we try to search for something from the archives using programs, it is better that the error message reported exactly matches the one that the program produces. (Even the case should be observed!) You should never try to remember what the error message was; instead, copy and paste the entire message into your report! If you have a problem with MyODBC, you should try to generate a MyODBC trace file. @xref{MyODBC bug report}. Please remember that many of the people who will read your report will do so using an 80-column display. When generating reports or examples using the @code{mysql} command line tool, you should therefore use the @code{--vertical} option (or the @code{\G} statement terminator) for output that would exceed the available width for such a display (for example, with the @code{EXPLAIN SELECT} statement; see the example below). @cindex bug reports, criteria for Please include the following information in your report: @itemize @bullet @item The version number of the MySQL distribution you are using (for example, MySQL Version 3.22.22). You can find out which version you are running by executing @code{mysqladmin version}. @code{mysqladmin} can be found in the @file{bin} directory under your MySQL installation directory. @item The manufacturer and model of the machine you are working on. @item The operating system name and version. For most operating systems, you can get this information by executing the Unix command @code{uname -a}. @item Sometimes the amount of memory (real and virtual) is relevant. If in doubt, include these values. @item If you are using a source distribution of MySQL, the name and version number of the compiler used is needed. If you have a binary distribution, the distribution name is needed. @item If the problem occurs during compilation, include the exact error message(s) and also a few lines of context around the offending code in the file where the error occurred. @item If @code{mysqld} died, you should also report the query that crashed @code{mysqld}. You can usually find this out by running @code{mysqld} with logging enabled. @xref{Using log files}. @item If any database table is related to the problem, include the output from @code{mysqldump --no-data db_name tbl_name1 tbl_name2 ...}. This is very easy to do and is a powerful way to get information about any table in a database that will help us create a situation matching the one you have. @item For speed-related bugs or problems with @code{SELECT} statements, you should always include the output of @code{EXPLAIN SELECT ...}, and at least the number of rows that the @code{SELECT} statement produces. The more information you give about your situation, the more likely it is that someone can help you! For example, the following is an example of a very good bug report (it should of course be posted with the @code{mysqlbug} script): Example run using the @code{mysql} command line tool (note the use of the @code{\G} statement terminator for statements whose output width would otherwise exceed that of an 80-column display device): @example mysql> SHOW VARIABLES; mysql> SHOW COLUMNS FROM ...\G <output from SHOW COLUMNS> mysql> EXPLAIN SELECT ...\G <output from EXPLAIN> mysql> FLUSH STATUS; mysql> SELECT ...; <A short version of the output from SELECT, including the time taken to run the query> mysql> SHOW STATUS; <output from SHOW STATUS> @end example @item If a bug or problem occurs while running @strong{mysqld}, try to provide an input script that will reproduce the anomaly. This script should include any necessary source files. The more closely the script can reproduce your situation, the better. If you can make a repeatable test case, you should post this to @email{bugs@@lists.mysql.com} for a high priority treatment! If you can't provide a script, you should at least include the output from @code{mysqladmin variables extended-status processlist} in your mail to provide some information of how your system is performing! @item If you can't produce a test case in a few rows, or if the test table is too big to be mailed to the mailing list (more than 10 rows), you should dump your tables using @code{mysqldump} and create a @file{README} file that describes your problem. Create a compressed archive of your files using @code{tar} and @code{gzip} or @code{zip}, and use @code{ftp} to transfer the archive to @uref{ftp://support.mysql.com/pub/mysql/secret/}. Then send a short description of the problem to @email{bugs@@lists.mysql.com}. @item If you think that MySQL produces a strange result from a query, include not only the result, but also your opinion of what the result should be, and an account describing the basis for your opinion. @item When giving an example of the problem, it's better to use the variable names, table names, etc., that exist in your actual situation than to come up with new names. The problem could be related to the name of a variable or table! These cases are rare, perhaps, but it is better to be safe than sorry. After all, it should be easier for you to provide an example that uses your actual situation, and it is by all means better for us. In case you have data you don't want to show to others, you can use @code{ftp} to transfer it to @uref{ftp://support.mysql.com/pub/mysql/secret/}. If the data are really top secret and you don't want to show them even to us, then go ahead and provide an example using other names, but please regard this as the last choice. @item Include all the options given to the relevant programs, if possible. For example, indicate the options that you use when you start the @code{mysqld} daemon and that you use to run any MySQL client programs. The options to programs like @code{mysqld} and @code{mysql}, and to the @code{configure} script, are often keys to answers and are very relevant! It is never a bad idea to include them anyway! If you use any modules, such as Perl or PHP, please include the version number(s) of those as well. @item If your question is related to the privilege system, please include the output of @code{mysqlaccess}, the output of @code{mysqladmin reload}, and all the error messages you get when trying to connect! When you test your privileges, you should first run @code{mysqlaccess}. After this, execute @code{mysqladmin reload version} and try to connect with the program that gives you trouble. @code{mysqlaccess} can be found in the @file{bin} directory under your MySQL installation directory. @item If you have a patch for a bug, that is good, but don't assume the patch is all we need, or that we will use it, if you don't provide some necessary information, such as test cases showing the bug that your patch fixes. We might find problems with your patch or we might not understand it at all; if so, we can't use it. If we can't verify exactly what the patch is meant for, we won't use it. Test cases will help us here. Show that the patch will handle all the situations that may occur. If we find a borderline case (even a rare one) where the patch won't work, it may be useless. @item Guesses about what the bug is, why it occurs, or what it depends on, are usually wrong. Even the MySQL team can't guess such things without first using a debugger to determine the real cause of a bug. @item Indicate in your mail message that you have checked the reference manual and mail archive so others know that you have tried to solve the problem yourself. @item If you get a @code{parse error}, please check your syntax closely! If you can't find something wrong with it, it's extremely likely that your current version of MySQL doesn't support the query you are using. If you are using the current version and the manual at @uref{http://www.mysql.com/documentation/manual.php} doesn't cover the syntax you are using, MySQL doesn't support your query. In this case, your only options are to implement the syntax yourself or e-mail @email{mysql-licensing@@mysql.com} and ask for an offer to implement it! If the manual covers the syntax you are using, but you have an older version of MySQL, you should check the MySQL change history to see when the syntax was implemented. In this case, you have the option of upgrading to a newer version of MySQL. @xref{News}. @item If you have a problem such that your data appears corrupt or you get errors when you access some particular table, you should first check and then try repairing your tables with @code{myisamchk} or @code{CHECK TABLE} and @code{REPAIR TABLE}. @xref{MySQL Database Administration}. @item If you often get corrupted tables you should try to find out when and why this happens! In this case, the @file{mysql-data-directory/'hostname'.err} file may contain some information about what happened. @xref{Error log}. Please include any relevant information from this file in your bug report! Normally @code{mysqld} should @strong{NEVER} crash a table if nothing killed it in the middle of an update! If you can find the cause of @code{mysqld} dying, it's much easier for us to provide you with a fix for the problem! @xref{What is crashing}. @item If possible, download and install the most recent version of MySQL and check whether or not it solves your problem. All versions of MySQL are thoroughly tested and should work without problems! We believe in making everything as backward compatible as possible, and you should be able to switch MySQL versions in minutes! @xref{Which version}. @end itemize @cindex technical support, mailing address @cindex support, mailing address @cindex customer support, mailing address @cindex mailing address, for customer support If you are a support customer, please cross-post the bug report to @email{mysql-support@@mysql.com} for higher priority treatment, as well as to the appropriate mailing list to see if someone else has experienced (and perhaps solved) the problem. For information on reporting bugs in @strong{MyODBC}, see @ref{ODBC Problems}. For solutions to some common problems, see @xref{Problems}. When answers are sent to you individually and not to the mailing list, it is considered good etiquette to summarize the answers and send the summary to the mailing list so that others may have the benefit of responses you received that helped you solve your problem! @node Answering questions, , Bug reports, Questions @subsubsection Guidelines for Answering Question on the Mailing List @cindex net etiquette @cindex questions, answering @cindex answering questions, etiquette @cindex mailing lists, guidelines If you consider your answer to have broad interest, you may want to post it to the mailing list instead of replying directly to the individual who asked. Try to make your answer general enough that people other than the original poster may benefit from it. When you post to the list, please make sure that your answer is not a duplication of a previous answer. Try to summarize the essential part of the question in your reply; don't feel obliged to quote the entire original message. Please don't post mail messages from your browser with HTML mode turned on! Many users don't read mail with a browser! @node Licensing and Support, Compatibility, MySQL Information Sources, Introduction @section MySQL Licensing and Support @cindex licensing terms @cindex support terms @menu * Licensing policy:: MySQL licensing policy * Copyright:: Copyrights used by MySQL * Licensing examples:: Example licensing situations * Cost:: MySQL licensing and support costs * Support:: Types of commercial support @end menu This section describes MySQL support and licensing arrangements: @itemize @bullet @item The copyrights under which MySQL is distributed (@pxref{Copyright}) @item Sample situations illustrating when a license is required (@pxref{Licensing examples}) @item Support costs (@pxref{Cost}) and support benefits (@pxref{Support}) @item Commercial licensing costs @end itemize @node Licensing policy, Copyright, Licensing and Support, Licensing and Support @subsection MySQL Licensing Policy @cindex licensing policy @cindex technical support, licensing @cindex support, licensing @cindex General Public License, MySQL The formal terms of the GPL license can be found at @ref{GPL license}. Basically, our licensing policy and interpretation of the GPL is as follows: Note that older versions of MySQL are still using a more @uref{http://www.mysql.com/support/arrangements/mypl.html, strict license}. See the documentation for that version for more information. If you need a commercial MySQL license, because the GPL license doesn't suit your application, you can buy one at @uref{https://order.mysql.com/}. For normal internal use, MySQL costs nothing. You do not have to pay us if you do not want to. A license is required if: @itemize @minus @item You link a program, that is not free software, with code from the MySQL server or clients that has the GPL copyright. This happens for example when you use MySQL as an embedded server in your applications or when you add not free extensions to the MySQL server. In this case your application/code would also become GPL through the GPL license that acts as a virus. By licensing MySQL server from MySQL AB under a commercial license you will avoid this problem. See @uref{http://www.gnu.org/copyleft/gpl-faq.html}. @item You have a commercial application that ONLY works with MySQL and ships the application with the MySQL server. This is because we view this as linking even if it is done over the network. @item You have a distribution of MySQL and you don't provide the source code for your copy of the MySQL server, as defined in the GPL license. @end itemize A license is @strong{NOT} required if: @itemize @minus @item You do not need a license to include the client code in commercial programs. The client part of MySQL licensed with the LGPL @code{GNU Library General Public License}. The @code{mysql} command-line client includes code from the @code{readline} library that is under the @code{GPL}. @item If your use of MySQL does not require a license, but you like MySQL and want to encourage further development, you are certainly welcome to purchase a license or MySQL support anyway. @item If you use MySQL in a commercial context such that you profit by its use, we ask that you further the development of MySQL by purchasing some level of support. We feel that if MySQL helps your business, it is reasonable to ask that you help MySQL. (Otherwise, if you ask us support questions, you are not only using for free something into which we've put a lot a work, you're asking us to provide free support, too.) @end itemize For circumstances under which a MySQL license is required, you need a license per machine that runs the @code{mysqld} server. However, a multiple-CPU machine counts as a single machine, and there is no restriction on the number of MySQL servers that run on one machine, or on the number of clients concurrently connected to a server running on that machine! If you have any questions as to whether or not a license is required for your particular use of MySQL, please read this again and then contact us. @xref{Contact information}. If you require a MySQL license, the easiest way to pay for it is to use the license form on MySQL's secure server at @uref{https://order.mysql.com/}. Other forms of payment are discussed in @ref{Payment information}. @node Copyright, Licensing examples, Licensing policy, Licensing and Support @subsection Copyrights Used by MySQL @cindex copyrights @menu * Copyright changes:: Possible future copyright changes @end menu There are several different copyrights on the MySQL distribution: @enumerate @item The MySQL-specific source needed to build the @code{mysqlclient} library is licensed under the @code{LGPL} and programs in the @file{client} directory is GPL. Each file has a header that shows which copyright is used for that file. @item The client library and the (GNU @code{getopt}) library are covered by the ``GNU LIBRARY GENERAL PUBLIC LICENSE.'' @xref{LGPL license}. @item Some parts of the source (the @code{regexp} library) are covered by a Berkeley-style copyright. @item All the source in the server and the (GNU @code{readline}) library is covered by the ``GNU GENERAL PUBLIC LICENSE.'' @xref{GPL license}. This is also available as the file @file{COPYING} in the distributions. @end enumerate One goal is that the SQL client library should be free enough that it is possible to add MySQL support into commercial products without a license. For this reason, we chose the LGPL license for the client code. @cindex licensing, free @cindex free licensing This means that you can use MySQL for free with any program that uses any of the free software licenses. MySQL is also free for any end user for his own or company usage. However, if you use MySQL for something important to you, you may want to help secure its development by purchasing licenses or a support contract. @xref{Support}. @node Copyright changes, , Copyright, Copyright @subsubsection Copyright Changes Version 3.22 of MySQL is still using a more strict license. See the documentation for that version for more information. @node Licensing examples, Cost, Copyright, Licensing and Support @subsection Example Licensing Situations @menu * Products that use MySQL:: Selling products that use MySQL * ISP:: ISP MySQL services * Web server:: Running a web server using MySQL. @end menu @cindex licensing, examples @cindex selling products @cindex products, selling This section describes some situations illustrating whether or not you must license the MySQL server. Generally these examples involve providing MySQL as an integral part of a product. Note that a single MySQL license covers any number of CPUs and @code{mysqld} servers on a machine! There is no artificial limit on the number of clients that connect to the server in any way. @node Products that use MySQL, ISP, Licensing examples, Licensing examples @subsubsection Selling Products that use MySQL To determine whether or not you need a MySQL license when selling your application, you should ask whether the proper functioning of your application is dependent on the use of MySQL and whether you include the MySQL server with your product. There are several cases to consider: @itemize @bullet @item Does your application require MySQL to function properly? @item If your product requires MySQL, you need a license for any machine that runs the @code{mysqld} server. For example, if you've designed your application around MySQL, then you've really made a commercial product that requires the engine, so you need a license. @item If your application does not require MySQL, you do not need to obtain a license. For example, if using MySQL just adds some new optional features to your product (such as adding logging to a database if MySQL is used rather than logging to a text file), it should fall within normal use, and a license would not be required. @item In other words, you need a license if you sell a product designed specifically for use with MySQL or that requires the MySQL server to function at all. This is true whether or not you provide MySQL for your client as part of your product distribution. @item It also depends on what you're doing for the client. Do you plan to provide your client with detailed instructions on installing MySQL with your software? Then your product may be contingent on the use of MySQL; if so, you need to buy a license. If you are simply tying into a database that you expect already to have been installed by the time your software is purchased, then you probably don't need a license. @end itemize @node ISP, Web server, Products that use MySQL, Licensing examples @subsubsection ISP MySQL Services @cindex ISP services @cindex services, ISP @cindex services, web @cindex Internet Service Providers Internet Service Providers (ISPs) often host MySQL servers for their customers. With the GPL license this does not require a license. On the other hand, we do encourage people to use ISPs that have MySQL support, as this will give them the confidence that if they have some problem with their MySQL installation, their ISP will be able to solve the problem for them (in some cases with the help from the MySQL development team). All ISPs that want to keep themselves up-to-date should subscribe to our @code{announce} mailing list so that they can be aware of fatal issues that may be relevant for their MySQL installations. Note that if the ISP doesn't have a license for MySQL, it should give its customers at least read access to the source of the MySQL installation so that its customer can verify that it is patched correctly. @node Web server, , ISP, Licensing examples @subsubsection Running a Web Server Using MySQL @cindex web server, running @cindex running, a web server If you use MySQL in conjunction with a Web server on Unix, you don't have to pay for a license. This is true even if you run a commercial Web server that uses MySQL, because you are not selling an embedded MySQL version yourself. However, in this case we would like you to purchase MySQL support, because MySQL is helping your enterprise. @node Cost, Support, Licensing examples, Licensing and Support @subsection MySQL Licensing and Support Costs @cindex costs, licensing and support @cindex licensing costs @cindex support costs @cindex prices, licensing and support @menu * Payment information:: Payment information * Contact information:: Contact information @end menu Our current license prices are shown below. To make a purchase, please visit @uref{https://order.mysql.com/}. If you pay by credit card, the currency is EURO (European Union Euro) so the prices will differ slightly. @multitable @columnfractions .3 .3 .3 @item @strong{Number of licenses} @tab @strong{Per copy} @item 1-9 @tab 230 EURO @item 10-24 @tab 138 EURO @item 25-49 @tab 117 EURO @item 50-99 @tab 102 EURO @item 100-249 @tab 91 EURO @item 250-499 @tab 76 EURO @item 500-999 @tab 66 EURO @end multitable For high volume (OEM) purchases, please contact @email{sales@@mysql.com}. For OEM purchases, you must act as the middle-man for eventual problems or extension requests from your users. We also require that OEM customers have at least an extended e-mail support contract. Note that OEM licenses only apply for products where the user doesn't have direct access to the MySQL server (embedded system). In other words, the MySQL server should only be used with the application that was supplied you. If you have a low-margin, high-volume product, you can always talk to us about other terms (for example, a percent of the sale price). If you do, please be informative about your product, pricing, market, and any other information that may be relevant. A full-price license is not a support agreement and includes very minimal support. This means that we try to answer any relevant questions. If the answer is in the documentation, we will direct you to the appropriate section. If you have not purchased a license or support, we probably will not answer at all. If you discover what we consider a real bug, we are likely to fix it in any case. But if you pay for support we will notify you about the fix status instead of just fixing it in a later release. More comprehensive support is sold separately. Descriptions of what each level of support includes are given in @ref{Support}. Costs for the various types of commercial support are shown below. Support level prices are in EURO (European Union Euro). One EURO is about 1.06 USD. @multitable @columnfractions .5 .5 @item @strong{Type of support} @tab @strong{Cost per year} @item Basic e-mail support. @xref{Basic email support}. @tab EURO 200 @item Extended e-mail support @xref{Extended email support}. @tab EURO 1000 @item Login support @xref{Login support}. @tab EURO 2000 @item Extended login support @xref{Extended login support}. @tab EURO 5000 @item Telephone support @xref{Telephone support}. @tab EURO 12000 @end multitable You may upgrade from any lower level of support to a higher level of support for the difference in price between the two support levels. We do also provide telephone support (mostly emergency support but also 24/7 support). This support option doesn't however have a fixed price but is negotiated for case to case. If you are interested in this option you can email @email{sales@@mysql.com} and tell us about your needs. Note that as our sales staff is very busy, it may take some time until your request is handled. Our support staff does however always answer promptly to support questions! @node Payment information, Contact information, Cost, Cost @subsubsection Payment information @cindex payment information Currently we can take SWIFT payments, checks, or credit cards. Payment should be made to: @example Postgirot Bank AB 105 06 STOCKHOLM, SWEDEN MySQL AB BOX 6434 11382 STOCKHOLM, SWEDEN SWIFT address: PGSI SESS Account number: 96 77 06 - 3 @end example Specify: license and/or support and your name and e-mail address. In Europe and Japan you can use EuroGiro (that should be less expensive) to the same account. If you want to pay by check, make it payable to ``MySQL Finland AB'' and mail it to the address below: @example MySQL AB BOX 6434, Torsgatan 21 11382 STOCKHOLM, SWEDEN @end example If you want to pay by credit card over the Internet, you can use @uref{https://order.mysql.com/, MySQL AB's secure license form}. You can also print a copy of the license form, fill it in, and send it by fax to: +46-8-729 69 05 If you want us to bill you, you can use the license form and write ``bill us'' in the comment field. You can also mail a message to @email{sales@@mysql.com} (@strong{not} @code{mysql@@lists.mysql.com}!) with your company information and ask us to bill you. @node Contact information, , Payment information, Cost @subsubsection Contact Information @cindex contact information @cindex licensing, contact information @cindex advertising, contact information @cindex employment, contact information @cindex partnering with MySQL @cindex employment with MySQL @cindex jobs at MySQL For commercial licensing, please contact the MySQL licensing team. The much preferred method is by e-mail to @email{licensing@@mysql.com}. Fax is also possible but handling of these may take much longer (Fax +46-8-729 69 05). If you represent a business that is interested in partnering with MySQL, please send e-mail to @email{partner@@mysql.com}. For timely, precise answers to technical questions about MySQL you should @uref{https://order.mysql.com/, order} one of our @uref{http://www.mysql.com/support/arrangements/types.html, support contracts}. MySQL support is provided by the MySQL developers so the standard is extremely high. If you are interested in placing a banner advertisement on our Web site, please send e-mail to @email{advertising@@mysql.com}. If you are interested in any of the jobs listed in our @uref{http://www.mysql.com/development/jobs/, jobs} section, please send e-mail to @email{jobs@@mysql.com}. For general discussion amongst our many users, please direct your attention to the appropriate @uref{http://www.mysql.com/documentation/lists.html, mailing list}. For general information inquires, please send e-mail to @email{info@@mysql.com}. For questions or comments about the workings or content of the Web site, please send e-mail to @email{webmaster@@mysql.com}. @node Support, , Cost, Licensing and Support @subsection Types of Commercial Support @cindex support, types @cindex types, of support @cindex commercial support, types @menu * Basic email support:: Basic email support * Extended email support:: Extended email support * Login support:: Login support * Extended login support:: Extended login support * Telephone support:: Telephone support * Table handler support:: Support for other table handlers @end menu The following is true of all support options: @itemize @bullet @item The support is per year. @item We will fix, or provide a reasonable workaround for any repeatable bug. @item We will give a reasonable effort to find and fix any other MySQL related bug. @item The higher level of support contract the more effort we will put into finding a solution to your problems. @item The following is true for all support contracts except Basic email support: For non-bug related things, like helping you optimize your queries or your system, extending MySQL with new functionality, etc., we charge 200 EURO/hour, which is deducted from your support contract. In other words, if you have login support (2000 EURO), you can expect us to work up to 10 hours to help you with things like this. @end itemize @node Basic email support, Extended email support, Support, Support @subsubsection Basic E-mail Support @cindex email, technical support @cindex technical support, by email Basic e-mail support is a very inexpensive support option and should be thought of more as a way to support our development of MySQL than as a real support option. We at MySQL do give a lot of free support in all the different MySQL lists, and the money we get from basic e-mail support is largely used to make this possible. At this support level, the MySQL mailing lists are the preferred means of communication. Questions normally should be mailed to the primary mailing list (@email{mysql@@lists.mysql.com}) or one of the other regular lists (for example, @email{win32@@lists.mysql.com} for Windows-related MySQL questions), as someone else already may have experienced and solved the problem you have. @xref{Asking questions}. However, by purchasing basic e-mail support, you also have access to the support address @email{mysql-support@@mysql.com}, which is not available as part of the minimal support that you get by purchasing a MySQL license. This means that for especially critical questions, you can cross-post your message to @email{mysql-support@@mysql.com}. (If the message contains sensitive data, you should post only to @email{mysql-support@@mysql.com}.) @strong{REMEMBER!} to ALWAYS include your registration number and expiration date when you send a message to @email{mysql-support@@mysql.com}. Note that if you have encountered a critical, repeatable bug, and follow the rules outlined in the manual section of how to report bugs and send it to @email{bugs@@lists.mysql.com}, we promise to try to fix this as soon as possible, regardless of your support level! @xref{Bug reports}. Basic e-mail support includes the following types of service: @itemize @bullet @item If your question is already answered in the manual, we will inform you of the correct section in which you can find the answer. If the answer is not in the manual, we will point you in the right direction to solve your problem. @item We guarantee a timely answer for your e-mail messages. We can't guarantee that we can solve any problem, but at least you will receive an answer if we can contact you by e-mail. @item We will help with unexpected problems when you install MySQL from a binary distribution on supported platforms. This level of support does not cover installing MySQL from a source distribution. Supported platforms are those for which MySQL is known to work. @xref{Which OS}. @item We will help you with bugs and missing features. Any bugs that are found are fixed for the next MySQL release. If the bug is critical for you, we will mail you a patch for it as soon the bug is fixed. Critical bugs always have the highest priority for us, and we ensure that they are fixed as soon as possible. @item Your suggestions for the further development of MySQL will be taken into consideration. By taking email support you have already helped the further development of MySQL. If you want to have more input, upgrade to a higher level of support. @item If you want us to help optimize your system, you must upgrade to a higher level of support. @end itemize @node Extended email support, Login support, Basic email support, Support @subsubsection Extended E-mail Support @cindex extended email support Extended e-mail support includes everything in basic e-mail support with these additions: @itemize @bullet @item Your e-mail will be dealt with before mail from basic e-mail support users and non-registered users. @item Your suggestions for the further development of MySQL will receive strong consideration. Simple extensions that suit the basic goals of MySQL are implemented in a matter of days. By taking extended e-mail support you have already helped the further development of MySQL. @item Typical situations that are covered by extended e-mail support are: @itemize @minus @item We will answer and (within reason) solve questions that relate to possible bugs in MySQL. As soon as the bug is found and corrected, we will mail a patch for it. @item We will help with unexpected problems when you install MySQL from a source or binary distribution on supported platforms. @item We will answer questions about missing features and offer hints how to work around them. @item We will provide hints on optimizing @code{mysqld} for your situation. @end itemize @item You are allowed to influence the priority of items on the MySQL TODO List. @xref{TODO}. This will ensure that the features you really need will be implemented sooner than they might be otherwise. @end itemize @node Login support, Extended login support, Extended email support, Support @subsubsection Login Support @cindex login support Login support includes everything in extended e-mail support with these additions: @itemize @bullet @item Your e-mail will be dealt with even before e-mail from extended e-mail support users. @item Your suggestions for the further development of MySQL will be taken into very high consideration. Realistic extensions that can be implemented in a couple of hours and that suit the basic goals of MySQL will be implemented as soon as possible. @item If you have a very specific problem, we can try to log in on your system to solve the problem ``in place.'' @item Like any database vendor, we can't guarantee that we can rescue any data from crashed tables, but if the worst happens, we will help you rescue as much as possible. MySQL has proven itself very reliable, but anything is possible due to circumstances beyond our control (for example, if your system crashes or someone kills the server by executing a @code{kill -9} command). @item We will provide hints on optimizing your system and your queries. @item You are allowed to call a MySQL developer (in moderation) and discuss your MySQL-related problems. This option is however only to be used as a last result during an emergency after we have failed to grasp the total problem with email. To make efficient use of our time we need to first get all facts about the problem, before talking on phone, to be able to work as efficiently as possible on solving the problem. @end itemize @node Extended login support, Telephone support, Login support, Support @subsubsection Extended Login Support Extended login support includes everything in login support with these additions: @itemize @bullet @item Your e-mail has the highest possible priority. @item We will actively examine your system and help you optimize it and your queries. We may also optimize and/or extend MySQL to better suit your needs. @item You may also request special extensions just for you. For example: @example mysql> select MY_FUNC(col1,col2) from table; @end example @item We will provide a binary distribution of all important MySQL releases for your system, as long as we can get an account on a similar system. In the worst case, we may require access to your system to be able to create a binary distribution. @item If you can provide accommodations and pay for traveler fares, you can even get a MySQL developer to visit you and offer you help with your troubles. Extended login support entitles you to one personal encounter per year, but we are always very flexible towards our customers! If the visit takes 16 hours or more, the first 8 hours is without charge. For the hours above 8 hours, you will be charged with a rate that is at least 20 % less than our standard rates. @end itemize @node Telephone support, Table handler support, Extended login support, Support @subsubsection Telephone Support Telephone support includes everything in extended login support with these additions: @itemize @bullet @item We will provide you with a dynamic web page showing the current list of @code{MySQL} developers that you can phone when you have a critical problem. @item For non critical problem, you can request a MySQL developer to phone back within 48 hours to discuss @code{MySQL} related issues. @end itemize @node Table handler support, , Telephone support, Support @subsubsection Support for other table handlers @cindex support, BDB Tables @cindex support, InnoDB Tables To get support for @code{BDB} tables, @code{InnoDB} tables you have to pay an additional 30% on the standard support price for each of the table handlers you would like to have support for. We at @code{MySQL AB} will help you create a proper bug report for the table handler and submit it to the developers for the specific table handler. We will also do our best to ensure that you will get a timely answer or solution from the developers of the table handler. Even if we are quite confident that we can solve most problems within a timely manner, we can't guarantee a quick solution for any problems you can get with the different table handlers. We will however do our best to help you get the problem solved. @node Compatibility, Comparisons, Licensing and Support, Introduction @section How Standards-compatible Is MySQL? @cindex compatibility, with ANSI SQL @cindex standards compatibility @cindex extensions, to ANSI SQL @cindex ANSI SQL92, extensions to @menu * Extensions to ANSI:: MySQL extensions to ANSI SQL92 * Differences from ANSI:: MySQL differences compared to ANSI SQL92 * ANSI mode:: Running MySQL in ANSI mode * Missing functions:: Functionality missing from MySQL * Standards:: What standards does MySQL follow? * Commit-rollback:: How to cope without @code{COMMIT}-@code{ROLLBACK} * Bugs:: Known errors and design deficiencies in MySQL @end menu This section describes how MySQL relates to the ANSI SQL standards. MySQL has many extensions to the ANSI SQL standards, and here you will find out what they are, and how to use them. You will also find information about functionality missing from MySQL, and how to work around some differences. @node Extensions to ANSI, Differences from ANSI, Compatibility, Compatibility @subsection MySQL Extensions to ANSI SQL92 MySQL includes some extensions that you probably will not find in other SQL databases. Be warned that if you use them, your code will not be portable to other SQL servers. In some cases, you can write code that includes MySQL extensions, but is still portable, by using comments of the form @code{/*! ... */}. In this case, MySQL will parse and execute the code within the comment as it would any other MySQL statement, but other SQL servers will ignore the extensions. For example: @example SELECT /*! STRAIGHT_JOIN */ col_name FROM table1,table2 WHERE ... @end example If you add a version number after the @code{'!'}, the syntax will only be executed if the MySQL version is equal to or newer than the used version number: @example CREATE /*!32302 TEMPORARY */ TABLE (a int); @end example The above means that if you have Version 3.23.02 or newer, then MySQL will use the @code{TEMPORARY} keyword. MySQL extensions are listed below: @itemize @bullet @item The field types @code{MEDIUMINT}, @code{SET}, @code{ENUM}, and the different @code{BLOB} and @code{TEXT} types. @item The field attributes @code{AUTO_INCREMENT}, @code{BINARY}, @code{NULL}, @code{UNSIGNED}, and @code{ZEROFILL}. @item All string comparisons are case insensitive by default, with sort ordering determined by the current character set (ISO-8859-1 Latin1 by default). If you don't like this, you should declare your columns with the @code{BINARY} attribute or use the @code{BINARY} cast, which causes comparisons to be done according to the ASCII order used on the MySQL server host. @item MySQL maps each database to a directory under the MySQL data directory, and tables within a database to filenames in the database directory. This has a few implications: @cindex database names, case sensitivity @cindex table names, case sensitivity @cindex case sensitivity, of database names @cindex case sensitivity, of table names @itemize @minus @item Database names and table names are case sensitive in MySQL on operating systems that have case-sensitive filenames (like most Unix systems). @xref{Name case sensitivity}. @item Database, table, index, column, or alias names may begin with a digit (but may not consist solely of digits). @item You can use standard system commands to backup, rename, move, delete, and copy tables. For example, to rename a table, rename the @file{.MYD}, @file{.MYI}, and @file{.frm} files to which the table corresponds. @end itemize @item In SQL statements, you can access tables from different databases with the @code{db_name.tbl_name} syntax. Some SQL servers provide the same functionality but call this @code{User space}. MySQL doesn't support tablespaces as in: @code{create table ralph.my_table...IN my_tablespace}. @item @code{LIKE} is allowed on numeric columns. @item Use of @code{INTO OUTFILE} and @code{STRAIGHT_JOIN} in a @code{SELECT} statement. @xref{SELECT, , @code{SELECT}}. @item The @code{SQL_SMALL_RESULT} option in a @code{SELECT} statement. @item @code{EXPLAIN SELECT} to get a description on how tables are joined. @item Use of index names, indexes on a prefix of a field, and use of @code{INDEX} or @code{KEY} in a @code{CREATE TABLE} statement. @xref{CREATE TABLE, , @code{CREATE TABLE}}. @item Use of @code{TEMPORARY} or @code{IF NOT EXISTS} with @code{CREATE TABLE}. @item Use of @code{COUNT(DISTINCT list)} where 'list' is more than one element. @item Use of @code{CHANGE col_name}, @code{DROP col_name}, or @code{DROP INDEX}, @code{IGNORE} or @code{RENAME} in an @code{ALTER TABLE} statement. @xref{ALTER TABLE, , @code{ALTER TABLE}}. @item Use of @code{RENAME TABLE}. @xref{RENAME TABLE, , @code{RENAME TABLE}}. @item Use of multiple @code{ADD}, @code{ALTER}, @code{DROP}, or @code{CHANGE} clauses in an @code{ALTER TABLE} statement. @item Use of @code{DROP TABLE} with the keywords @code{IF EXISTS}. @item You can drop multiple tables with a single @code{DROP TABLE} statement. @item The @code{LIMIT} clause of the @code{DELETE} statement. @item The @code{DELAYED} clause of the @code{INSERT} and @code{REPLACE} statements. @item The @code{LOW_PRIORITY} clause of the @code{INSERT}, @code{REPLACE}, @code{DELETE}, and @code{UPDATE} statements. @cindex Oracle compatibility @cindex compatibility, with Oracle @item Use of @code{LOAD DATA INFILE}. In many cases, this syntax is compatible with Oracle's @code{LOAD DATA INFILE}. @xref{LOAD DATA, , @code{LOAD DATA}}. @item The @code{ANALYZE TABLE}, @code{CHECK TABLE}, @code{OPTIMIZE TABLE}, and @code{REPAIR TABLE} statements. @item The @code{SHOW} statement. @xref{SHOW, , @code{SHOW}}. @item Strings may be enclosed by either @samp{"} or @samp{'}, not just by @samp{'}. @item Use of the escape @samp{\} character. @item The @code{SET OPTION} statement. @xref{SET OPTION, , @code{SET OPTION}}. @item You don't need to name all selected columns in the @code{GROUP BY} part. This gives better performance for some very specific, but quite normal queries. @xref{Group by functions}. @item One can specify @code{ASC} and @code{DESC} with @code{GROUP BY}. @item To make it easier for users who come from other SQL environments, MySQL supports aliases for many functions. For example, all string functions support both ANSI SQL syntax and ODBC syntax. @item MySQL understands the @code{||} and @code{&&} operators to mean logical OR and AND, as in the C programming language. In MySQL, @code{||} and @code{OR} are synonyms, as are @code{&&} and @code{AND}. Because of this nice syntax, MySQL doesn't support the ANSI SQL @code{||} operator for string concatenation; use @code{CONCAT()} instead. Because @code{CONCAT()} takes any number of arguments, it's easy to convert use of the @code{||} operator to MySQL. @item @code{CREATE DATABASE} or @code{DROP DATABASE}. @xref{CREATE DATABASE, , @code{CREATE DATABASE}}. @cindex PostgreSQL compatibility @cindex compatibility, with PostgreSQL @item The @code{%} operator is a synonym for @code{MOD()}. That is, @code{N % M} is equivalent to @code{MOD(N,M)}. @code{%} is supported for C programmers and for compatibility with PostgreSQL. @item The @code{=}, @code{<>}, @code{<=} ,@code{<}, @code{>=},@code{>}, @code{<<}, @code{>>}, @code{<=>}, @code{AND}, @code{OR}, or @code{LIKE} operators may be used in column comparisons to the left of the @code{FROM} in @code{SELECT} statements. For example: @example mysql> SELECT col1=1 AND col2=2 FROM tbl_name; @end example @item The @code{LAST_INSERT_ID()} function. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @item The @code{REGEXP} and @code{NOT REGEXP} extended regular expression operators. @item @code{CONCAT()} or @code{CHAR()} with one argument or more than two arguments. (In MySQL, these functions can take any number of arguments.) @item The @code{BIT_COUNT()}, @code{CASE}, @code{ELT()}, @code{FROM_DAYS()}, @code{FORMAT()}, @code{IF()}, @code{PASSWORD()}, @code{ENCRYPT()}, @code{md5()}, @code{ENCODE()}, @code{DECODE()}, @code{PERIOD_ADD()}, @code{PERIOD_DIFF()}, @code{TO_DAYS()}, or @code{WEEKDAY()} functions. @item Use of @code{TRIM()} to trim substrings. ANSI SQL only supports removal of single characters. @item The @code{GROUP BY} functions @code{STD()}, @code{BIT_OR()}, and @code{BIT_AND()}. @item Use of @code{REPLACE} instead of @code{DELETE} + @code{INSERT}. @xref{REPLACE, , @code{REPLACE}}. @item The @code{FLUSH}, @code{RESET} and @code{DO} statements. @item The possibility to set variables in a statement with @code{:=}: @example SELECT @@a:=SUM(total),@@b=COUNT(*),@@a/@@b AS avg FROM test_table; SELECT @@t1:=(@@t2:=1)+@@t3:=4,@@t1,@@t2,@@t3; @end example @end itemize @node Differences from ANSI, ANSI mode, Extensions to ANSI, Compatibility @subsection MySQL Differences Compared to ANSI SQL92 We try to make MySQL follow the ANSI SQL standard and the ODBC SQL standard, but in some cases MySQL does some things differently: @itemize @bullet @item @code{--} is only a comment if followed by a white space. @xref{Missing comments}. @item For @code{VARCHAR} columns, trailing spaces are removed when the value is stored. @xref{Bugs}. @item In some cases, @code{CHAR} columns are silently changed to @code{VARCHAR} columns. @xref{Silent column changes}. @item Privileges for a table are not automatically revoked when you delete a table. You must explicitly issue a @code{REVOKE} to revoke privileges for a table. @xref{GRANT, , @code{GRANT}}. @item @code{NULL AND FALSE} will evaluate to @code{NULL} and not to @code{FALSE}. This is because we don't think it's good to have to evaluate a lot of extra conditions in this case. @end itemize @node ANSI mode, Missing functions, Differences from ANSI, Compatibility @subsection Running MySQL in ANSI Mode @cindex running, ANSI mode @cindex ANSI mode, running If you start @code{mysqld} with the @code{--ansi} option, the following behavior of MySQL changes: @itemize @bullet @item @code{||} is string concatenation instead of @code{OR}. @item You can have any number of spaces between a function name and the @samp{(}. This forces all function names to be treated as reserved words. @item @samp{"} will be an identifier quote character (like the MySQL @samp{`} quote character) and not a string quote character. @item @code{REAL} will be a synonym for @code{FLOAT} instead of a synonym of @code{DOUBLE}. @item The default transaction isolation level is @code{SERIALIZABLE}. @xref{SET TRANSACTION}. @end itemize This is the same as using @code{--sql-mode=REAL_AS_FLOAT,PIPES_AS_CONCAT,ANSI_QUOTES,IGNORE_SPACE,SERIALIZE,ONLY_FULL_GROUP_BY}. @node Missing functions, Standards, ANSI mode, Compatibility @subsection Functionality Missing from MySQL @cindex missing functionality @cindex functionality, missing The following functionality is missing in the current version of MySQL. For a prioritized list indicating when new extensions may be added to MySQL, you should consult @uref{http://www.mysql.com/documentation/manual.php?section=TODO, the online MySQL TODO list}. That is the latest version of the TODO list in this manual. @xref{TODO}. @menu * Missing Sub-selects:: Sub-selects * Missing SELECT INTO TABLE:: @code{SELECT INTO TABLE} * Missing Transactions:: Transactions * Missing Triggers:: Triggers * Missing Foreign Keys:: Foreign Keys * Broken Foreign KEY:: Reasons NOT to Use Foreign Keys constraints * Missing Views:: Views * Missing comments:: @samp{--} as the start of a comment @end menu @node Missing Sub-selects, Missing SELECT INTO TABLE, Missing functions, Missing functions @subsubsection Sub-selects @cindex sub-selects MySQL currently only supports sub selects of the form @code{INSERT ... SELECT ...} and @code{REPLACE ... SELECT ...}. You can however use the function @code{IN()} in other contexts. In many cases you can rewrite the query without a sub-select: @example SELECT * FROM table1 WHERE id IN (SELECT id FROM table2); @end example This can be re-written as: @example SELECT table1.* FROM table1,table2 WHERE table1.id=table2.id; @end example The queries: @example SELECT * FROM table1 WHERE id NOT IN (SELECT id FROM table2); SELECT * FROM table1 WHERE NOT EXISTS (SELECT id FROM table2 where table1.id=table2.id); @end example Can be rewritten as: @example SELECT table1.* FROM table1 LEFT JOIN table2 ON table1.id=table2.id where table2.id IS NULL @end example For more complicated subqueries you can often create temporary tables to hold the subquery. In some cases, however this option will not work. The most frequently encountered of these cases arises with @code{DELETE} statements, for which standard SQL does not support joins (except in sub-selects). For this situation there are two options available until subqueries are supported by MySQL. The first option is to use a procedural programming language (such as Perl or PHP) to submit a @code{SELECT} query to obtain the primary keys for the records to be deleted, and then use these values to construct the @code{DELETE} statement (@code{DELETE FROM ... WHERE ... IN (key1, key2, ...)}). The second option is to use interactive SQL to contruct a set of @code{DELETE} statements automatically, using the MySQL extension @code{CONCAT()} (in lieu of the standard @code{||} operator). For example: @example SELECT CONCAT('DELETE FROM tab1 WHERE pkid = ', tab1.pkid, ';') FROM tab1, tab2 WHERE tab1.col1 = tab2.col2; @end example You can place this query in a script file and redirect input from it to the @code{mysql} command-line interpreter, piping its output back to a second instance of the interpreter: @example prompt> mysql --skip-column-names mydb < myscript.sql | mysql mydb @end example MySQL 4.0 supports multi-table deletes that can be used to efficiently delete rows based on information from one table or even from many tables at the same time. @node Missing SELECT INTO TABLE, Missing Transactions, Missing Sub-selects, Missing functions @subsubsection @code{SELECT INTO TABLE} @findex SELECT INTO TABLE MySQL doesn't yet support the Oracle SQL extension: @code{SELECT ... INTO TABLE ...}. MySQL supports instead the ANSI SQL syntax @code{INSERT INTO ... SELECT ...}, which is basically the same thing. @xref{INSERT SELECT}. @example INSERT INTO tblTemp2 (fldID) SELECT tblTemp1.fldOrder_ID FROM tblTemp1 WHERE tblTemp1.fldOrder_ID > 100; @end example Alternatively, you can use @code{SELECT INTO OUTFILE...} or @code{CREATE TABLE ... SELECT} to solve your problem. @node Missing Transactions, Missing Triggers, Missing SELECT INTO TABLE, Missing functions @subsubsection Transactions @cindex transactions, support As MySQL does nowadays support transactions, the following discussion is only valid if you are only using the non-transaction-safe table types. @xref{COMMIT}. The question is often asked, by the curious and the critical, ``Why is MySQL not a transactional database?'' or ``Why does MySQL not support transactions?'' MySQL has made a conscious decision to support another paradigm for data integrity, ``atomic operations.'' It is our thinking and experience that atomic operations offer equal or even better integrity with much better performance. We, nonetheless, appreciate and understand the transactional database paradigm and plan, within the next few releases, to introduce transaction-safe tables on a per table basis. We will be giving our users the possibility to decide if they need the speed of atomic operations or if they need to use transactional features in their applications. How does one use the features of MySQL to maintain rigorous integrity and how do these features compare with the transactional paradigm? First, in the transactional paradigm, if your applications are written in a way that is dependent on the calling of ``rollback'' instead of ``commit'' in critical situations, then transactions are more convenient. Moreover, transactions ensure that unfinished updates or corrupting activities are not committed to the database; the server is given the opportunity to do an automatic rollback and your database is saved. MySQL, in almost all cases, allows you to solve for potential problems by including simple checks before updates and by running simple scripts that check the databases for inconsistencies and automatically repair or warn if such occurs. Note that just by using the MySQL log or even adding one extra log, one can normally fix tables perfectly with no data integrity loss. Moreover, fatal transactional updates can be rewritten to be atomic. In fact,we will go so far as to say that all integrity problems that transactions solve can be done with @code{LOCK TABLES} or atomic updates, ensuring that you never will get an automatic abort from the database, which is a common problem with transactional databases. Not even transactions can prevent all loss if the server goes down. In such cases even a transactional system can lose data. The difference between different systems lies in just how small the time-lap is where they could lose data. No system is 100% secure, only ``secure enough.'' Even Oracle, reputed to be the safest of transactional databases, is reported to sometimes lose data in such situations. To be safe with MySQL, you only need to have backups and have the update logging turned on. With this you can recover from any situation that you could with any transactional database. It is, of course, always good to have backups, independent of which database you use. The transactional paradigm has its benefits and its drawbacks. Many users and application developers depend on the ease with which they can code around problems where an abort appears to be, or is necessary, and they may have to do a little more work with MySQL to either think differently or write more. If you are new to the atomic operations paradigm, or more familiar or more comfortable with transactions, do not jump to the conclusion that MySQL has not addressed these issues. Reliability and integrity are foremost in our minds. Recent estimates indicate that there are more than 1,000,000 @code{mysqld} servers currently running, many of which are in production environments. We hear very, very seldom from our users that they have lost any data, and in almost all of those cases user error is involved. This is, in our opinion, the best proof of MySQL's stability and reliability. Lastly, in situations where integrity is of highest importance, MySQL's current features allow for transaction-level or better reliability and integrity. If you lock tables with @code{LOCK TABLES}, all updates will stall until any integrity checks are made. If you only obtain a read lock (as opposed to a write lock), then reads and inserts are still allowed to happen. The new inserted records will not be seen by any of the clients that have a @code{READ} lock until they release their read locks. With @code{INSERT DELAYED} you can queue inserts into a local queue, until the locks are released, without having the client wait for the insert to complete. @xref{INSERT DELAYED}. ``Atomic,'' in the sense that we mean it, is nothing magical. It only means that you can be sure that while each specific update is running, no other user can interfere with it, and there will never be an automatic rollback (which can happen on transaction based systems if you are not very careful). MySQL also guarantees that there will not be any dirty reads. You can find some example of how to write atomic updates in the commit-rollback section. @xref{Commit-rollback}. We have thought quite a bit about integrity and performance, and we believe that our atomic operations paradigm allows for both high reliability and extremely high performance, on the order of three to five times the speed of the fastest and most optimally tuned of transactional databases. We didn't leave out transactions because they are hard to do. The main reason we went with atomic operations as opposed to transactions is that by doing this we could apply many speed optimizations that would not otherwise have been possible. Many of our users who have speed foremost in their minds are not at all concerned about transactions. For them transactions are not an issue. For those of our users who are concerned with or have wondered about transactions vis-a-vis MySQL, there is a ``MySQL way'' as we have outlined above. For those where safety is more important than speed, we recommend them to use the @code{BDB}, or @code{InnoDB} tables for all their critical data. @xref{Table types}. One final note: We are currently working on a safe replication schema that we believe to be better than any commercial replication system we know of. This system will work most reliably under the atomic operations, non-transactional, paradigm. Stay tuned. @node Missing Triggers, Missing Foreign Keys, Missing Transactions, Missing functions @subsubsection Stored Procedures and Triggers @cindex stored procedures and triggers, defined @cindex procedures, stored @cindex triggers, stored A stored procedure is a set of SQL commands that can be compiled and stored in the server. Once this has been done, clients don't need to keep reissuing the entire query but can refer to the stored procedure. This provides better performance because the query has to be parsed only once, and less information needs to be sent between the server and the client. You can also raise the conceptual level by having libraries of functions in the server. A trigger is a stored procedure that is invoked when a particular event occurs. For example, you can install a stored procedure that is triggered each time a record is deleted from a transaction table and that automatically deletes the corresponding customer from a customer table when all his transactions are deleted. The planned update language will be able to handle stored procedures, but without triggers. Triggers usually slow down everything, even queries for which they are not needed. To see when MySQL might get stored procedures, see @ref{TODO}. @node Missing Foreign Keys, Broken Foreign KEY, Missing Triggers, Missing functions @subsubsection Foreign Keys @cindex foreign keys @cindex keys, foreign Note that foreign keys in SQL are not used to join tables, but are used mostly for checking referential integrity (foreign key constraints). If you want to get results from multiple tables from a @code{SELECT} statement, you do this by joining tables: @example SELECT * from table1,table2 where table1.id = table2.id; @end example @xref{JOIN, , @code{JOIN}}. @xref{example-Foreign keys}. The @code{FOREIGN KEY} syntax in MySQL exists only for compatibility with other SQL vendors' @code{CREATE TABLE} commands; it doesn't do anything. The @code{FOREIGN KEY} syntax without @code{ON DELETE ...} is mostly used for documentation purposes. Some ODBC applications may use this to produce automatic @code{WHERE} clauses, but this is usually easy to override. @code{FOREIGN KEY} is sometimes used as a constraint check, but this check is unnecessary in practice if rows are inserted into the tables in the right order. MySQL only supports these clauses because some applications require them to exist (regardless of whether or not they work). In MySQL, you can work around the problem of @code{ON DELETE ...} not being implemented by adding the appropriate @code{DELETE} statement to an application when you delete records from a table that has a foreign key. In practice this is as quick (in some cases quicker) and much more portable than using foreign keys. In the near future we will extend the @code{FOREIGN KEY} implementation so that at least the information will be saved in the table specification file and may be retrieved by @code{mysqldump} and ODBC. At a later stage we will implement the foreign key constraints for application that can't easily be coded to avoid them. @menu * Broken Foreign KEY:: Reasons NOT to use foreign keys constraints @end menu @node Broken Foreign KEY, Missing Views, Missing Foreign Keys, Missing functions @subsubsection Reasons NOT to Use Foreign Keys constraints @cindex foreign keys, reasons not to use There are so many problems with foreign key constraints that we don't know where to start: @itemize @bullet @item Foreign key constraints make life very complicated, because the foreign key definitions must be stored in a database and implementing them would destroy the whole ``nice approach'' of using files that can be moved, copied, and removed. @item The speed impact is terrible for @code{INSERT} and @code{UPDATE} statements, and in this case almost all @code{FOREIGN KEY} constraint checks are useless because you usually insert records in the right tables in the right order, anyway. @item There is also a need to hold locks on many more tables when updating one table, because the side effects can cascade through the entire database. It's MUCH faster to delete records from one table first and subsequently delete them from the other tables. @item You can no longer restore a table by doing a full delete from the table and then restoring all records (from a new source or from a backup). @item If you use foreign key constraints you can't dump and restore tables unless you do so in a very specific order. @item It's very easy to do ``allowed'' circular definitions that make the tables impossible to re-create each table with a single create statement, even if the definition works and is usable. @item It's very easy to overlook @code{FOREIGN KEY ... ON DELETE} rules when one codes an application. It's not unusual that one loses a lot of important information just because a wrong or misused @code{ON DELETE} rule. @end itemize The only nice aspect of @code{FOREIGN KEY} is that it gives ODBC and some other client programs the ability to see how a table is connected and to use this to show connection diagrams and to help in building applications. MySQL will soon store @code{FOREIGN KEY} definitions so that a client can ask for and receive an answer about how the original connection was made. The current @file{.frm} file format does not have any place for it. At a later stage we will implement the foreign key constraints for application that can't easily be coded to avoid them. @node Missing Views, Missing comments, Broken Foreign KEY, Missing functions @subsubsection Views @cindex views MySQL doesn't yet support views, but we plan to implement these to about 4.1. Views are mostly useful for letting users access a set of relations as one table (in read-only mode). Many SQL databases don't allow one to update any rows in a view, but you have to do the updates in the separate tables. As MySQL is mostly used in applications and on web system where the application writer has full control on the database usage, most of our users haven't regarded views to be very important. (At least no one has been interested enough in this to be prepared to finance the implementation of views). One doesn't need views in MySQL to restrict access to columns as MySQL has a very sophisticated privilege system. @xref{Privilege system}. @node Missing comments, , Missing Views, Missing functions @subsubsection @samp{--} as the Start of a Comment @cindex comments, starting @cindex starting, comments Some other SQL databases use @samp{--} to start comments. MySQL has @samp{#} as the start comment character, even if the @code{mysql} command-line tool removes all lines that start with @samp{--}. You can also use the C comment style @code{/* this is a comment */} with MySQL. @xref{Comments}. MySQL Version 3.23.3 and above supports the @samp{--} comment style only if the comment is followed by a space. This is because this degenerate comment style has caused many problems with automatically generated SQL queries that have used something like the following code, where we automatically insert the value of the payment for @code{!payment!}: @example UPDATE tbl_name SET credit=credit-!payment! @end example What do you think will happen when the value of @code{payment} is negative? Because @code{1--1} is legal in SQL, we think it is terrible that @samp{--} means start comment. In MySQL Version 3.23 you can, however, use: @code{1-- This is a comment} The following discussion only concerns you if you are running a MySQL version earlier than Version 3.23: If you have a SQL program in a text file that contains @samp{--} comments you should use: @example shell> replace " --" " #" < text-file-with-funny-comments.sql \ | mysql database @end example instead of the usual: @example shell> mysql database < text-file-with-funny-comments.sql @end example You can also edit the command file ``in place'' to change the @samp{--} comments to @samp{#} comments: @example shell> replace " --" " #" -- text-file-with-funny-comments.sql @end example Change them back with this command: @example shell> replace " #" " --" -- text-file-with-funny-comments.sql @end example @node Standards, Commit-rollback, Missing functions, Compatibility @subsection What Standards Does MySQL Follow? Entry level SQL92. ODBC levels 0-2. @node Commit-rollback, Bugs, Standards, Compatibility @subsection How to Cope Without @code{COMMIT}/@code{ROLLBACK} @findex COMMIT @findex ROLLBACK @cindex transaction-safe tables @cindex tables, updating @cindex updating, tables @cindex @code{BDB} tables @cindex @code{InnoDB} tables The following mostly applies only for @code{ISAM}, @code{MyISAM}, and @code{HEAP} tables. If you only use transaction-safe tables (@code{BDB}, or @code{InnoDB} tables) in an an update, you can do @code{COMMIT} and @code{ROLLBACK} also with MySQL. @xref{COMMIT}. The problem with handling @code{COMMIT}-@code{ROLLBACK} efficiently with the above table types would require a completely different table layout than MySQL uses today. The table type would also need extra threads that do automatic cleanups on the tables, and the disk usage would be much higher. This would make these table types about 2-4 times slower than they are today. For the moment, we prefer implementing the SQL server language (something like stored procedures). With this you would very seldom really need @code{COMMIT}-@code{ROLLBACK.} This would also give much better performance. Loops that need transactions normally can be coded with the help of @code{LOCK TABLES}, and you don't need cursors when you can update records on the fly. We at TcX had a greater need for a real fast database than a 100% general database. Whenever we find a way to implement these features without any speed loss, we will probably do it. For the moment, there are many more important things to do. Check the TODO for how we prioritize things at the moment. (Customers with higher levels of support can alter this, so things may be reprioritized.) The current problem is actually @code{ROLLBACK}. Without @code{ROLLBACK}, you can do any kind of @code{COMMIT} action with @code{LOCK TABLES}. To support @code{ROLLBACK} with the above table types, MySQL would have to be changed to store all old records that were updated and revert everything back to the starting point if @code{ROLLBACK} was issued. For simple cases, this isn't that hard to do (the current @code{isamlog} could be used for this purpose), but it would be much more difficult to implement @code{ROLLBACK} for @code{ALTER/DROP/CREATE TABLE}. To avoid using @code{ROLLBACK}, you can use the following strategy: @enumerate @item Use @code{LOCK TABLES ...} to lock all the tables you want to access. @item Test conditions. @item Update if everything is okay. @item Use @code{UNLOCK TABLES} to release your locks. @end enumerate This is usually a much faster method than using transactions with possible @code{ROLLBACK}s, although not always. The only situation this solution doesn't handle is when someone kills the threads in the middle of an update. In this case, all locks will be released but some of the updates may not have been executed. You can also use functions to update records in a single operation. You can get a very efficient application by using the following techniques: @itemize @bullet @item Modify fields relative to their current value. @item Update only those fields that actually have changed. @end itemize For example, when we are doing updates to some customer information, we update only the customer data that has changed and test only that none of the changed data, or data that depend on the changed data, has changed compared to the original row. The test for changed data is done with the @code{WHERE} clause in the @code{UPDATE} statement. If the record wasn't updated, we give the client a message: "Some of the data you have changed have been changed by another user". Then we show the old row versus the new row in a window, so the user can decide which version of the customer record he should use. This gives us something that is similar to column locking but is actually even better, because we only update some of the columns, using values that are relative to their current values. This means that typical @code{UPDATE} statements look something like these: @example UPDATE tablename SET pay_back=pay_back+'relative change'; UPDATE customer SET customer_date='current_date', address='new address', phone='new phone', money_he_owes_us=money_he_owes_us+'new_money' WHERE customer_id=id AND address='old address' AND phone='old phone'; @end example As you can see, this is very efficient and works even if another client has changed the values in the @code{pay_back} or @code{money_he_owes_us} columns. @findex mysql_insert_id() @findex LAST_INSERT_ID() In many cases, users have wanted @code{ROLLBACK} and/or @code{LOCK TABLES} for the purpose of managing unique identifiers for some tables. This can be handled much more efficiently by using an @code{AUTO_INCREMENT} column and either the SQL function @code{LAST_INSERT_ID()} or the C API function @code{mysql_insert_id()}. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @cindex rows, locking At MySQL AB, we have never had any need for row-level locking because we have always been able to code around it. Some cases really need row locking, but they are very few. If you want row-level locking, you can use a flag column in the table and do something like this: @example UPDATE tbl_name SET row_flag=1 WHERE id=ID; @end example MySQL returns 1 for the number of affected rows if the row was found and @code{row_flag} wasn't already 1 in the original row. You can think of it as MySQL changed the above query to: @example UPDATE tbl_name SET row_flag=1 WHERE id=ID and row_flag <> 1; @end example @node Bugs, , Commit-rollback, Compatibility @subsection Known errors and design deficiencies in MySQL @cindex bugs, known @cindex errors, known @cindex design, issues @cindex known errors The following problems are known and have a very high priority to get fixed: @itemize @bullet @item @code{ANALYZE TABLE} on a BDB table may in some case make the table unusable until one has restarted @code{mysqld}. When this happens you will see errors like the following in the MySQL error file: @example 001207 22:07:56 bdb: log_flush: LSN past current end-of-log @end example @item Don't execute @code{ALTER TABLE} on a @code{BDB} table on which you are running not completed multi-statement transactions. (The transaction will probably be ignored). @item @code{ANALYZE TABLE}, @code{OPTIMIZE TABLE} and @code{REPAIR TABLE} may cause problems on tables for which you are using @code{INSERT DELAYED}. @item Doing a @code{LOCK TABLE ..} and @code{FLUSH TABLES ..} doesn't guarantee that there isn't a half-finished transaction in progress on the table. @item BDB tables are a bit slow to open. If you have many BDB tables in a database, it will take a long time to use the @code{mysql} client on the database if you are not using the @code{-A} option or if you are using @code{rehash}. This is especially notable when you have a big table cache. @item Th current replication protocol cannot deal with @code{LOAD DATA INFILE} and line terminator characters of more than 1 character. @end itemize The following problems are known and will be fixed in due time: @itemize @bullet @item For the moment @code{MATCH} only works with @code{SELECT} statements. @item When using @code{SET CHARACTER SET}, one can't use translated characters in database, table and column names. @item @code{DELETE FROM merge_table} used without a @code{WHERE} will only clear the mapping for the table, not delete everything in the mapped tables @item You cannot build in another directory when using MIT-pthreads. Because this requires changes to MIT-pthreads, we are not likely to fix this. @item @code{BLOB} values can't ``reliably'' be used in @code{GROUP BY} or @code{ORDER BY} or @code{DISTINCT}. Only the first @code{max_sort_length} bytes (default 1024) are used when comparing @code{BLOB}bs in these cases. This can be changed with the @code{-O max_sort_length} option to @code{mysqld}. A workaround for most cases is to use a substring: @code{SELECT DISTINCT LEFT(blob,2048) FROM tbl_name}. @item Calculation is done with @code{BIGINT} or @code{DOUBLE} (both are normally 64 bits long). It depends on the function which precision one gets. The general rule is that bit functions are done with @code{BIGINT} precision, @code{IF}, and @code{ELT()} with @code{BIGINT} or @code{DOUBLE} precision and the rest with @code{DOUBLE} precision. One should try to avoid using bigger unsigned long long values than 63 bits (9223372036854775807) for anything else than bit fields! @item All string columns, except @code{BLOB} and @code{TEXT} columns, automatically have all trailing spaces removed when retrieved. For @code{CHAR} types this is okay, and may be regarded as a feature according to ANSI SQL92. The bug is that in MySQL, @code{VARCHAR} columns are treated the same way. @item You can only have up to 255 @code{ENUM} and @code{SET} columns in one table. @item @code{safe_mysqld} re-directs all messages from @code{mysqld} to the @code{mysqld} log. One problem with this is that if you execute @code{mysqladmin refresh} to close and reopen the log, @code{stdout} and @code{stderr} are still redirected to the old log. If you use @code{--log} extensively, you should edit @code{safe_mysqld} to log to @file{'hostname'.err} instead of @file{'hostname'.log} so you can easily reclaim the space for the old log by deleting the old one and executing @code{mysqladmin refresh}. @item In the @code{UPDATE} statement, columns are updated from left to right. If you refer to an updated column, you will get the updated value instead of the original value. For example: @example mysql> UPDATE tbl_name SET KEY=KEY+1,KEY=KEY+1; @end example This will update @code{KEY} with @code{2} instead of with @code{1}. @item You can't use temporary tables more than once in the same query. For example, the following doesn't work. @example select * from temporary_table, temporary_table as t2; @end example @item @code{RENAME} doesn't work with @code{TEMPORARY} tables. @item The optimizer may handle @code{DISTINCT} differently if you are using 'hidden' columns in a join or not. In a join, hidden columns are counted as part of the result (even if they are not shown) while in normal queries hidden columns doesn't participate in the @code{DISTINCT} comparison. We will probably change this in the future to never compare the hidden columns when executing @code{DISTINCT} An example of this is: @example SELECT DISTINCT mp3id FROM band_downloads WHERE userid = 9 ORDER BY id DESC; @end example and @example SELECT DISTINCT band_downloads.mp3id, FROM band_downloads,band_mp3 WHERE band_downloads.userid = 9 AND band_mp3.id = band_downloads.mp3id ORDER BY band_downloads.id DESC; @end example In the second case you may in MySQL 3.23.x get two identical rows in the result set (because the hidden 'id' column may differ). Note that the this only happens for queries where you don't have the ORDER BY columns in the result, something that is you are not allowed to do in ANSI SQL. @item Because MySQL allows you to work with table types that doesn't support transactions (and thus can't @code{rollback} data) some things behaves a little different in MySQL than in other SQL servers: (This is just to ensure that MySQL never need to do a rollback for a SQL command). This may be a little awkward at times as column Because MySQL allows you to work with table types that don't support transactions, and thus can't @code{rollback} data, some things behave a little differently in MySQL than in other SQL servers. This is just to ensure that MySQL never need to do a rollback for a SQL command. This may be a little awkward at times as column values must be checked in the application, but this will actually give you a nice speed increase as it allows MySQL to do some optimizations that otherwise would be very hard to do. If you set a column to an incorrect value, MySQL will, instead of doing a rollback, store the @code{best possible value} in the column: @itemize @minus @item If you try to store a value outside the range in a numerical column, MySQL will instead store the smallest or biggest possible value in the column. @item If you try to store a string that doesn't start with a number into a numerical column, MySQL will store 0 into it. @item If you try to store @code{NULL} into a column that doesn't take @code{NULL} values, MySQL will store 0 or @code{''} (empty string) in it instead. (This behavior can, however, be changed with the -DDONT_USE_DEFAULT_FIELDS compile option). @item MySQL allows you to store some wrong date values into @code{DATE} and @code{DATETIME} columns. (Like 2000-02-31 or 2000-02-00). If the date is totally wrong, MySQL will store the special 0000-00-00 date value in the column. @item If you set an @code{enum} to an unsupported value, it will be set to the error value 'empty string', with numeric value 0. @end itemize @item If you execute a @code{PROCEDURE} on a query that returns an empty set, in some cases the @code{PROCEDURE} will not transform the columns. @item Creation of a table of type @code{MERGE} doesn't check if the underlying tables are of compatible types. @item MySQL can't yet handle @code{NaN}, @code{-Inf} and @code{Inf} values in double. Using these will cause problems when trying to export and import data. We should as an intermediate solution change @code{NaN} to @code{NULL} (if possible) and @code{-Inf} and @code{Inf} to the Minimum respective maximum possible @code{double} value. @item @code{LIMIT} on negative numbers are treated as big positive numbers. @item If you use @code{ALTER TABLE} to first add an @code{UNIQUE} index to a table used in a @code{MERGE} table and then use @code{ALTER TABLE} to add a normal index on the @code{MERGE} table, the key order will be different for the tables if there was an old not-unique key in the table. This is because @code{ALTER TABLE} puts @code{UNIQUE} keys before normal keys to be able to detect duplicate keys as early as possible. @end itemize The following are known bugs in earlier versions of MySQL: @itemize @bullet @item You can get a hung thread if you do a @code{DROP TABLE} on a table that is one among many tables that is locked with @code{LOCK TABLES}. @item In the following case you can get a core dump: @itemize @minus @item Delayed insert handler has pending inserts to a table. @item @code{LOCK table} with @code{WRITE} @item @code{FLUSH TABLES} @end itemize @item Before MySQL Version 3.23.2 an @code{UPDATE} that updated a key with a @code{WHERE} on the same key may have failed because the key was used to search for records and the same row may have been found multiple times: @example UPDATE tbl_name SET KEY=KEY+1 WHERE KEY > 100; @end example A workaround is to use: @example mysql> UPDATE tbl_name SET KEY=KEY+1 WHERE KEY+0 > 100; @end example This will work because MySQL will not use index on expressions in the @code{WHERE} clause. @item Before MySQL Version 3.23, all numeric types where treated as fixed-point fields. That means you had to specify how many decimals a floating-point field shall have. All results were returned with the correct number of decimals. @end itemize For platform-specific bugs, see the sections about compiling and porting. @node Comparisons, TODO, Compatibility, Introduction @section How MySQL Compares to Other Databases @cindex databases, MySQL vs. others @cindex comparisons, MySQL vs. others @menu * Compare mSQL:: How MySQL compares to @code{mSQL} * Compare PostgreSQL:: How MySQL compares with PostgreSQL @end menu This section compares MySQL to other popular databases. This section has been written by the MySQL developers, so it should be read with that in mind. There are no factual errors contained in this section that we know of. If you find something which you believe to be an error, please contact us about it at @email{docs@@mysql.com}. For a list of all supported limits, functions, and types, see the @code{crash-me} Web page at @uref{http://www.mysql.com/information/crash-me.php}. @node Compare mSQL, Compare PostgreSQL, Comparisons, Comparisons @subsection How MySQL Compares to @code{mSQL} @table @strong @item Performance For a true comparison of speed, consult the growing MySQL benchmark suite. @xref{MySQL Benchmarks}. Because there is no thread creation overhead, a small parser, few features, and simple security, @code{mSQL} should be quicker at: @itemize @bullet @item Tests that perform repeated connects and disconnects, running a very simple query during each connection. @item @code{INSERT} operations into very simple tables with few columns and keys. @item @code{CREATE TABLE} and @code{DROP TABLE}. @item @code{SELECT} on something that isn't an index. (A table scan is very easy.) @end itemize Because these operations are so simple, it is hard to be better at them when you have a higher startup overhead. After the connection is established, MySQL should perform much better. On the other hand, MySQL is much faster than @code{mSQL} (and most other SQL implementations) on the following: @itemize @bullet @item Complex @code{SELECT} operations. @item Retrieving large results (MySQL has a better, faster, and safer protocol). @item Tables with variable-length strings, because MySQL has more efficient handling and can have indexes on @code{VARCHAR} columns. @item Handling tables with many columns. @item Handling tables with large record lengths. @item @code{SELECT} with many expressions. @item @code{SELECT} on large tables. @item Handling many connections at the same time. MySQL is fully multi-threaded. Each connection has its own thread, which means that no thread has to wait for another (unless a thread is modifying a table another thread wants to access). In @code{mSQL}, once one connection is established, all others must wait until the first has finished, regardless of whether the connection is running a query that is short or long. When the first connection terminates, the next can be served, while all the others wait again, etc. @item Joins. @code{mSQL} can become pathologically slow if you change the order of tables in a @code{SELECT}. In the benchmark suite, a time more than 15000 times slower than MySQL was seen. This is due to @code{mSQL}'s lack of a join optimizer to order tables in the optimal order. However, if you put the tables in exactly the right order in @code{mSQL}2 and the @code{WHERE} is simple and uses index columns, the join will be relatively fast! @xref{MySQL Benchmarks}. @item @code{ORDER BY} and @code{GROUP BY}. @item @code{DISTINCT}. @item Using @code{TEXT} or @code{BLOB} columns. @end itemize @item SQL Features @itemize @bullet @item @code{GROUP BY} and @code{HAVING}. @code{mSQL} does not support @code{GROUP BY} at all. MySQL supports a full @code{GROUP BY} with both @code{HAVING} and the following functions: @code{COUNT()}, @code{AVG()}, @code{MIN()}, @code{MAX()}, @code{SUM()}, and @code{STD()}. @code{COUNT(*)} is optimized to return very quickly if the @code{SELECT} retrieves from one table, no other columns are retrieved, and there is no @code{WHERE} clause. @code{MIN()} and @code{MAX()} may take string arguments. @item @code{INSERT} and @code{UPDATE} with calculations. MySQL can do calculations in an @code{INSERT} or @code{UPDATE}. For example: @example mysql> UPDATE SET x=x*10+y WHERE x<20; @end example @item Aliasing. MySQL has column aliasing. @item Qualifying column names. In MySQL, if a column name is unique among the tables used in a query, you do not have to use the full qualifier. @item @code{SELECT} with functions. MySQL has many functions (too many to list here; see @ref{Functions}). @end itemize @item Disk Space Efficiency That is, how small can you make your tables? MySQL has very precise types, so you can create tables that take very little space. An example of a useful MySQL datatype is the @code{MEDIUMINT} that is 3 bytes long. If you have 100,000,000 records, saving even one byte per record is very important. @code{mSQL2} has a more limited set of column types, so it is more difficult to get small tables. @item Stability This is harder to judge objectively. For a discussion of MySQL stability, see @ref{Stability}. We have no experience with @code{mSQL} stability, so we cannot say anything about that. @item Price Another important issue is the license. MySQL has a more flexible license than @code{mSQL}, and is also less expensive than @code{mSQL}. Whichever product you choose to use, remember to at least consider paying for a license or e-mail support. (You are required to get a license if you include MySQL with a product that you sell, of course.) @item Perl Interfaces MySQL has basically the same interfaces to Perl as @code{mSQL} with some added features. @item JDBC (Java) MySQL currently has a lot of different JDBC drivers: @itemize @bullet @item The mm driver: A type 4 JDBC driver by Mark Matthews @email{mmatthew@@ecn.purdue.edu}. This is released under the LGPL. @item The Resin driver. This is a commercial JDBC driver released under open source. @uref{http://www.caucho.com/projects/jdbc-mysql/index.xtp} @item The gwe driver: A Java interface by GWE technologies (not supported anymore). @item The jms driver: An improved gwe driver by Xiaokun Kelvin ZHU @email{X.Zhu@@brad.ac.uk} (not supported anymore). @item The twz driver: A type 4 JDBC driver by Terrence W. Zellers @email{zellert@@voicenet.com}. This is commercial but is free for private and educational use (not supported anymore). @end itemize The recommended driver is the mm driver. The Resin driver may also be good (at least the benchmarks looks good), but we haven't received that much information about this yet. We know that @code{mSQL} has a JDBC driver, but we have too little experience with it to compare. @item Rate of Development MySQL has a very small team of developers, but we are quite used to coding C and C++ very rapidly. Because threads, functions, @code{GROUP BY}, and so on are still not implemented in @code{mSQL}, it has a lot of catching up to do. To get some perspective on this, you can view the @code{mSQL} @file{HISTORY} file for the last year and compare it with the News section of the MySQL Reference Manual (@pxref{News}). It should be pretty obvious which one has developed most rapidly. @item Utility Programs Both @code{mSQL} and MySQL have many interesting third-party tools. Because it is very easy to port upward (from @code{mSQL} to MySQL), almost all the interesting applications that are available for @code{mSQL} are also available for MySQL. MySQL comes with a simple @code{msql2mysql} program that fixes differences in spelling between @code{mSQL} and MySQL for the most-used C API functions. For example, it changes instances of @code{msqlConnect()} to @code{mysql_connect()}. Converting a client program from @code{mSQL} to MySQL usually takes a couple of minutes. @end table @menu * Using mSQL tools:: How to convert @code{mSQL} tools for MySQL * Protocol differences:: How @code{mSQL} and MySQL Client/Server Communications Protocols Differ * Syntax differences:: How @code{mSQL} 2.0 SQL Syntax Differs from MySQL @end menu @node Using mSQL tools, Protocol differences, Compare mSQL, Compare mSQL @subsubsection How to Convert @code{mSQL} Tools for MySQL @cindex MySQL tools, conversion @cindex converting, tools @cindex tools, converting According to our experience, it would just take a few hours to convert tools such as @code{msql-tcl} and @code{msqljava} that use the @code{mSQL} C API so that they work with the MySQL C API. The conversion procedure is: @enumerate @item Run the shell script @code{msql2mysql} on the source. This requires the @code{replace} program, which is distributed with MySQL. @item Compile. @item Fix all compiler errors. @end enumerate Differences between the @code{mSQL} C API and the MySQL C API are: @itemize @bullet @item MySQL uses a @code{MYSQL} structure as a connection type (@code{mSQL} uses an @code{int}). @item @code{mysql_connect()} takes a pointer to a @code{MYSQL} structure as a parameter. It is easy to define one globally or to use @code{malloc()} to get one. @code{mysql_connect()} also takes two parameters for specifying the user and password. You may set these to @code{NULL, NULL} for default use. @item @code{mysql_error()} takes the @code{MYSQL} structure as a parameter. Just add the parameter to your old @code{msql_error()} code if you are porting old code. @item MySQL returns an error number and a text error message for all errors. @code{mSQL} returns only a text error message. @item Some incompatibilities exist as a result of MySQL supporting multiple connections to the server from the same process. @end itemize @node Protocol differences, Syntax differences, Using mSQL tools, Compare mSQL @subsubsection How @code{mSQL} and MySQL Client/Server Communications Protocols Differ @cindex communications protocols @cindex mSQL vs. MySQL There are enough differences that it is impossible (or at least not easy) to support both. The most significant ways in which the MySQL protocol differs from the @code{mSQL} protocol are listed below: @itemize @bullet @item A message buffer may contain many result rows. @item The message buffers are dynamically enlarged if the query or the result is bigger than the current buffer, up to a configurable server and client limit. @item All packets are numbered to catch duplicated or missing packets. @item All column values are sent in ASCII. The lengths of columns and rows are sent in packed binary coding (1, 2, or 3 bytes). @item MySQL can read in the result unbuffered (without having to store the full set in the client). @item If a single read/write takes more than 30 seconds, the server closes the connection. @item If a connection is idle for 8 hours, the server closes the connection. @end itemize @menu * Syntax differences:: How @code{mSQL} 2.0 SQL Syntax Differs from MySQL @end menu @node Syntax differences, , Protocol differences, Compare mSQL @subsubsection How @code{mSQL} 2.0 SQL Syntax Differs from MySQL @noindent @strong{Column types} @table @code @item MySQL Has the following additional types (among others; @pxref{CREATE TABLE, , @code{CREATE TABLE}}): @itemize @bullet @item @code{ENUM} type for one of a set of strings. @item @code{SET} type for many of a set of strings. @item @code{BIGINT} type for 64-bit integers. @end itemize @item MySQL also supports the following additional type attributes: @itemize @bullet @item @code{UNSIGNED} option for integer columns. @item @code{ZEROFILL} option for integer columns. @item @code{AUTO_INCREMENT} option for integer columns that are a @code{PRIMARY KEY}. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @item @code{DEFAULT} value for all columns. @end itemize @item mSQL2 @code{mSQL} column types correspond to the MySQL types shown below: @multitable @columnfractions .15 .85 @item @code{mSQL} @strong{type} @tab @strong{Corresponding MySQL type} @item @code{CHAR(len)} @tab @code{CHAR(len)} @item @code{TEXT(len)} @tab @code{TEXT(len)}. @code{len} is the maximal length. And @code{LIKE} works. @item @code{INT} @tab @code{INT}. With many more options! @item @code{REAL} @tab @code{REAL}. Or @code{FLOAT}. Both 4- and 8-byte versions are available. @item @code{UINT} @tab @code{INT UNSIGNED} @item @code{DATE} @tab @code{DATE}. Uses ANSI SQL format rather than @code{mSQL}'s own format. @item @code{TIME} @tab @code{TIME} @item @code{MONEY} @tab @code{DECIMAL(12,2)}. A fixed-point value with two decimals. @end multitable @end table @noindent @strong{Index Creation} @table @code @item MySQL Indexes may be specified at table creation time with the @code{CREATE TABLE} statement. @item mSQL Indexes must be created after the table has been created, with separate @code{CREATE INDEX} statements. @end table @noindent @strong{To Insert a Unique Identifier into a Table} @table @code @item MySQL Use @code{AUTO_INCREMENT} as a column type specifier. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @item mSQL Create a @code{SEQUENCE} on a table and select the @code{_seq} column. @end table @noindent @strong{To Obtain a Unique Identifier for a Row} @table @code @item MySQL Add a @code{PRIMARY KEY} or @code{UNIQUE} key to the table and use this. New in Version 3.23.11: If the @code{PRIMARY} or @code{UNIQUE} key consists of only one column and this is of type integer, one can also refer to it as @code{_rowid}. @item mSQL Use the @code{_rowid} column. Observe that @code{_rowid} may change over time depending on many factors. @end table @noindent @strong{To Get the Time a Column Was Last Modified} @table @code @item MySQL Add a @code{TIMESTAMP} column to the table. This column is automatically set to the current date and time for @code{INSERT} or @code{UPDATE} statements if you don't give the column a value or if you give it a @code{NULL} value. @item mSQL Use the @code{_timestamp} column. @end table @noindent @strong{@code{NULL} Value Comparisons} @table @code @item MySQL MySQL follows ANSI SQL, and a comparison with @code{NULL} is always @code{NULL}. @item mSQL In @code{mSQL}, @code{NULL = NULL} is TRUE. You must change @code{=NULL} to @code{IS NULL} and @code{<>NULL} to @code{IS NOT NULL} when porting old code from @code{mSQL} to MySQL. @end table @noindent @strong{String Comparisons} @table @code @item MySQL Normally, string comparisons are performed in case-independent fashion with the sort order determined by the current character set (ISO-8859-1 Latin1 by default). If you don't like this, declare your columns with the @code{BINARY} attribute, which causes comparisons to be done according to the ASCII order used on the MySQL server host. @item mSQL All string comparisons are performed in case-sensitive fashion with sorting in ASCII order. @end table @noindent @strong{Case-insensitive Searching} @table @code @item MySQL @code{LIKE} is a case-insensitive or case-sensitive operator, depending on the columns involved. If possible, MySQL uses indexes if the @code{LIKE} argument doesn't start with a wild-card character. @item mSQL Use @code{CLIKE}. @end table @noindent @strong{Handling of Trailing Spaces} @table @code @item MySQL Strips all spaces at the end of @code{CHAR} and @code{VARCHAR} columns. Use a @code{TEXT} column if this behavior is not desired. @item mSQL Retains trailing space. @end table @noindent @strong{@code{WHERE} Clauses} @table @code @item MySQL MySQL correctly prioritizes everything (@code{AND} is evaluated before @code{OR}). To get @code{mSQL} behavior in MySQL, use parentheses (as shown in an example below). @item mSQL Evaluates everything from left to right. This means that some logical calculations with more than three arguments cannot be expressed in any way. It also means you must change some queries when you upgrade to MySQL. You do this easily by adding parentheses. Suppose you have the following @code{mSQL} query: @example mysql> SELECT * FROM table WHERE a=1 AND b=2 OR a=3 AND b=4; @end example To make MySQL evaluate this the way that @code{mSQL} would, you must add parentheses: @example mysql> SELECT * FROM table WHERE (a=1 AND (b=2 OR (a=3 AND (b=4)))); @end example @end table @noindent @strong{Access Control} @table @code @item MySQL Has tables to store grant (permission) options per user, host, and database. @xref{Privileges}. @item mSQL Has a file @file{mSQL.acl} in which you can grant read/write privileges for users. @item @end table @node Compare PostgreSQL, , Compare mSQL, Comparisons @subsection How MySQL Compares to PostgreSQL @cindex PostgreSQL vs. MySQL, overview When reading the following, please note that both products are continually evolving. We at MySQL AB and the PostgreSQL developers are both working on making our respective database as good as possible, so we are both a serious choice to any commercial database. The following comparison is made by us at MySQL AB. We have tried to be as accurate and fair as possible, but because we don't have a full knowledge of all PostgreSQL features while we know MySQL througly, we may have got some things wrong. We will however correct these when they come to our attention. We would first like to note that PostgreSQL and MySQL are both widely used products, but with different design goals, even if we are both striving to be ANSI SQL compatible. This means that for some applications MySQL is more suited, while for others PostgreSQL is more suited. When choosing which database to use, you should first check if the database's feature set satisfies your application. If you need raw speed, MySQL is probably your best choice. If you need some of the extra features that only PostgreSQL can offer, you should use @code{PostgreSQL}. @cindex PostgreSQL/MySQL, strategies @menu * MySQL-PostgreSQL goals:: MySQL and PostgreSQL development strategies * MySQL-PostgreSQL features:: Featurewise Comparison of MySQL and PostgreSQL * MySQL-PostgreSQL benchmarks:: Benchmarking MySQL and PostgreSQL @end menu @node MySQL-PostgreSQL goals, MySQL-PostgreSQL features, Compare PostgreSQL, Compare PostgreSQL @subsubsection MySQL and PostgreSQL development strategies When adding things to MySQL we take pride to do an optimal, definite solution. The code should be so good that we shouldn't have any need to change it in the foreseeable future. We also do not like to sacrifice speed for features but instead will do our utmost to find a solution that will give maximal throughput. This means that development will take a little longer, but the end result will be well worth this. This kind of development is only possible because all server code are checked by one of a few (currently two) persons before it's included in the MySQL server. We at MySQL AB believe in frequent releases to be able to push out new features quickly to our users. Because of this we do a new small release about every three weeks, and a major branch every year. All releases are throughly tested with our testing tools on a lot of different platforms. PostgreSQL is based on a kernel with lots of contributors. In this setup it makes sense to prioritize adding a lot of new features, instead of implementing them optimally, because one can always optimize things later if there arises a need for this. Another big difference between MySQL and PostgreSQL is that nearly all of the code in the MySQL server are coded by developers that are employed by MySQL AB and are still working on the server code. The exceptions are the transaction engines, and the regexp library. This is in sharp contrast to the PostgreSQL code where the majority of the code is coded by a big group of people with different backgrounds. It was only recently that the PostgreSQL developers announced that their current developer group had finally had time to take a look at all the code in the current PostgreSQL release. Both of the above development methods has it's own merits and drawbacks. We here at MySQL AB think of course that our model is better because our model gives better code consistency, more optimal and reusable code, and in our opinion, fewer bugs. Because we are the authors of the MySQL server code, we are better able to coordinate new features and releases. @node MySQL-PostgreSQL features, MySQL-PostgreSQL benchmarks, MySQL-PostgreSQL goals, Compare PostgreSQL @subsubsection Featurewise Comparison of MySQL and PostgreSQL @cindex PostgreSQL/MySQL, features On the @uref{http://www.mysql.com/information/crash-me.php, crash-me} page you can find a list of those database constructs and limits that one can detect automatically with a program. Note however that a lot of the numerical limits may be changed with startup options for respective database. The above web page is however extremely useful when you want to ensure that your applications works with many different databases or when you want to convert your application from one datbase to another. MySQL offers the following advantages over PostgreSQL: @itemize @bullet @item @code{MySQL} is generally much faster than PostgreSQL. @xref{MySQL-PostgreSQL benchmarks}. @item MySQL has a much larger user base than PostgreSQL, therefor the code is more tested and has historically been more stable than PostgreSQL. MySQL is the much more used in production environments than PostgreSQL, mostly thanks to that MySQL AB, formerly TCX DataKonsult AB, has provided top quality commercial support for MySQL from the day it was released, whereas until recently PostgreSQL was unsupported. @item MySQL works better on Windows than PostgreSQL does. MySQL runs as a native Windows application (a service on NT/Win2000/WinXP), while PostgreSQL is run under the cygwin emulation. We have heard that PostgreSQL is not yet that stable on Windows but we haven't been able to verify this ourselves. @item MySQL has more APIs to other languages and is supported by more existing programs than PostgreSQL. @xref{Contrib}. @item MySQL works on 24/7 heavy duty systems. In most circumstances you never have to run any cleanups on MySQL. PostgreSQL doesn't yet support 24/7 systems because you have to run @code{VACUUM()} once in a while to reclaim space from @code{UPDATE} and @code{DELETE} commands and to perform statistics analyzes that are critical to get good performance with PostgreSQL. @code{VACUUM()} is also needed after adding a lot of new rows to a table. On a busy system with lots of changes, @code{VACUUM()} must be run very frequently, in the worst cases even many times a day. During the @code{VACUUM()} run, which may take hours if the database is big, the database is from a production standpoint, practically dead. The PostgreSQL team has fixing this on their TODO, but we assume that this is not an easy thing to fix permanently. @item A working, tested replication feature used by sites like: @itemize @minus @item Yahoo Finance (@uref{http://finance.yahoo.com}) @item Mobile.de (@uref{http://www.mobile.de/}) @item Slashdot (@uref{http://www.slashdot.org}) @end itemize @item Included in the MySQL distribution are two different testing suites, @file{mysql-test-run} and @uref{http://www.mysql.com/information/crash-me.php,crash-me}, as well as a benchmark suite. The test system is actively updated with code to test each new feature and almost all repeatable bugs that have come to our attention. We test MySQL with these on a lot of platforms before every release. These tests are more sophisticated than anything we have seen from PostgreSQL, and they ensures that the MySQL is kept to a high standard. @item There are far more books in print about MySQL than about PostgreSQL. O'Reilly, Sams, Que, and New Riders are all major publishers with books about MySQL. All MySQL features are also documented in the MySQL on-line manual, because when a new feature is implemented, the MySQL developers are required to document it before it's included in the source. @item MySQL supports more of the standard ODBC functions than @code{PostgreSQL}. @item MySQL has a much more sophisticated @code{ALTER TABLE}. @item MySQL has support for tables without transactions for applications that need all speed they can get. The tables may be memory based, @code{HEAP} tables or disk based @code{MyISAM}. @xref{Table types}. @item MySQL has support for two different table handlers that support transactions, @code{BerkeleyDB} and @code{InnoDB}. Because every transaction engine performs differently under different conditions, this gives the application writer more options to find an optimal solution for his or her setup. @xref{Table types}. @item @code{MERGE} tables gives you a unique way to instantly make a view over a set of identical tables and use these as one. This is perfect for systems where you have log files that you order for example by month. @xref{MERGE}. @item The option to compress read-only tables, but still have direct access to the rows in the table, gives you better performance by minimizing disk reads. This is very useful when you are archiving things. @xref{myisampack}. @item MySQL has internal support for fulltext search. @xref{Fulltext Search}. @item You can access many databases from the same connection (depending of course on your privileges). @item MySQL is coded from the start to be multi-threaded while PostgreSQL uses processes. Context switching and access to common storage areas is much faster between threads than between separate processes, this gives MySQL a big speed advantage in multi-user applications and also makes it easier for MySQL to take full advantage of symmetric multiprocessor (SMP) systems. @item MySQL has a much more sophisticated privilege system than PostgreSQL. While PostgreSQL only supports @code{INSERT}, @code{SELECT}, and @code{UPDATE/DELETE} grants per user on a database or a table, MySQL allows you to define a full set of different privileges on database, table and column level. MySQL also allows you to specify the privilege on host and user combinations. @xref{GRANT}. @item MySQL supports a compressed client/server protocol which improves performance over slow links. @item MySQL employs a ``table handler'' concept, and is the only relational database we know of built around this concept. This allows different low-level table types to be swapped into the SQL engine, and each table type can be optimized for different performance characteristics. @item All MySQL table types (except @strong{InnoDB}) are implemented as files (one table per file), which makes it really easy to backup, move, delete and even symlink databases and tables, even when the server is down. @item Tools to repair and optimize @strong{MyISAM} tables (the most common MySQL table type). A repair tool is only needed when a physical corruption of a data file happens, usually from a hardware failure. It allows a majority of the data to be recovered. @item Upgrading MySQL is painless. When you are upgrading MySQL, you don't need to dump/restore your data, as you have to do with most PostgreSQL upgrades. @end itemize Drawbacks with MySQL compared to PostgreSQL: @itemize @bullet @item The transaction support in MySQL is not yet as well tested as PostgreSQL's system. @item Because MySQL uses threads, which are not yet flawless on many OSes, one must either use binaries from @uref{http://www.mysql.com/downloads}, or carefully follow our instructions on @uref{http://www.mysql.com/doc/I/n/Installing_source.html} to get an optimal binary that works in all cases. @item Table locking, as used by the non-transactional @code{MyISAM} tables, is in many cases faster than page locks, row locks or versioning. The drawback however is that if one doesn't take into account how table locks work, a single long-running query can block a table for updates for a long time. This can usable be avoided when designing the application. If not, one can always switch the trouble table to use one of the transactional table types. @xref{Table locking}. @item With UDF (user defined functions) one can extend MySQL with both normal SQL functions and aggregates, but this is not yet as easy or as flexible as in PostgreSQL. @xref{Adding functions}. @item Updates and deletes that run over multiple tables is harder to do in MySQL. This will, hoever, be fixed in MySQL 4.0 with multi-table @code{DELETE} and multi-table @code{UPDATE} and in MySQL 4.1 with subselects. @end itemize PostgreSQL currently offers the following advantages over MySQL: Note that because we know the MySQL road map, we have included in the following table the version when MySQL should support this feature. Unfortunately we couldn't do this for previous comparison, because we don't know the PostgreSQL roadmap. @multitable @columnfractions .70 .30 @item @strong{Feature} @tab @strong{MySQL version} @item Subselects @tab 4.1 @item Foreign keys @tab 4.0 and 4.1 @item Views @tab 4.2 @item Stored procedures @tab 4.1 @item Extensible type system @tab Not planned @item Unions @tab 4.0 @item Full join @tab 4.0 or 4.1 @item Triggers @tab 4.1 @item Constrainst @tab 4.1 @item Cursors @tab 4.1 or 4.2 @item Extensible index types like R-trees @tab R-trees are planned for 4.2 @item Inherited tables @tab Not planned @end multitable Other reasons to use PostgreSQL: @itemize @bullet @item Standard usage in PostgreSQL is closer to ANSI SQL in some cases. @item One can speed up PostgreSQL by coding things as stored procedures. @item PostgreSQL has a bigger team of developers that contribute to the server. @end itemize Drawbacks with PostgreSQL compared to MySQL: @itemize @bullet @item @code{VACUUM()} makes PostgreSQL hard to use in a 24/7 environment. @item Only transactional tables. @item Much slower @code{INSERT}, @code{DELETE}, and @code{UPDATE}. @end itemize For a complete list of drawbacks, you should also examine the first table in this section. @menu * MySQL-PostgreSQL benchmarks:: Benchmarking MySQL and PostgreSQL @end menu @node MySQL-PostgreSQL benchmarks, , MySQL-PostgreSQL features, Compare PostgreSQL @subsubsection Benchmarking MySQL and PostgreSQL @cindex PostgreSQL vs. MySQL, benchmarks The only open source benchmark that we know of that can be used to benchmark MySQL and PostgreSQL (and other databases) is our own. It can be found at @uref{http://www.mysql.com/information/benchmarks.html}. We have many times asked the PostgreSQL developers and some PostgreSQL users to help us extend this benchmark to make it the definitive benchmark for databases, but unfortunately we haven't gotten any feedback for this. We the MySQL developers have, because of this, spent a lot of hours to get maximum performance from PostgreSQL for the benchmarks, but because we don't know PostgreSQL intimately, we are sure that there are things that we have missed. We have on the benchmark page documented exactly how we did run the benchmark so that it should be easy for anyone to repeat and verify our results. The benchmarks are usually run with and without the @code{--fast} option. When run with @code{--fast} we are trying to use every trick the server can do to get the code to execute as fast as possible. The idea is that the normal run should show how the server would work in a default setup and the @code{--fast} run shows how the server would do if the application developer would use extensions in the server to make his application run faster. When running with PostgreSQL and @code{--fast} we do a @code{VACUUM()} after every major table @code{UPDATE} and @code{DROP TABLE} to make the database in perfect shape for the following @code{SELECT}s. The time for @code{VACUUM()} is measured separately. When running with PostgreSQL 7.1.1 we could, however, not run with @code{--fast} because during the @code{INSERT} test, the postmaster (the PostgreSQL deamon) died and the database was so corrupted that it was impossible to restart postmaster. After this happened twice, we decided to postpone the @code{--fast} test until next PostgreSQL release. The details about the machine we run the benchmark can be found on the benchmark page. Before going to the other benchmarks we know of, we would like to give some background on benchmarks: It's very easy to write a test that shows ANY database to be best database in the world, by just restricting the test to something the database is very good at and not test anything that the database is not good at. If one after this publishes the result with a single figure, things are even easier. This would be like us measuring the speed of MySQL compared to PostgreSQL by looking at the summary time of the MySQL benchmarks on our web page. Based on this MySQL would be more than 40 times faster than PostgreSQL, something that is of course not true. We could make things even worse by just taking the test where PostgreSQL performs worst and claim that MySQL is more than 2000 times faster than PostgreSQL. The case is that MySQL does a lot of optimizations that PostgreSQL doesn't do. This is of course also true the other way around. An SQL optimizer is a very complex thing, and a company could spend years on just making the optimizer faster and faster. When looking at the benchmark results you should look for things that you do in your application and just use these results to decide which database would be best suited for your application. The benchmark results also shows things a particular database is not good at and should give you a notion about things to avoid and what you may have to do in other ways. We know of two benchmark tests that claims that PostgreSQL performs better than MySQL. These both where multi-user tests, a test that we here at MySQL AB haven't had time to write and include in the benchmark suite, mainly because it's a big task to do this in a manner that is fair against all databases. One is the benchmark paid for by Great Bridge, which you can read about at: @uref{http://www.greatbridge.com/about/press.php?content_id=4}. This is the probably worst benchmark we have ever seen anyone conduct. This was not only tuned to only test what PostgreSQL is absolutely best at, it was also totally unfair against every other database involved in the test. @strong{NOTE}: We know that not even some of the main PostgreSQL developers did like the way Great Bridge conducted the benchmark, so we don't blame them for the way the benchmark was made. This benchmark has been condemned in a lot of postings and newsgroups so we will here just shortly repeat some things that where wrong with it. @itemize @bullet @item The tests were run with an expensive commercial tool, that makes it impossible for an open source company like us to verify the benchmarks, or even check how the benchmarks were really done. The tool is not even a true benchmark tool, but an application/setup testing tool. To refer this as a ``standard'' benchmark tool is to stretch the truth a long way. @item Great Bridge admitted that they had optimized the PostgreSQL database (with @code{VACUUM()} before the test) and tuned the startup for the tests, something they hadn't done for any of the other databases involved. To say ``This process optimizes indexes and frees up disk space a bit. The optimized indexes boost performance by some margin.'' Our benchmarks clearly indicate that the difference in running a lot of selects on a database with and without @code{VACUUM()} can easily differ by a factor of ten. @item The test results were also strange. The AS3AP test documentation mentions that the test does ``selections, simple joins, projections, aggregates, one-tuple updates, and bulk updates''. PostgreSQL is good at doing @code{SELECT}s and @code{JOIN}s (especially after a @code{VACUUM()}), but doesn't perform as well on @code{INSERT}s or @code{UPDATE}s. The benchmarks seem to indicate that only @code{SELECT}s were done (or very few updates). This could easily explain they good results for PostgreSQL in this test. The bad results for MySQL will be obvious a bit down in this document. @item They did run the so-called benchmark from a Windows machine against a Linux machine over ODBC, a setup that no normal database user would ever do when running a heavy multi-user application. This tested more the ODBC driver and the Windows protocol used between the clients than the database itself. @item When running the database against Oracle and MS-SQL (Great Bridge has indirectly indicated that the databases they used in the test), they didn't use the native protocol but instead ODBC. Anyone that has ever used Oracle knows that all real application uses the native interface instead of ODBC. Doing a test through ODBC and claiming that the results had anything to do with using the database in a real-world situation can't be regarded as fair. They should have done two tests with and without ODBC to provide the right facts (after having got experts to tune all involved databases of course). @item They refer to the TPC-C tests, but they don't mention anywhere that the test they did was not a true TPC-C test and they were not even allowed to call it a TPC-C test. A TPC-C test can only be conducted by the rules approved by the TPC Council (@uref{http://www.tpc.org}). Great Bridge didn't do that. By doing this they have both violated the TPC trademark and miscredited their own benchmarks. The rules set by the TPC Council are very strict to ensure that no one can produce false results or make unprovable statements. Apparently Great Bridge wasn't interested in doing this. @item After the first test, we contacted Great Bridge and mentioned to them some of the obvious mistakes they had done with MySQL: @itemize @minus @item Running with a debug version of our ODBC driver @item Running on a Linux system that wasn't optimized for threads @item Using an old MySQL version when there was a recommended newer one available @item Not starting MySQL with the right options for heavy multi-user use (the default installation of MySQL is tuned for minimal resource use). @end itemize Great Bridge did run a new test, with our optimized ODBC driver and with better startup options for MySQL, but refused to either use our updated glibc library or our standard binary (used by 80% of our users), which was statically linked with a fixed glibc library. According to what we know, Great Bridge did nothing to ensure that the other databases were set up correctly to run well in their test environment. We are sure however that they didn't contact Oracle or Microsoft to ask for their advice in this matter ;) @item The benchmark was paid for by Great Bridge, and they decided to publish only partial, chosen results (instead of publishing it all). @end itemize Tim Perdue, a long time PostgreSQL fan and a reluctant MySQL user published a comparison on @uref{http://www.phpbuilder.com/columns/tim20001112.php3,phpbuider}. When we became aware of the comparison, we phoned Tim Perdue about this because there were a lot of strange things in his results. For example, he claimed that MySQL had a problem with five users in his tests, when we know that there are users with similar machines as his that are using MySQL with 2000 simultaneous connections doing 400 queries per second. (In this case the limit was the web bandwidth, not the database.) It sounded like he was using a Linux kernel that either had some problems with many threads, such as kernels before 2.4, which had a problem with many threads on multi-CPU machines. We have documented in this manual how to fix this and Tim should be aware of this problem. The other possible problem could have been an old glibc library and that Tim didn't use a MySQL binary from our site, which is linked with a corrected glibc library, but had compiled a version of his own with. In any of the above cases, the symptom would have been exactly what Tim had measured. We asked Tim if we could get access to his data so that we could repeat the benchmark and if he could check the MySQL version on the machine to find out what was wrong and he promised to come back to us about this. He has not done that yet. Because of this we can't put any trust in this benchmark either :( Over time things also changes and the above benchmarks are not that relevant anymore. MySQL now have a couple of different table handlers with different speed/concurrency tradeoffs. @xref{Table types}. It would be interesting to see how the above tests would run with the different transactional table types in MySQL. PostgreSQL has of course also got new features since the test was made. As the above test are not publicly available there is no way for us to know how the database would preform in the same tests today. Conclusion: The only benchmarks that exist today that anyone can download and run against MySQL and PostgreSQL is the MySQL benchmarks. We here at MySQL believe that open source databases should be tested with open source tools! This is the only way to ensure that no one does tests that nobody can reproduce and use this to claim that a database is better than another. Without knowing all the facts it's impossible to answer the claims of the tester. The thing we find strange is that every test we have seen about PostgreSQL, that is impossible to reproduce, claims that PostgreSQL is better in most cases while our tests, which anyone can reproduce, clearly shows otherwise. With this we don't want to say that PostgreSQL isn't good at many things (it is!) or that it isn't faster than MySQL under certain conditions. We would just like to see a fair test where they are very good so that we could get some friendly competition going! For more information about our benchmarks suite @xref{MySQL Benchmarks}. We are working on an even better benchmark suite, including multi user tests, and a better documentation of what the individual tests really do and how to add more tests to the suite. @node TODO, , Comparisons, Introduction @section MySQL and the future (The TODO) @cindex ToDo list for MySQL @menu * TODO MySQL 4.0:: Things that should be in Version 4.0 * TODO future:: Things that must be done in the near future * TODO sometime:: Things that have to be done sometime * TODO unplanned:: Some things we don't have any plans to do @end menu This appendix lists the features that we plan to implement in MySQL. Everything in this list is approximately in the order it will be done. If you want to affect the priority order, please register a license or support us and tell us what you want to have done more quickly. @xref{Licensing and Support}. The plan is that we in the future will support the full ANSI SQL99 standard, but with a lot of useful extensions. The challenge is to do this without sacrifying the speed or compromise the code. @node TODO MySQL 4.0, TODO future, TODO, TODO @subsection Things that should be in 4.0 We plan to make MySQL Version 4.0 a ``quick'' release where we only add some new stuff to enable others to help us with developing new features into Version 4.1. The MySQL 4.0 version should only take us about a month to make after which we want to stabilize it and start working on Version 4.1. Version 4.0 should have the following new features: The news section for 4.0 includes a list of the features we have already implemented in the 4.0 tree. @xref{News-4.0.x}. @itemize @bullet @item New table definition file format (@code{.frm} files) This will enable us to not run out of bits when adding more table options. One will still be able to use the old @code{.frm} file format with 4.0. All newly created tables will, however, use the new format. The new file format will enable us to add new column types, more options for keys and @code{FOREIGN KEY} support. @item @code{mysqld} as a library. This will have the same interface as the standard MySQL client (with an extra function to just set up startup parameters) but will be faster (no TCP/IP or socket overhead), smaller and much easier to use for embedded products. One will be able to define at link time if one wants to use the client/server model or a stand-alone application just by defining which library to link with. The @code{mysqld} will support all standard MySQL features and one can use it in a threaded client to run different queries in each thread. @item Replication should work with @code{RAND()} and user variables @code{@@var}. @item Online backup with very low performance penalty. The online backup will make it easy to add a new replication slave without taking down the master. @item @code{DELETE FROM table_name} will return the number of deleted rows. For fast execution one should use @code{TRUNCATE table_name}. @item Allow @code{DELETE} on @code{MyISAM} tables to use the record cache. To do this, we need to update the threads record cache when we update the @code{.MYD} file. @item Better replication. @item More functions for full-text search. @xref{Fulltext Features to Appear in MySQL 4.0}. @item Character set casts and syntax for handling multiple character sets. @item Allow users to change startup options without taking down the server. @item Help for all commands from the client. @item Secure connections (with SSL). @item Extend the optimizer to be able to optimize some @code{ORDER BY key_name DESC} queries. @item @code{SHOW COLUMNS FROM table_name} (used by @code{mysql} client to allow expansions of column names) should not open the table, but only the definition file. This will require less memory and be much faster. @item New key cache @item When using @code{SET CHARACTER SET} we should translate the whole query at once and not only strings. This will enable users to use the translated characters in database, table and column names. @item Add a portable interface over @code{gethostbyaddr_r()} so that we can change @code{ip_to_hostname()} to not block other threads while doing DNS lookups. @item Add @code{record_in_range()} method to @code{MERGE} tables to be able to choose the right index when there is many to choose from. We should also extend the info interface to get the key distribution for each index, of @code{analyze} is run on all sub tables. @item @code{SET SQL_DEFAULT_TABLE_TYPE=[MyISAM | INNODB | BDB | HEAP]}. @end itemize @node TODO future, TODO sometime, TODO MySQL 4.0, TODO @subsection Things that must be done in the real near future @itemize @bullet @item Fail safe replication. @item Subqueries. @code{select id from t where grp in (select grp from g where u > 100)} @item Derived tables. @example select a.col1, b.col2 from (select max(col1) as col1 from root_table ) a, other_table b where a.col1=b.col1 @end example This could be done by automatically creating temporary tables for the derived tables for the duration of the query. @item Add @code{PREPARE} of statements and sending of parameters to @code{mysqld}. @item Extend the server/client protocol to support warnings. @item Add options to the server/protocol protocol to get progress notes for long running commands. @item Add database and real table name (in case of alias) to the MYSQL_FIELD structure. @item Don't allow more than a defined number of threads to run MyISAM recover at the same time. @item Change @code{INSERT ... SELECT} to optionally use concurrent inserts. @item Implement @code{RENAME DATABASE}. To make this safe for all table handlers, it should work as follows: @itemize @bullet @item Create the new database. @item For every table do a rename of the table to another database, as we do with the @code{RENAME} command. @item Drop the old database. @end itemize @item Return the original field types() when doing @code{SELECT MIN(column) ... GROUP BY}. @item Multiple result sets. @item Change the protocol to allow binary transfer of values. To do this efficiently, we need to add an API to allow binding of variables. @item Make it possible to specify @code{long_query_time} with a granularity in microseconds. @item Add a configurable prompt to the @code{mysql} command line client, with options like database in use, time and date... @item Add range checking to @code{MERGE} tables. @item Link the @code{myisampack} code into the server. @item Port of MySQL to BeOS. @item Port of the MySQL clients to LynxOS. @item Add a temporary key buffer cache during @code{INSERT/DELETE/UPDATE} so that we can gracefully recover if the index file gets full. @item If you perform an @code{ALTER TABLE} on a table that is symlinked to another disk, create temporary tables on this disk. @item Implement a @code{DATE/DATETIME} type that handles time zone information properly, so that dealing with dates in different time zones is easier. @item FreeBSD and MIT-pthreads; Do sleeping threads take CPU? @item Check if locked threads take any CPU. @item Fix configure so that one can compile all libraries (like @code{MyISAM}) without threads. @item Add an option to periodically flush key pages for tables with delayed keys if they haven't been used in a while. @item Allow join on key parts (optimization issue). @item @code{INSERT SQL_CONCURRENT} and @code{mysqld --concurrent-insert} to do a concurrent insert at the end of the file if the file is read-locked. @item Remember @code{FOREIGN} key definitions in the @file{.frm} file. @item Cascading @code{DELETE} @item Server side cursors. @item Check if @code{lockd} works with modern Linux kernels; If not, we have to fix @code{lockd}! To test this, start @code{mysqld} with @code{--enable-locking} and run the different fork* test suits. They shouldn't give any errors if @code{lockd} works. @item Allow SQL variables in @code{LIMIT}, like in @code{LIMIT @@a,@@b}. @item Allow update of variables in @code{UPDATE} statements. For example: @code{UPDATE TABLE foo SET @@a=a+b,a=@@a, b=@@a+c} @item Change when user variables are updated so that one can use them with @code{GROUP BY}, as in the following example: @code{SELECT id, @@a:=count(*), sum(sum_col)/@@a FROM table_name GROUP BY id}. @item Don't add automatic @code{DEFAULT} values to columns. Give an error when using an @code{INSERT} that doesn't contain a column that doesn't have a @code{DEFAULT}. @item Caching of queries and results. This should be done as a separated module that examines each query and if this is query is in the cache the cached result should be returned. When one updates a table one should remove as few queries as possible from the cache. This should give a big speed bost on machines with much RAM where queries are often repeated (like WWW applications). One idea would be to only cache queries of type: @code{SELECT CACHED ...} @item Fix @file{libmysql.c} to allow two @code{mysql_query()} commands in a row without reading results or give a nice error message when one does this. @item Optimize @code{BIT} type to take 1 bit (now @code{BIT} takes 1 char). @item Check why MIT-pthreads @code{ctime()} doesn't work on some FreeBSD systems. @item Add an @code{IMAGE} option to @code{LOAD DATA INFILE} to not update @code{TIMESTAMP} and @code{AUTO_INCREMENT} fields. @item Added @code{LOAD DATE INFILE.. UPDATE} syntax. @itemize @bullet @item For tables with primary keys, if the data contains the primary key, entries matching that primary key are updated from the remainder of the columns. However, columns MISSING from the incoming data feed are not touched. @item For tables tables with primary keys that are missing some part of the key in the incoming data stream, or that have no primary key, the feed is treated as a @code{LOAD DATA INFILE ... REPLACE INTO} is now. @end itemize @item Make @code{LOAD DATA INFILE} understand syntax like: @example LOAD DATA INFILE 'file_name.txt' INTO TABLE tbl_name TEXT_FIELDS (text_field1, text_field2, text_field3) SET table_field1=concatenate(text_field1, text_field2), table_field3=23 IGNORE text_field3 This can be used to skip over extra columns in the text file, or update columns based on expressions of the read data... @end example @item @code{LOAD DATA INFILE 'file_name' INTO TABLE 'table_name' ERRORS TO err_table_name} This would cause any errors and warnings to be logged into the err_table_name table. That table would have a structure like: @example line_number - line number in data file error_message - the error/warning message and maybe data_line - the line from the data file @end example @item Add true @code{VARCHAR} support (There is already support for this in MyISAM). @item Automatic output from @code{mysql} to netscape. @item @code{LOCK DATABASES}. (with various options) @item Change sort to allocate memory in ``hunks'' to get better memory utilization. @item @code{DECIMAL} and @code{NUMERIC} types can't read exponential numbers; @code{Field_decimal::store(const char *from,uint len)} must be recoded to fix this. @item Functions: ADD_TO_SET(value,set) and REMOVE_FROM_SET(value,set) @item Add use of @code{t1 JOIN t2 ON ...} and @code{t1 JOIN t2 USING ...} Currently, you can only use this syntax with @code{LEFT JOIN}. @item Add full support for @code{unsigned long long} type. @item Many more variables for @code{show status}. Counts for: @code{INSERT}/@code{DELETE}/@code{UPDATE} statements. Records reads and updated. Selects on 1 table and selects with joins. Mean number of tables in select. Number of @code{ORDER BY} and @code{GROUP BY} queries. @item If you abort @code{mysql} in the middle of a query, you should open another connection and kill the old running query. Alternatively, an attempt should be made to detect this in the server. @item Add a handler interface for table information so you can use it as a system table. This would be a bit slow if you requested information about all tables, but very flexible. @code{SHOW INFO FROM tbl_name} for basic table information should be implemented. @item Add support for UNICODE. @item @code{NATURAL JOIN} and @code{UNION JOIN} @item Allow @code{select a from crash_me left join crash_me2 using (a)}; In this case a is assumed to come from the crash_me table. @item Fix that @code{ON} and @code{USING} works with the @code{JOIN} join type. @item Oracle like @code{CONNECT BY PRIOR ...} to search hierarchy structures. @item @code{mysqladmin copy database new-database}. -- Requires COPY command to be added to @code{mysqld} @item Processlist should show number of queries/thread. @item @code{SHOW HOSTS} for printing information about the hostname cache. @item @code{DELETE} and @code{REPLACE} options to the @code{UPDATE} statement (this will delete rows when one gets a duplicate key error while updating). @item Change the format of @code{DATETIME} to store fractions of seconds. @item Add all missing ANSI92 and ODBC 3.0 types. @item Change table names from empty strings to @code{NULL} for calculated columns. @item Don't use 'Item_copy_string' on numerical values to avoid number->string->number conversion in case of: @code{SELECT COUNT(*)*(id+0) FROM table_name GROUP BY id} @item Make it possible to use the new GNU regexp library instead of the current one (The GNU library should be much faster than the old one). @item Change that @code{ALTER TABLE} doesn't abort clients that executes @code{INSERT DELAYED}. @item Fix that when columns referenced in an @code{UPDATE} clause contains the old values before the update started. @item @code{myisamchk}, @code{REPAIR} and @code{OPTIMIZE TABLE} should be able to handle cases where the data and/or index files are symbolic links. @item Add simulation of @code{pread()}/@code{pwrite()} on Windows to enable concurrent inserts. @item A logfile analyzer that could parsed out information about which tables are hit most often, how often multi-table joins are executed, etc. It should help users identify areas or table design that could be optimized to execute much more efficient queries. @item Add @code{SUM(DISTINCT)} @item Add @code{ANY()},@code{EVERY()} and @code{SOME()} group functions. In ANSI SQL these only works on boolean columns, but we can extend these to work on any columns/expressions by applying: value == 0 -> FALSE and value <> 0 -> TRUE. @item Fix that the type for @code{MAX(column)} is the same as the column type. @example create table t1 (a DATE); insert into t1 values (now()); create table t2 select max(a) from t1; show columns from t2; @end example @item Come up with a nice syntax for a statement that will @code{UPDATE} the row if it exists and @code{INSERT} a new row if the row didn't exist. (Like @code{REPLACE} works with @code{INSERT} / @code{DELETE}) @end itemize @node TODO sometime, TODO unplanned, TODO future, TODO @subsection Things that have to be done sometime @itemize @bullet @item Implement function: @code{get_changed_tables(timeout,table1,table2,...)} @item Atomic multi-table updates, eg @code{update items,month set items.price=month.price where items.id=month.id;}; @item Change reading through tables to use memmap when possible. Now only compressed tables use memmap. @item Add a new privilege @strong{'Show_priv'} for @code{SHOW} commands. @item Make the automatic timestamp code nicer. Add timestamps to the update log with @code{SET TIMESTAMP=#;} @item Use read/write mutex in some places to get more speed. @item Full foreign key support. One probably wants to implement a procedural language first. @item Simple views (first on one table, later on any expression). @item Automatically close some tables if a table, temporary table or temporary files gets error 23 (not enough open files). @item When one finds a field=#, change all occurrences of field to #. Now this is only done for some simple cases. @item Change all const expressions with calculated expressions if possible. @item Optimize key = expression. At the moment only key = field or key = constant are optimized. @item Join some of the copy functions for nicer code. @item Change @file{sql_yacc.yy} to an inline parser to reduce its size and get better error messages (5 days). @item Change the parser to use only one rule per different number of arguments in function. @item Use of full calculation names in the order part. (For ACCESS97) @item @code{UNION}, @code{MINUS}, @code{INTERSECT} and @code{FULL OUTER JOIN}. (Currently only @code{LEFT OUTER JOIN} is supported) @item Allow @code{UNIQUE} on fields that can be @code{NULL}. @item @code{SQL_OPTION MAX_SELECT_TIME=#} to put a time limit on a query. @item Make the update log to a database. @item Negative @code{LIMIT} to retrieve data from the end. @item Alarm around client connect/read/write functions. @item Please note the changes to @code{safe_mysqld}: according to FSSTND (which Debian tries to follow) PID files should go into @file{/var/run/<progname>.pid} and log files into @file{/var/log}. It would be nice if you could put the "DATADIR" in the first declaration of "pidfile" and "log", so the placement of these files can be changed with a single statement. @item Allow a client to request logging. @item Add use of @code{zlib()} for @code{gzip}-ed files to @code{LOAD DATA INFILE}. @item Fix sorting and grouping of @code{BLOB} columns (partly solved now). @item Stored procedures. This is currently not regarded to be very important as stored procedures are not very standardized yet. Another problem is that true stored procedures make it much harder for the optimizer and in many cases the result is slower than before We will, on the other hand, add a simple (atomic) update language that can be used to write loops and such in the MySQL server. @item Change to use semaphores when counting threads. One should first implement a semaphore library to MIT-pthreads. @item Don't assign a new @code{AUTO_INCREMENT} value when one sets a column to 0. Use @code{NULL} instead. @item Add full support for @code{JOIN} with parentheses. @item As an alternative for one thread / connection manage a pool of threads to handle the queries. @item Allow one to get more than one lock with @code{GET_LOCK}. When doing this, one must also handle the possible deadlocks this change will introduce. @end itemize Time is given according to amount of work, not real time. @node TODO unplanned, , TODO sometime, TODO @subsection Some things we don't have any plans to do @itemize @bullet @item Nothing; In the long run we plan to be fully ANSI 92 / ANSI 99 compliant. @end itemize @node Installing, Tutorial, Introduction, Top @chapter MySQL Installation @cindex installing, overview @menu * Quick Standard Installation:: Quick Standard Installation of MySQL * General Installation Issues:: General Installation Issues * Installing source:: Installing a MySQL source distribution * Post-installation:: Post-installation setup and testing * Upgrade:: Upgrading/Downgrading MySQL * Operating System Specific Notes:: Operating System Specific Notes @end menu This chapter describes how to obtain and install MySQL: @itemize @bullet @item For a list of sites from which you can obtain MySQL, see @ref{Getting MySQL, , Getting MySQL}. @item To see which platforms are supported, see @ref{Which OS}. Please note that not all supported system are equally good for running MySQL on them. On some it is much more robust and efficient than others - see @ref{Which OS} for details. @item Several versions of MySQL are available in both binary and source distributions. We also provide public access to our current source tree for those who want to see our most recent developments and help us test new code. To determine which version and type of distribution you should use, see @ref{Which version}. When in doubt, use the binary distribution. @item Installation instructions for binary and source distributions are described in @ref{Installing binary} and @ref{Installing source}. Each set of instructions includes a section on system-specific problems you may run into. @item For post-installation procedures, see @ref{Post-installation}. These procedures apply whether you install MySQL using a binary or source distribution. @end itemize @node Quick Standard Installation, General Installation Issues, Installing, Installing @section Quick Standard Installation of MySQL @c This node name is special @menu * Linux-RPM:: Installing MySQL on Linux * Windows installation:: Installing MySQL on Windows @end menu @node Linux-RPM, Windows installation, Quick Standard Installation, Quick Standard Installation @subsection Installing MySQL on Linux @cindex RPM file @cindex RedHat Package Manager The recommended way to install MySQL on Linux is by using an RPM file. The MySQL RPMs are currently being built on a RedHat Version 6.2 system but should work on other versions of Linux that support @code{rpm} and use @code{glibc}. If you have problems with an RPM file, for example, if you receive the error ``@code{Sorry, the host 'xxxx' could not be looked up}'', see @ref{Binary notes-Linux}. The RPM files you may want to use are: @itemize @bullet @item @code{MySQL-VERSION.i386.rpm} The MySQL server. You will need this unless you only want to connect to a MySQL server running on another machine. @item @code{MySQL-client-VERSION.i386.rpm} The standard MySQL client programs. You probably always want to install this package. @item @code{MySQL-bench-VERSION.i386.rpm} Tests and benchmarks. Requires Perl and msql-mysql-modules RPMs. @item @code{MySQL-devel-VERSION.i386.rpm} Libraries and include files needed if you want to compile other MySQL clients, such as the Perl modules. @item @code{MySQL-VERSION.src.rpm} This contains the source code for all of the above packages. It can also be used to try to build RPMs for other architectures (for example, Alpha or SPARC). @end itemize To see all files in an RPM package, run: @example shell> rpm -qpl MySQL-VERSION.i386.rpm @end example To perform a standard minimal installation, run: @example shell> rpm -i MySQL-VERSION.i386.rpm MySQL-client-VERSION.i386.rpm @end example To install just the client package, run: @example shell> rpm -i MySQL-client-VERSION.i386.rpm @end example The RPM places data in @file{/var/lib/mysql}. The RPM also creates the appropriate entries in @file{/etc/rc.d/} to start the server automatically at boot time. (This means that if you have performed a previous installation, you may want to make a copy of your previously installed MySQL startup file if you made any changes to it, so you don't lose your changes.) After installing the RPM file(s), the @code{mysqld} daemon should be running and you should now be able to start using MySQL. @xref{Post-installation}. If something goes wrong, you can find more information in the binary installation chapter. @xref{Installing binary}. @node Windows installation, , Linux-RPM, Quick Standard Installation @subsection Installing MySQL on Windows The following instructions apply to precompiled binary distributions. If you download a source distribution, you will have to compile and install it yourself. If you don't have a copy of the MySQL distribution, you should first download one from @uref{http://www.mysql.com/downloads/mysql-3.23.html}. If you plan to connect to MySQL from some other program, you will probably also need the @strong{MyODBC} driver. You can find this at the @strong{MyODBC} download page (@uref{http://www.mysql.com/downloads/api-myodbc.html}). To install either distribution, unzip it in some empty directory and run the @code{Setup.exe} program. By default, MySQL-Windows is configured to be installed in @file{C:\mysql}. If you want to install MySQL elsewhere, install it in @file{C:\mysql} first, then move the installation to where you want it. If you do move MySQL, you must indicate where everything is located by supplying a @code{--basedir} option when you start the server. For example, if you have moved the MySQL distribution to @file{D:\programs\mysql}, you must start @code{mysqld} like this: @example C:\> D:\programs\mysql\bin\mysqld --basedir D:\programs\mysql @end example Use @code{mysqld --help} to display all the options that @code{mysqld} understands! With all newer MySQL versions, you can also create a @file{C:\my.cnf} file that holds any default options for the MySQL server. Copy the file @file{\mysql\my-xxxxx.cnf} to @file{C:\my.cnf} and edit it to suit your setup. Note that you should specify all paths with @samp{/} instead of @samp{\}. If you use @samp{\}, you need to specify it twice, because @samp{\} is the escape character in MySQL. @xref{Option files}. Starting with MySQL 3.23.38, the Windows distribution includes both the normal and the @strong{MySQL-Max} binaries. The main benefit of using the normal @code{mysqld.exe} binary is that it's a little faster and uses less resources. Here is a list of the different MySQL servers you can use: @multitable @columnfractions .25 .75 @item @code{mysqld} @tab Compiled with full debugging and automatic memory allocation checking, symbolic links, BDB and InnoDB tables. @item @code{mysqld-opt} @tab Optimized binary with no support for transactional tables. @item @code{mysqld-nt} @tab Optimized binary for NT with support for named pipes. You can run this version on Win98, but in this case no named pipes are created and you must have TCP/IP installed. @item @code{mysqld-max} @tab Optimized binary with support for symbolic links, BDB and InnoDB tables. @item @code{mysqld-max-nt} @tab Like @code{mysqld-max}, but compiled with support for named pipes. @end multitable Starting from 3.23.50, named pipes are only enabled if mysqld is started with @code{--enable-named-pipe}. All of the above binaries are optimized for the Pentium Pro processor but should work on any Intel processor >= i386. NOTE: If you want to use InnoDB tables, there are certain startup options that must be specified in your @file{my.ini} file! @xref{InnoDB start}. @node General Installation Issues, Installing source, Quick Standard Installation, Installing @section General Installation Issues @c @node Methods of Installation, , , @c @subsection Methods of Installation @c FIX: this needs to be written? @menu * Getting MySQL:: How to Get MySQL * Which OS:: Operating Systems Supported by MySQL * Which version:: Which MySQL Version to Use * Installation layouts:: Installation Layouts * Many versions:: How and When Updates Are Released * MySQL binaries:: MySQL Binaries Compiled by MySQL AB @end menu @node Getting MySQL, Which OS, General Installation Issues, General Installation Issues @subsection How to Get MySQL @cindex downloading @cindex MySQL version @cindex version, latest @cindex getting MySQL @cindex mirror sites @cindex URLS for downloading MySQL Check the @uref{http://www.mysql.com/, MySQL home page} for information about the current version and for downloading instructions. Our main download mirror is located at: @uref{http://download.sourceforge.net/mirrors/mysql/} If you are interested in becoming a MySQL mirror site, you may anonymously rsync with: @code{rsync://download.sourceforge.net/mysql/}. Please send e-mail to @email{webmaster@@mysql.com} notifying us of your mirror to be added to the list below. If you have problems downloading from our main site, try using one of the mirrors listed below. Please report bad or out-of-date mirrors to @email{webmaster@@mysql.com}. @c START_OF_MIRROR_LISTING @strong{Europe:} @itemize @bullet @item @c EMAIL: sl@iuinfo.tuwien.ac.at (Tony Sprinzl) @image{Flags/austria} Austria [Univ. of Technology/Vienna] @ @uref{http://gd.tuwien.ac.at/db/mysql/, WWW} @uref{ftp://gd.tuwien.ac.at/db/mysql/, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c EMAIL: delian@naturella.com (Delian Delchev) @c @image{Flags/bulgaria} Bulgaria [Naturella] @ @c @uref{http://archive.nat.bg/pub/mirror/mysql/, WWW} @c @uref{ftp://ftp.ntrl.net/pub/mirror/mysql/, FTP} @item @c EMAIL: salle@online.bg (Admin) @image{Flags/bulgaria} Bulgaria [online.bg/Sofia] @ @uref{http://mysql.online.bg/, WWW} @uref{ftp://mysql.online.bg/, FTP} @c @item @c Added: 990614 @c Not ok 20000919; Out of date (Matt) @c EMAIL: vuksan@veus.hr (Vladimir Vuksan) @c @image{Flags/croatia} Croatia [HULK] @ @c @uref{http://ftp.linux.hr/pub/mysql/, WWW} @c @uref{ftp://ftp.linux.hr/pub/mysql/, FTP} @item @c Added: 990614 @c EMAIL: kas@informatics.muni.cz (Jan Kasprzak) @image{Flags/czech-republic} Czech Republic [Masaryk University in Brno] @ @uref{http://mysql.linux.cz/index.html, WWW} @uref{ftp://ftp.fi.muni.cz/pub/mysql/, FTP} @item @c Added: 990920 @c EMAIL: radek@sopik.cz (Radek Libovicky) @image{Flags/czech-republic} Czech Republic [www.sopik.cz] @ @uref{http://www.mysql.cz/, WWW} @item @c Added: 000418 @c EMAIL: feela@ipex.cz (Ondrej Feela Filip) @image{Flags/czech-republic} Czech Republic [www.gin.cz] @ @uref{http://mysql.gin.cz/, WWW} @uref{ftp://ftp.gin.cz/pub/MIRRORS/www.mysql.com/, FTP} @item @c removed 991020 (no DNS entry). New name 991026. Added 991121 @c Statistics at http://mirror.borsen.dk/ @c EMAIL: guru@borsen.dk (Jesper Angelo) @image{Flags/denmark} Denmark [Borsen] @ @uref{ http://mysql.borsen.dk/, WWW} @item @c EMAIL: mkp@socsci.auc.dk (Martin Kasper Petersen) @image{Flags/denmark} Denmark [SunSITE] @ @uref{http://mirrors.sunsite.dk/mysql/, WWW} @uref{ftp://sunsite.dk/mirrors/mysql/, FTP} @c @item @c T�nu does not work there anymore :-) 990720 @c EMAIL: tonu@tradenet.ee (T�nu Samuel) @c @image{Flags/estonia} Estonia [Tradenet] @ @c @uref{http://mysql.tradenet.ee, WWW} @item @c EMAIL: tonu@spam.ee (T�nu Samuel) @image{Flags/estonia} Estonia [OKinteractive] @ @uref{http://mysql.mirror.ok.ee, WWW} @c @item @c Changed 990531 @c Not ok 20000919; Old site (Matt) @c EMAIL: Steeve.Devergne@minet.net (Steeve Devergne) @c @image{Flags/france} France [minet] @ @c @uref{http://www.minet.net/devel/mysql/, WWW} @item @c EMAIL: alex@mtesa.net (Alexandre Dupouy) @image{Flags/france} France [mtesa.net] @ @uref{http://mysql.mtesa.net/, WWW} @item @c EMAIL: mrjack@fastorama.com (MrJacK) @image{Flags/france} France [fastorama.com, Chatenois] @ @uref{http://mysql.fastorama.com/, WWW} @uref{ftp://ftp.fastorama.com/unix/mysql/, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c EMAIL: Jaakko.Hyvatti@eunet.fi @c @image{Flags/finland} Finland [EUnet] @ @c @uref{http://mysql.eunet.fi/, WWW} @c @item @c Added 990829 @c Not ok 20000919; Non-existent (Matt) @c EMAIL: tomi.hakala@clinet.fi (Tomi Hakala) @c @image{Flags/finland} Finland [clinet] @ @c @uref{ftp://ftp.clinet.fi/mirrors/ftp.mysql.org/pub/mysql/, FTP} @item @c Added 20001019 @c EMAIL: d@d.tj (Dominique L . Bouix) @image{Flags/finland} Finland [tonnikala.net] @ @uref{http://mysql.tonnikala.org/, WWW} @item @c Added 981208 @c EMAIL: noel@uni-bonn.de (Noel Koethe) @image{Flags/germany} Germany [Kernelnotes.de, Bonn] @ @uref{http://www.kernelnotes.de/mysql/, WWW} @uref{ftp://ftp.kernelnotes.de/pub/mirror/mysql.org/, FTP} @item @c EMAIL: th@rz.fh-wolfenbuettel.de (Thorsten Ludewig) @image{Flags/germany} Germany [Wolfenbuettel] @ @uref{http://www.fh-wolfenbuettel.de/ftp/pub/database/mysql/, WWW} @uref{ftp://ftp.fh-wolfenbuettel.de/pub/database/mysql/, FTP} @c @item @c Ok 980114. Removed 981208 (down > 3 days) ok 981214 @c Not ok 20000919; Non-existent (Matt) @c EMAIL: straub@gks.de (Hans-Peter Straub) @c @image{Flags/germany} Germany [Staufen] @ @c @uref{http://mysql.staufen.de/, WWW} @c @item @c Added 990614 @c Not ok 20000919; Non-existent (Matt) @c EMAIL: thomas.rohde@ecrc.de (Thomas Rohde) @c @image{Flags/germany} Germany [Cable & Wireless] @ @c @uref{ftp://ftp.ecrc.net/pub/database/mysql/, FTP} @item @c Added 981208 @c EMAIL: christias@noc.ntua.gr (Panagiotis Christias) @image{Flags/greece} Greece [NTUA, Athens] @ @uref{http://www.ntua.gr/mysql/, WWW} @uref{ftp://ftp.ntua.gr/pub/databases/mysql/, FTP} @item @c EMAIL: torlasz@xenia.sote.hu (Laszlo L. Tornoc) @image{Flags/hungary} Hungary [Xenia] @ @uref{http://mysql.sote.hu/, WWW} @uref{ftp://xenia.sote.hu/pub/mirrors/www.mysql.com/, FTP} @item @c EMAIL: hendlein@tiszanet.hu (Hendlein Peter) @image{Flags/hungary} Hungary [TiszaneT] @ @uref{http://mysql.tiszanet.hu/, WWW} @uref{ftp://mysql.tiszanet.hu/pub/mirrors/mysql/, FTP} @item @c EMAIL: i.habencius@telnet.hu (Habencius Istvan) @image{Flags/hungary} Hungary [stop.hu] @ @uref{http://mysql.mirror.stop.hu/, WWW} @item @c EMAIL: mirrors@gm.is (Tomas Edwardsson) @image{Flags/iceland} Iceland [GM] @ @uref{http://mysql.gm.is/, WWW} @uref{ftp://ftp.gm.is/pub/mysql/, FTP} @c @item @c Out of date 990906 @c EMAIL: bourbon@netvision.net.il (Zeev Suraski) @c @image{Flags/israel} Israel [Netvision] @ @c @uref{http://mysql.netvision.net.il/, WWW} @c @item @c Not working 99.03.06 @c EMAIL: maruzz@matrice.it (Giovanni Maruzzelli) @c @image{Flags/italy} Italy [Matrice] @ @c @uref{http://www.matrice.it/risorse/mysql/, WWW} @item @c EMAIL: gabriele@feelinglinux.com (Giansante Gabriele) @image{Flags/italy} Italy [feelinglinux.com] @ @uref{http://mysql.feelinglinux.com/, WWW} @item @c EMAIL: irena@yacc.it @image{Flags/italy} Italy [Teta Srl] @ @uref{http://www.teta.it/mysql/, WWW} @item @c EMAIL: rudy@comm2000.it (Rudy) @image{Flags/italy} Italy [tzone.it] @ @uref{http://mysql.tzone.it/, WWW} @c @item @c Added 991121 @c Not ok 20000919; Old site (Matt) @c EMAIL: nick@iol.ie (Nick Hilliard) @c @image{Flags/ireland} Ireland [Ireland On-Line/Dublin] @ @c @uref{http://mysql.iol.ie, WWW} @c @uref{ftp://ftp.iol.ie/pub/mysql, FTP} @item @c Added 20001031 @c EMAIL: dave@esat.net (Dave Rynne) @image{Flags/ireland} Ireland [Esat Net] @ @uref{http://ftp.esat.net/mirrors/download.sourceforge.net/pub/mirrors/mysql/, WWW} @uref{ftp://ftp.esat.net/mirrors/download.sourceforge.net/pub/mirrors/mysql/, FTP} @item @c Added 20010524 @c EMAIL: arvids@parks.lv (Arvids) @image{Flags/latvia} Latvia [linux.lv] @ @uref{ftp://ftp.linux.lv/pub/software/mysql/, FTP} @item @c Added 20001125 @c EMAIL: mleicher@silverpoint.nl (Marcel Leicher) @image{Flags/netherlands} Netherlands [Silverpoint] @ @uref{http://mysql.silverpoint.nl/, WWW} @item @c EMAIL: mysql@widexs.nl (Wouter de Jong) @image{Flags/netherlands} Netherlands [Widexs BV] @ @uref{http://mysql.widexs.nl/, WWW} @uref{ftp://mysql.widexs.nl/pub/mysql/, FTP} @item @c EMAIL: peter@proserve.nl (ProServe - Peter Batenburg) @image{Flags/netherlands} Netherlands [ProServe] @ @uref{http://mysql.proserve.nl/, WWW} @item @c EMAIL: W.Sylwestrzak@icm.edu.pl (Wojtek Sylwestrzak) @c mirroring nightly at 05:25 @image{Flags/poland} Poland [Sunsite] @ @uref{http://sunsite.icm.edu.pl/mysql/, WWW} @uref{ftp://sunsite.icm.edu.pl/pub/unix/mysql/, FTP} @item @c EMAIL: admin@net.ncservice.com.pl (Marian Witkowski) @image{Flags/poland} Poland [ncservice.com/Gdansk] @ @uref{http://mysql.service.net.pl/, WWW} @c @item @c EMAIL: melo@co.telenet.pt (Pedro Melo) @c Temp out of service (email from Pedro) @c @image{Flags/portugal} Portugal [IP] @ @c @uref{http://mysql.ip.pt, WWW} @c @item @c Not ok 20000919; Non-existent (Matt) @c EMAIL: support@leirianet.pt (Equipa de suporte do Leirianet) @c @image{Flags/portugal} Portugal [lerianet] @ @c @uref{http://mysql.leirianet.pt, WWW} @c @uref{ftp://ftp.leirianet.pt/pub/mysql/,FTP} @item @c Added 20001031 @c bofh@netc.pt (Bruno Rodrigues) @image{Flags/portugal} Portugal [Netc] @ @uref{http://ftp.netc.pt/pub/mysql/, WWW} @uref{ftp://ftp.netc.pt/pub/mysql/, FTP} @item @c EMAIL: keeper@roedu.net (Mihai RUSU) @image{Flags/romania} Romania [roedu.net/Bucharest] @ @uref{ftp://ftp.roedu.net/pub/mirrors/ftp.mysql.com/, FTP} @item @c EMAIL: kuzmin@dn.ru (Roma Kuzmin) @image{Flags/russia} Russia [DirectNet] @ @uref{http://mysql.directnet.ru/, WWW} @uref{ftp://ftp.dn.ru/pub/MySQL/, FTP} @c @item @c down 990113 @c EMAIL: nikkic@cityline.ru (Nikki Chumakov) @c @image{Flags/russia} Russia [Cityline] @ @c @uref{ftp://mysql.cityline.ru/pub/mysql, FTP} @c @uref{http://mysql.cityline.ru, WWW} @c EMAIL: bar@izhcom.ru (Alexander I Barkov) @c @item @c Not ok 20000919; Non-existent (Matt) @c @image{Flags/russia} Russia [IZHCOM] @ @c @uref{http://mysql.udm.net/, WWW} @c @uref{ftp://ftp.izhcom.ru/pub/mysql/,FTP} @item @c Added 990507 @c EMAIL: demon@gpad.ac.ru (Dima Sivachenko) @image{Flags/russia} Russia [Scientific Center/Chernogolovka] @ @uref{ftp://ftp.chg.ru/pub/databases/mysql/, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c EMAIL: sebi@dnttm.ro (Sebastian DEAC) @c @image{Flags/romania} Romania [Timisoara] @ @c @uref{http://download.dnttm.ro/mysql, WWW} @c @uref{ftp://ftp.dnttm.ro/pub/mysql, FTP} @c @item @c Not ok 20001123; Non-existent (Matt) @c EMAIL: tim@lbi.ro (Bogdan Surdu) @c @image{Flags/romania} @c Romania [Bucharest] @ @c @uref{http://www.lbi.ro/MySQL/, WWW} @c @uref{ftp://ftp.lbi.ro/mirrors/ftp.tcx.se, FTP} @c @item @c Removed 20000521 because there is no mirror here. @c EMAIL: jips@masterd.es (Juan Ignacio P�rez Sacrist�n) @c @image{Flags/spain} Spain [MasterD] @c @uref{http://mysql.masterd.es, WWW} @item @c EMAIL: Patrik.Karen@sdi.slu.se (Patrik Karen) @c ftp -> remove old files @image{Flags/sweden} Sweden [Sunet] @ @uref{http://ftp.sunet.se/pub/unix/databases/relational/mysql/, WWW} @uref{ftp://ftp.sunet.se/pub/unix/databases/relational/mysql/, FTP} @item @c EMAIL: archive@sunsite.cnlab-switch.ch (Thomas Lenggenhager) @image{Flags/switzerland} Switzerland [Sunsite] @ @uref{http://sunsite.cnlab-switch.ch/ftp/mirror/mysql/, WWW} @uref{ftp://sunsite.cnlab-switch.ch/mirror/mysql/, FTP} @c @item @c Not ok 20000919; Non-existent (Matt) @c simon@oyster.co.uk (Simon Gornall) @c @image{Flags/great-britain} UK [Oyster/UK] @ @c @uref{ftp://ftp.oyster.co.uk/pub/mysql, FTP} @item @c EMAIL: keet@mordor.plig.net (Christiaan Keet) @image{Flags/great-britain} UK [PLiG/UK] @ @uref{http://ftp.plig.org/pub/mysql/, WWW} @uref{ftp://ftp.plig.org/pub/mysql/, FTP} @c @item @c Not ok 20010808; Non-existent (Matt) @c EMAIL: sean@telekon.co.uk (Sean Gibson) @c @image{Flags/great-britain} UK [Telekon Internet/UK] @ @c @uref{ftp://ftp.telekon.co.uk/pub/mysql/, FTP} @c @item @c lance@uklinux.net (Lance) @c @image{Flags/great-britain} UK [uklinux.net] @ @c @uref{http://mirror.uklinux.net/mysql/, WWW} @c @uref{ftp://mirror.uklinux.net/pub/mysql/, FTP} @c @item @c unknown @c @image{Flags/great-britain} UK [MicroMuse] @ @c @uref{ftp://ftp.micromuse.co.uk/pub/packages/unix/databases/mysql/, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c lmjm@icparc.ic.ac.uk (Lee McLoughlin) @c @image{Flags/great-britain} UK [SunSITE] @ @c @uref{http://sunsite.org.uk/packages/mysql/, WWW} @c @uref{ftp://sunsite.org.uk/packages/mysql/, FTP} @item @c EMAIL: sander@paco.net (Alexander Ivanov) @image{Flags/ukraine} Ukraine [PACO] @ @uref{http://mysql.paco.net.ua, WWW} @uref{ftp://mysql.paco.net.ua/, FTP} @item @c EMAIL: mizi@alkar.net (Alexander Ryumshin) @image{Flags/ukraine} Ukraine [ISP Alkar Teleport/Dnepropetrovsk] @ @uref{http://mysql.dp.ua/, WWW} @item @c EMAIL: bole@bolex.bolex.co.yu (Bosko Radivojevic) @image{Flags/yugoslavia} Yugoslavia [bolex.co.yu] @ @uref{http://mysql.boa.org.yu/, WWW} @uref{ftp://ftp.linux.org.yu/pub/MySQL/, FTP} @end itemize @strong{North America:} @itemize @bullet @c @item @c Not ok 990101 (only to 981007) @c EMAIL: sysop@polarcom.com (Seamus Venasse) @c @image{Flags/canada} Canada [Polaris Computing] @ @c @uref{http://mysql.polaris.ca/, WWW} @item @c Ok 980109 @c EMAIL: wojtek@tryc.on.ca (Wojtjeck Tryc) @image{Flags/canada} Canada [Tryc] @ @uref{http://web.tryc.on.ca/mysql/, WWW} @c @item @c not updated 990218. Added again 990918 @c Not ok 20000919; Non-existent (Matt) @c EMAIL: rhooper@cyberus.ca (Roy Hooper) @c @image{Flags/canada} Canada [Cyberus] @ @c @uref{http://mysql.cyberus.ca/, WWW} @c @uref{ftp://mysql.cyberus.ca/, FTP} @item @c EMAIL: mleber@he.net (Mike Leber) @c Added 980312 @image{Flags/usa} USA [Hurricane Electric/San Jose] @ @uref{http://mysql.he.net/, WWW} @item @c EMAIL: ask@valueclick.com (Ask Bjoern Hansen) @c Added 20000925 @image{Flags/usa} USA [ValueClick, Los Angeles CA] @ @uref{http://mysql.valueclick.com/, WWW} @uref{ftp://mysql.valueclick.com/pub/mysql/Downloads/, FTP} @c @item @c Not ok 20000919; Non-existent (Matt) @c EMAIL: meltzer@icsnet.com (Jeffrey Meltzer) @c Added 000108 @c @image{Flags/usa} USA [Meltzer/New York State] @ @c @uref{ftp://ftp.meltzer.org/pub/mysql/, FTP} @c @item @c No such directory 990830 @c EMAIL: tps@users.buoy.com (Tim Sailer) @c @image{Flags/usa} USA [Buoy/New York] @ @c @uref{http://www.buoy.com/mysql/, WWW} @c @item @c EMAIL: db@hpnc.com (Douglas Bowyer) @c Added 980107, removed 981124 because of 'file not found' @c @image{Flags/usa} USA [Hypernet Communications/Dallas] @ @c @uref{http://epsilon.hpnc.com/mysql, WWW} @c @item @c ********************************** @c Not updated 980106 @c EX: twh@iquest.net (Thomas Holt) who no longer works there @c @image{Flags/usa} USA [IQuest/Indiana] @ @c @uref{http://mirrors.iquest.net/mysql/, WWW} @c @item @c ********************************** @c Only a partial mirror so we exclude it from the list @c EX: lindberg@id.wustl.edu (Fred Lindberg) @c @image{Flags/usa} USA [Washington University/St. Louis] @ @c @uref{ftp://ftp.id.wustl.edu/pub/database/mysql/, FTP} @c removed 991111 -> no answer @c @item @c EMAIL: andrew@netcasting.net (Andrew Sawyers) @c @image{Flags/usa} USA [Netcasting/West Coast] @ @c @uref{ftp://ftp.netcasting.net/pub/mysql/, FTP} @c @item @c No mirror! 980809 David @c EMAIL: savages@savages.com (Shaun Savage) @c @image{Flags/usa} USA [Savages/Oregon] @ @c @uref{http://mysql.savages.com, WWW} @c @item @c Not ok 20000919; Non-existent (Matt) @c EMAIL: paul@gina.net (Paul Vining) @c mirrors ftp.sunet.se @c @image{Flags/usa} USA [Gina net/Florida] @ @c @uref{http://www.gina.net/mysql/, WWW} @c Out of date 2000-01-08 (Not updated since 1999-10) @c @item @c EMAIL: wswanson@pingzero.net (Wylie Swanson) @c mirrors mysql.org @c @image{Flags/usa} USA [pingzero/Los Angeles] @ @c @uref{http://mysql.pingzero.net/, WWW} @item @c EMAIL: ftpkeeper@mirror.sit.wisc.edu @image{Flags/usa} USA [Wisconsin University/Wisconsin] @ @uref{http://mirror.sit.wisc.edu/mysql/, WWW} @uref{ftp://mirror.sit.wisc.edu/mirrors/mysql/, FTP} @c @item @c Not ok 20000919; Out of date (Matt) @c EMAIL: ftp-admin@digex.net @c @image{Flags/usa} USA [DIGEX] @ @c @uref{ftp://ftp.digex.net/pub/packages/database/mysql/, FTP} @item @c EMAIL: andrew.sawyers@thelinuxstore.com @image{Flags/usa} USA [LinuxWired/Scottsdale, AZ] @ @uref{http://mysql.linuxwired.net/, WWW} @uref{ftp://ftp.linuxwired.net/pub/mirrors/mysql/, FTP} @c @item @c EMAIL: dan@surfsouth.com (Dan Muntz) @c @image{Flags/usa} USA [Venoma.Org/Valdosta, GA] @ @c @uref{http://mysql.venoma.org/, WWW} @item @c EMAIL: hkind@adgrafix.com (Hans Kind) @image{Flags/usa} USA [adgrafix.com/Boston, MA] @ @uref{http://mysql.adgrafix.com/, WWW} @item @c EMAIL: Pjacob@netnumina.com (Philip Jacob) @image{Flags/usa} USA [netNumina/Cambridge, MA] @ @uref{http://mysql.mirrors.netnumina.com/, WWW} @item @c EMAIL: hagler@ahaza.com (Mark Hagler) @image{Flags/usa} USA [Ahaza Systems/Seattle, WA] @ @uref{http://mysql.mirrortree.com/, WWW} @uref{ftp://mysql.mirrortree.com/pub/mysql/, FTP} @end itemize @strong{South America:} @itemize @bullet @item @c Added 20001102 @c EMAIL: nico@bannerlandia.com (Nicolas Moldavsky) @image{Flags/argentina} Argentina [bannerlandia.com] @ @uref{http://mysql.bannerlandia.com.ar/, WWW} @uref{ftp://mysql.bannerlandia.com.ar/mirrors/mysql/, FTP} @c @item @c Not ok 20000919; Non-existent (Matt) @c EMAIL: gaiser@matrix.com.br (Roberto Gaiser) @c @image{Flags/brazil} Brazil [Matrix] @ @c @uref{http://mysql.matrix.com.br, WWW} @item @c jpabuyer@vision.cl @image{Flags/chile} Chile [Vision] @ @uref{http://mysql.vision.cl/, WWW} @item @c EMAIL: bartschm@psi.com (Marcelo Bartsch) @image{Flags/chile} Chile [PSINet] @ @uref{http://mysql.psinet.cl/, WWW} @uref{ftp://ftp.psinet.cl/pub/database/mysql/, FTP} @item @c EMAIL: jpabuyer@tecnoera.com (Juan Pablo Abuyeres) @image{Flags/chile} Chile [Tecnoera] @ @uref{http://mysql.tecnoera.com/, WWW} @c @item @c Removed 990730 @c @c EMAIL: dan@amerikanclaris.com (Danilo Lotina F.) @c @image{Flags/chile} Chile [Amerikanclaris] @ @c @uref{http://www.labs.amerikanclaris.cl/mysql, WWW} @c @uref{ftp://ftp.amerikanclaris.cl/pub/mysql, FTP} @end itemize @strong{Asia:} @itemize @bullet @c @item @c Host cannot be resolved for last week at least 20010809 Tonu @c EMAIL: mirnshi@263.net (Meng Lingbo) @c @image{Flags/china} China [Freecode] @ @c @uref{http://www.freecode.net.cn/mirror/mysql/, WWW} @item @c EMAIL: yusun@atwell.co.jp (yu sun) @image{Flags/china} China [linuxforum.net] @ @uref{http://www2.linuxforum.net/mirror/mysql/, WWW} @item @c EMAIL: vfong@hklpg.org (Vincent Fong) @image{Flags/china} China [HKLPG/Hong Kong] @ @uref{http://mysql.hklpg.org, WWW} @item @c EMAIL: jason-wong@gremlins.com.hk (Gremlins Jason Wong) @image{Flags/china} China [Gremlins/Hong Kong] @ @uref{http://mysql.gremlins.com.hk/, WWW} @uref{ftp://ftp.mirrors.gremlins.com.hk/mysql/, FTP} @item @c EMAIL: shell@linux.org.hk (Shell Hung) @image{Flags/china} China [shellhung.org/Hong Kong] @ @uref{http://mysql.shellhung.org/, WWW} @uref{ftp://ftp.shellhung.org/pub/Mirror/mysql/, FTP} @c @item @c Not ok 20010308; Other content! (Tonu) @c EMAIL: xcyber@yahoo.com (xcyber) @c @image{Flags/china} China [xcyber.org/Hong Kong] @ @c @uref{http://mysql.xcyber.org/, WWW} @c @item @c Not ok 20010330; Non-existent! (Matt) @c EMAIL: marquischan@hotmail.com (Marquis Chan) @c @image{Flags/china} China [TraLand.com/Hong Kong] @ @c @uref{http://www.traland.com/mysql/, WWW} @c @item @c Not ok 20000919; Non-existent (Matt) @c EMAIL: george@netfirm.net (Hongsheng Zhu) @c @image{Flags/china} China [Netfirm] @ @c @uref{http://mysql.netfirm.net, WWW} @item @c EMAIL: dnata@incaf.net (Denie Nataprawira) @image{Flags/indonesia} Indonesia [incaf.net] @ @uref{http://mysql.incaf.net/, WWW} @item @c EMAIL: andika@piksi.itb.ac.id (Andika Triwidada) @image{Flags/indonesia} Indonesia [web.id] @ @uref{http://mysql.itb.web.id/, WWW} @uref{ftp://mysql.itb.web.id/pub/MySQL/, FTP} @item @c Ok 980805 @c EMAIL: takeshi@SoftAgency.co.jp @image{Flags/japan} Japan [Soft Agency] @ @uref{http://www.softagency.co.jp/MySQL, WWW} @item @c EMAIL: takafumi@u-aizu.ac.jp (Takafumi Hayashi) @image{Flags/japan} Japan [u-aizu.ac.jp/Aizu] @ @uref{ftp://ftp.u-aizu.ac.jp/ftp/pub/dbms/mysql/mysql.com, FTP} @c @item @c Ok 980109 Removed 990730 @c EMAIL: satoshi@HappySize.co.jp (Satoshi TATSUOKA) @c @image{Flags/japan} Japan [HappySize] @ @c @uref{http://www.happysize.co.jp/mysql/, WWW} @c @uref{ftp://ftp.happysize.co.jp/pub/mysql/, FTP} @c @item @c Ok 981204 @c Not ok 20000919; Old site (Matt) @c EMAIL: hiroyuki@nucba.ac.jp (hiroyuki kurimoto) @c @image{Flags/japan} Japan [Nagoya Syouka University] @ @c @uref{http://mirror.nucba.ac.jp/mirror/mysql, WWW} @c @uref{ftp://mirror.nucba.ac.jp/mirror/mysql, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c EMAIL: ahmlhs@nmsvr.chosun.com (Ho-sun Lee) @c @image{Flags/south-korea} South Korea [KREONet] @ @c @uref{http://linux.kreonet.re.kr/mysql/, WWW} @item @c EMAIL: jasper@webiiz.com (Kang, Tae-jin) @image{Flags/south-korea} South Korea [Webiiz] @ @uref{http://mysql.webiiz.com/, WWW} @item @c EMAIL: hollywar@holywar.net (Oh Junseon) @image{Flags/south-korea} South Korea [PanworldNet] @ @uref{http://mysql.holywar.net/, WWW} @c @item @c ftp -> remove old files @c EX: ahmlhs@nmsvr.chosun.com (Ho-sun Lee) @c @image{Flags/south-korea} South Korea [KREONet] @ @c @uref{ftp://linux.kreonet.re.kr/pub/tools/db/mysql/, FTP} @c @item @c Removed 990308 @c EMAIL: terence@com5.net (Terence Chan) @c @image{Flags/singapore} Singapore [Com5 Productions] @ @c @uref{http://mysql.com5.net, WWW} @c @uref{ftp://ftp.com5.net/pub/mysql, FTP} @item @c EMAIL: csy@hjc.edu.sg @image{Flags/singapore} Singapore [HJC] @ @uref{http://mysql.hjc.edu.sg, WWW} @uref{ftp://ftp.hjc.edu.sg/mysql, FTP} @c @item @c Not ok 20000919; Old site (Matt) @c 991118: Removed because a user complained about that the page contains @c nothing about MySQL. 991119: Added again because it is a mirror again @c EMAIL: dean@ht.net.tw (Dean Lin) @c @image{Flags/taiwan} Taiwan [HT] @ @c @uref{http://mysql.ht.net.tw, WWW} @item @c EMAIL: linda@ttn.com.tw (Linda Hu) @image{Flags/taiwan} Taiwan [TTN] @ @uref{http://mysql.ttn.net, WWW} @c @item @c Ok 980321 No connect -> removed 990730 @c EMAIL: tby@ccca.nctu.edu.tw (Bao-Yi Tuang) @c @image{Flags/taiwan} Taiwan [NCTU] @ @c @uref{http://mysql.taconet.com.tw, WWW} @c @item @c Out of date 990905 (~2 months) @c @item @c ********************************** @c Error 980106 @c EX: WolfySu@acer.net (Wolfy Su) @c @image{Flags/taiwan} Taiwan [Acer] @ @c @uref{http://mysql.acer.net/, WWW} @c @item @c ********************************** @c files to delete @c EX: service@wownet.net @c @image{Flags/taiwan} Taiwan [Wownet] @ @c @uref{ftp://ftp.wownet.net/mysql/, FTP} @c @item @c ********************************** @c No contact 980106 @c EX: serge@oneway.net @c @image{Flags/taiwan} Taiwan [Oneway] @ @c @uref{ftp://ftp.oneway.com.tw/pub/mysql/, FTP} @item @c Added 20001031 @c EMAIL: ijliao@php.nctu.edu.tw (Ying-Chieh Liao) @image{Flags/taiwan} Taiwan [nctu.edu/HsinChu] @ @uref{http://mysql.nctu.edu.tw/, WWW} @end itemize @c @strong{Australia:} @c @itemize @bullet @c @item @c Added 980610 @c EMAIL: jason@dstc.edu.au (Jason Andrade) @c @image{Flags/australia} Australia [AARNet/Queensland] @ @c @uref{http://mysql.mirror.aarnet.edu.au/, WWW} @c @uref{ftp://mysql.mirror.aarnet.edu.au/, FTP} @c @item @c Added 980805. Removed 000102 'no such directory' @c EMAIL: sdd@ntccc.tas.gov.au (Scott Donovan) @c @image{Flags/australia} Australia [Tas] @ @c @uref{http://ftp.tas.gov.au/mysql, WWW} @c @uref{ftp://ftp.tas.gov.au/pub/mysql, FTP} @c @item @c Ok 980623 @c Not ok 20000919; Old site (Matt) @c EMAIL: samh@bluep.com (Sam Hadzajlic) @c @image{Flags/australia} Australia [Blue Planet/Melbourne] @ @c @uref{http://mysql.bluep.com/, WWW} @c removed because ftp was not working 990729 & 30 @c @uref{ftp://mysql.bluep.com/pub/mirror1/mysql/, FTP} @c @item @c Added 990531 @c EMAIL: gavin@itworks.com.au (Gavin Cameron) @c @image{Flags/australia} Australia [ITworks Consulting/Victoria] @ @c @uref{http://mysql.itworks.com.au, WWW} @c @item @c 980610 Only the toplevel dir! @c EMAIL: lucifer@maths.uq.edu.au (David Conran) @c @image{Flags/australia} Australia FTP @ @c @uref{ftp://ftp.sage-au.org.au/pub/database/mysql, [Sage]} @c @end itemize @strong{Africa:} @itemize @bullet @item @c EMAIL: ftp-admin@mweb.com (Warren Baker) @image{Flags/south-africa1} South-Africa [Mweb] @ @uref{http://www.mysql.mweb.co.za/, WWW} @item @c Ok 981010 @c EMAIL: oskar@is.co.za (Oskar Pearson) @image{Flags/south-africa1} South Africa [The Internet Solution/Johannesburg] @ @uref{ftp://ftp.is.co.za/linux/mysql/, FTP} @end itemize @c END_OF_MIRROR_LISTING @node Which OS, Which version, Getting MySQL, General Installation Issues @subsection Operating Systems Supported by MySQL @cindex operating systems, supported @cindex native thread support @cindex thread support @cindex process support @cindex support, for operating systems We use GNU Autoconf, so it is possible to port MySQL to all modern systems with working Posix threads and a C++ compiler. (To compile only the client code, a C++ compiler is required but not threads.) We use and develop the software ourselves primarily on Sun Solaris (Versions 2.5 - 2.7) and SuSE Linux Version 7.x. Note that for many operating systems, the native thread support works only in the latest versions. MySQL has been reported to compile successfully on the following operating system/thread package combinations: @itemize @bullet @item AIX 4.x with native threads. @xref{IBM-AIX}. @item Amiga. @item BSDI 2.x with the included MIT-pthreads package. @xref{BSDI}. @item BSDI 3.0, 3.1 and 4.x with native threads. @xref{BSDI}. @item DEC Unix 4.x with native threads. @xref{Alpha-DEC-UNIX}. @item FreeBSD 2.x with the included MIT-pthreads package. @xref{FreeBSD}. @item FreeBSD 3.x and 4.x with native threads. @xref{FreeBSD}. @item HP-UX 10.20 with the included MIT-pthreads package. @xref{HP-UX 10.20}. @item HP-UX 11.x with the native threads. @xref{HP-UX 11.x}. @item Linux 2.0+ with LinuxThreads 0.7.1+ or @code{glibc} 2.0.7+. @xref{Linux}. @item Mac OS X Server. @xref{Mac OS X}. @item NetBSD 1.3/1.4 Intel and NetBSD 1.3 Alpha (Requires GNU make). @xref{NetBSD}. @item OpenBSD > 2.5 with native therads. OpenBSD < 2.5 with the included MIT-pthreads package. @xref{OpenBSD}. @item OS/2 Warp 3, FixPack 29 and OS/2 Warp 4, FixPack 4. @xref{OS/2}. @item SGI Irix 6.x with native threads. @xref{SGI-Irix}. @item Solaris 2.5 and above with native threads on SPARC and x86. @xref{Solaris}. @item SunOS 4.x with the included MIT-pthreads package. @xref{Solaris}. @item SCO OpenServer with a recent port of the FSU Pthreads package. @xref{SCO}. @item SCO UnixWare 7.0.1. @xref{SCO Unixware}. @item Tru64 Unix @item Win95, Win98, NT, and Win2000. @xref{Windows}. @end itemize Note that not all platforms are suited equally well for running MySQL. How well a certain platform is suited for a high-load mission critical MySQL server is determined by the following factors: @itemize @bullet @item General stability of the thread library. A platform may have excellent reputation otherwise, but if the thread library is unstable in the code that is called by MySQL, even if everything else is perfect, MySQL will be only as stable as the thread library. @item The ability of the kernel and/or thread library to take advantage of @strong{SMP} on multi-processor systems. In other words, when a process creates a thread, it should be possible for that thread to run on a different CPU than the original process. @item The ability of the kernel and/or the thread library to run many threads which acquire/release a mutex over a short critical region frequently without excessive context switches. In other words, if the implementation of @code{pthread_mutex_lock()} is too anxious to yield CPU, this will hurt MySQL tremendously. If this issue is not taken care of, adding extra CPUs will actually make MySQL slower. @item General file system stability/performance. @item Ability of the file system to deal with large files at all and deal with them efficiently, if your tables are big. @item Our level of expertise here at MySQL AB with the platform. If we know a platform well, we introduce platform-specific optimizations/fixes enabled at compile time. We can also provide advice on configuring your system optimally for MySQL. @item The amount of testing of similar configurations we have done internally. @item The number of users that have successfully run MySQL on that platform in similar configurations. If this number is high, the chances of hitting some platform-specific surprise are much smaller. @end itemize Based on the above criteria, the best platforms for running MySQL at this point are x86 with SuSE Linux 7.1, 2.4 kernel and ReiserFS (or any similar Linux distribution) and Sparc with Solaris 2.7 or 2.8. FreeBSD comes third, but we really hope it will join the top club once the thread library is improved. We also hope that at some point we will be able to include all other platforms on which MySQL compiles, runs ok, but not quite with the same level of stability and performance, into the top category. This will require some effort on our part in cooperation with the developers of the OS/library components MySQL depends upon. If you are interested in making one of those components better, are in a position to influence their development, and need more detailed instructions on what MySQL needs to run better, send an e-mail to @email{internals@@lists.mysql.com}. Please note that the comparison above is not to say that one OS is better or worse than the other in general. We are talking about choosing a particular OS for a dedicated purpose - running MySQL, and compare platforms in that regard only. With this in mind, the result of this comparison would be different if we included more issues into it. And in some cases, the reason one OS is better than the other could simply be that we have put forth more effort into testing on and optimizing for that particular platform. We are just stating our observations to help you make a decision on which platform to use MySQL on in your setup. @node Which version, Installation layouts, Which OS, General Installation Issues @subsection Which MySQL Version to Use @cindex MySQL binary distribution @cindex MySQL source distribution @cindex release numbers @cindex version, choosing @cindex choosing, a MySQL version The first decision to make is whether you want to use the latest development release or the last stable release: @itemize @bullet @item Normally, if you are beginning to use MySQL for the first time or trying to port it to some system for which there is no binary distribution, we recommend going with the stable release (currently Version @value{mysql_version}. Note that all MySQL releases are checked with the MySQL benchmarks and an extensive test suite before each release. @item Otherwise, if you are running an old system and want to upgrade, but don't want to take chances with a non-seamless upgrade, you should upgrade to the latest in the same branch you are using (where only the last version number is newer than yours). We have tried to fix only fatal bugs and make small, relatively safe changes to that version. @end itemize The second decision to make is whether you want to use a source distribution or a binary distribution. In most cases you should probably use a binary distribution, if one exists for your platform, as this generally will be easier to install than a source distribution. In the following cases you probably will be better off with a source installation: @itemize @bullet @item If you want to install MySQL at some explicit location. (The standard binary distributions are ``ready to run'' at any place, but you may want to get even more flexibility). @item To be able to satisfy different user requirements, we are providing two different binary versions; One compiled with the non-transactional table handlers, (a small, fast binary), and one configured with the most important extended options like transaction-safe tables. Both versions are compiled from the same source distribution. All native @code{MySQL} clients can connect to both MySQL versions. The extended MySQL binary distribution is marked with the @code{-max} suffix and is configured with the same options as @code{mysqld-max}. @xref{mysqld-max, , @code{mysqld-max}}. If you want to use the @code{MySQL-Max} RPM, you must first install the standard @code{MySQL} RPM. @item If you want to configure @code{mysqld} with some extra features that are NOT in the standard binary distributions. Here is a list of the most common extra options that you may want to use: @itemize @bullet @item @code{--with-berkeley-db} @item @code{--with-innodb} @item @code{--with-raid} @item @code{--with-libwrap} @item @code{--with-named-z-lib (This is done for some of the binaries)} @item @code{--with-debug[=full]} @end itemize @item The default binary distribution is normally compiled with support for all characters sets and should work on a variety of processors from the same processor family. If you want a faster MySQL server you may want to recompile it with support for only the character sets you need, use a better compiler (like @code{pgcc}) or use compiler options that are better optimized for your processor. @item If you have found a bug and reported it to the MySQL development team you will probably receive a patch that you need to apply to the source distribution to get the bug fixed. @item If you want to read (and/or modify) the C and C++ code that makes up MySQL, you should get a source distribution. The source code is always the ultimate manual. Source distributions also contain more tests and examples than binary distributions. @end itemize @cindex naming, releases of MySQL @cindex releases, naming scheme The MySQL naming scheme uses release numbers that consist of three numbers and a suffix. For example, a release name like @code{mysql-3.21.17-beta} is interpreted like this: @itemize @bullet @item The first number (@code{3}) describes the file format. All Version 3 releases have the same file format. @item The second number (@code{21}) is the release level. Normally there are two to choose from. One is the release/stable branch (currently @code{23}) and the other is the development branch (currently @code{4.0}). Normally both are stable, but the development version may have quirks, missing documentation on new features, or may fail to compile on some systems. @item The third number (@code{17}) is the version number within the release level. This is incremented for each new distribution. Usually you want the latest version for the release level you have chosen. @item The suffix (@code{beta}) indicates the stability level of the release. The possible suffixes are: @itemize @minus @item @code{alpha} indicates that the release contains some large section of new code that hasn't been 100% tested. Known bugs (usually there are none) should be documented in the News section. @xref{News}. There are also new commands and extensions in most alpha releases. Active development that may involve major code changes can occur on an alpha release, but everything will be tested before doing a release. There should be no known bugs in any MySQL release. @item @code{beta} means that all new code has been tested. No major new features that could cause corruption on old code are added. There should be no known bugs. A version changes from alpha to beta when there haven't been any reported fatal bugs within an alpha version for at least a month and we don't plan to add any features that could make any old command more unreliable. @item @code{gamma} is a beta that has been around a while and seems to work fine. Only minor fixes are added. This is what many other companies call a release. @item If there is no suffix, it means that the version has been run for a while at many different sites with no reports of bugs other than platform-specific bugs. Only critical bug fixes are applied to the release. This is what we call a stable release. @end itemize @end itemize All versions of MySQL are run through our standard tests and benchmarks to ensure that they are relatively safe to use. Because the standard tests are extended over time to check for all previously found bugs, the test suite keeps getting better. @cindex releases, testing @cindex testing, of MySQL releases Note that all releases have been tested at least with: @table @asis @item An internal test suite This is part of a production system for a customer. It has many tables with hundreds of megabytes of data. @item The MySQL benchmark suite This runs a range of common queries. It is also a test to see whether the latest batch of optimizations actually made the code faster. @xref{MySQL Benchmarks}. @item The @code{crash-me} test This tries to determine what features the database supports and what its capabilities and limitations are. @xref{MySQL Benchmarks}. @end table Another test is that we use the newest MySQL version in our internal production environment, on at least one machine. We have more than 100 gigabytes of data to work with. @node Installation layouts, Many versions, Which version, General Installation Issues @subsection Installation Layouts @cindex installation layouts @cindex layout of installation @cindex directory structure, default @cindex default installation location This section describes the default layout of the directories created by installing binary and source distributions. A binary distribution is installed by unpacking it at the installation location you choose (typically @file{/usr/local/mysql}) and creates the following directories in that location: @multitable @columnfractions .3 .7 @item @strong{Directory} @tab @strong{Contents of directory} @item @file{bin} @tab Client programs and the @code{mysqld} server @item @file{data} @tab Log files, databases @item @file{include} @tab Include (header) files @item @file{lib} @tab Libraries @item @file{scripts} @tab @code{mysql_install_db} @item @file{share/mysql} @tab Error message files @item @file{sql-bench} @tab Benchmarks @end multitable A source distribution is installed after you configure and compile it. By default, the installation step installs files under @file{/usr/local}, in the following subdirectories: @multitable @columnfractions .3 .7 @item @strong{Directory} @tab @strong{Contents of directory} @item @file{bin} @tab Client programs and scripts @item @file{include/mysql} @tab Include (header) files @item @file{info} @tab Documentation in Info format @item @file{lib/mysql} @tab Libraries @item @file{libexec} @tab The @code{mysqld} server @item @file{share/mysql} @tab Error message files @item @file{sql-bench} @tab Benchmarks and @code{crash-me} test @item @file{var} @tab Databases and log files @end multitable Within an installation directory, the layout of a source installation differs from that of a binary installation in the following ways: @itemize @bullet @item The @code{mysqld} server is installed in the @file{libexec} directory rather than in the @file{bin} directory. @item The data directory is @file{var} rather than @file{data}. @item @code{mysql_install_db} is installed in the @file{/usr/local/bin} directory rather than in @file{/usr/local/mysql/scripts}. @item The header file and library directories are @file{include/mysql} and @file{lib/mysql} rather than @file{include} and @file{lib}. @end itemize You can create your own binary installation from a compiled source distribution by executing the script @file{scripts/make_binary_distribution}. @node Many versions, MySQL binaries, Installation layouts, General Installation Issues @subsection How and When Updates Are Released @cindex releases, updating @cindex updating, releases of MySQL MySQL is evolving quite rapidly here at MySQL AB and we want to share this with other MySQL users. We try to make a release when we have very useful features that others seem to have a need for. We also try to help out users who request features that are easy to implement. We take note of what our licensed users want to have, and we especially take note of what our extended e-mail supported customers want and try to help them out. No one has to download a new release. The News section will tell you if the new release has something you really want. @xref{News}. We use the following policy when updating MySQL: @itemize @bullet @item For each minor update, the last number in the version string is incremented. When there are major new features or minor incompatibilities with previous versions, the second number in the version string is incremented. When the file format changes, the first number is increased. @item Stable tested releases are meant to appear about 1-2 times a year, but if small bugs are found, a release with only bug fixes will be released. @item Working releases are meant to appear about every 1-8 weeks. @item Binary distributions for some platforms will be made by us for major releases. Other people may make binary distributions for other systems but probably less frequently. @item We usually make patches available as soon as we have located and fixed small bugs. @item For non-critical but annoying bugs, we will make patches available if they are sent to us. Otherwise we will combine many of them into a larger patch. @item If there is, by any chance, a fatal bug in a release we will make a new release as soon as possible. We would like other companies to do this, too. @end itemize The current stable release is Version 3.23; We have already moved active development to Version 4.0. Bugs will still be fixed in the stable version. We don't believe in a complete freeze, as this also leaves out bug fixes and things that ``must be done.'' ``Somewhat frozen'' means that we may add small things that ``almost surely will not affect anything that's already working.'' @node MySQL binaries, , Many versions, General Installation Issues @subsection MySQL Binaries Compiled by MySQL AB @cindex binary distributions As a service, we at MySQL AB provide a set of binary distributions of MySQL that are compiled at our site or at sites where customers kindly have given us access to their machines. These distributions are generated with @code{scripts/make_binary_distribution} and are configured with the following compilers and options: @table @asis @item SunOS 4.1.4 2 sun4c with @code{gcc} 2.7.2.1 @code{CC=gcc CXX=gcc CXXFLAGS="-O3 -felide-constructors" ./configure --prefix=/usr/local/mysql --disable-shared --with-extra-charsets=complex --enable-assembler} @item SunOS 5.5.1 (and above) sun4u with @code{egcs} 1.0.3a or 2.90.27 or gcc 2.95.2 and newer @code{CC=gcc CFLAGS="-O3" CXX=gcc CXXFLAGS="-O3 -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --with-low-memory --with-extra-charsets=complex --enable-assembler} @item SunOS 5.6 i86pc with @code{gcc} 2.8.1 @code{CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-low-memory --with-extra-charsets=complex} @item Linux 2.0.33 i386 with @code{pgcc} 2.90.29 (@code{egcs} 1.0.3a) @code{CFLAGS="-O3 -mpentium -mstack-align-double" CXX=gcc CXXFLAGS="-O3 -mpentium -mstack-align-double -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static --with-extra-charsets=complex} @item Linux 2.2.x with x686 with @code{gcc} 2.95.2 @code{CFLAGS="-O3 -mpentiumpro" CXX=gcc CXXFLAGS="-O3 -mpentiumpro -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static --disable-shared --with-extra-charset=complex} @item SCO 3.2v5.0.4 i386 with @code{gcc} 2.7-95q4 @code{CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-extra-charsets=complex} @item AIX 2 4 with @code{gcc} 2.7.2.2 @code{CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-extra-charsets=complex} @item OSF1 V4.0 564 alpha with @code{gcc} 2.8.1 @code{CC=gcc CFLAGS=-O CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-low-memory --with-extra-charsets=complex} @item Irix 6.3 IP32 with @code{gcc} 2.8.0 @code{CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-extra-charsets=complex} @item BSDI BSD/OS 3.1 i386 with @code{gcc} 2.7.2.1 @code{CC=gcc CXX=gcc CXXFLAGS=-O ./configure --prefix=/usr/local/mysql --with-extra-charsets=complex} @item BSDI BSD/OS 2.1 i386 with @code{gcc} 2.7.2 @code{CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql --with-extra-charsets=complex} @end table Anyone who has more optimal options for any of the configurations listed above can always mail them to the developer's mailing list at @email{internals@@lists.mysql.com}. RPM distributions prior to MySQL Version 3.22 are user-contributed. Beginning with Version 3.22, the RPMs are generated by us at MySQL AB. If you want to compile a debug version of MySQL, you should add @code{--with-debug} or @code{--with-debug=full} to the above configure lines and remove any @code{-fomit-frame-pointer} options. @node Installing source, Post-installation, General Installation Issues, Installing @section Installing a MySQL Source Distribution @cindex installing, source distribution @cindex source distribution, installing @cindex installation overview Before you proceed with the source installation, check first to see if our binary is available for your platform and if it will work for you. We put in a lot of effort into making sure that our binaries are built with the best possible options. You need the following tools to build and install MySQL from source: @itemize @bullet @item GNU @code{gunzip} to uncompress the distribution. @item A reasonable @code{tar} to unpack the distribution. GNU @code{tar} is known to work. Sun @code{tar} is known to have problems. @item A working ANSI C++ compiler. @code{gcc} >= 2.95.2, @code{egcs} >= 1.0.2 or @code{egcs 2.91.66}, SGI C++, and SunPro C++ are some of the compilers that are known to work. @code{libg++} is not needed when using @code{gcc}. @code{gcc} 2.7.x has a bug that makes it impossible to compile some perfectly legal C++ files, such as @file{sql/sql_base.cc}. If you only have @code{gcc} 2.7.x, you must upgrade your @code{gcc} to be able to compile MySQL. @code{gcc} 2.8.1 is also known to have problems on some platforms so it should be avoided if there exists a new compiler for the platform.. @code{gcc} >= 2.95.2 is recommended when compiling MySQL Version 3.23.x. @item A good @code{make} program. GNU @code{make} is always recommended and is sometimes required. If you have problems, we recommend trying GNU @code{make} 3.75 or newer. @end itemize If you are using a recent version of @strong{gcc}, recent enough to understand @code{-fno-exceptions} option, it is @strong{VERY IMPORTANT} that you use it. Otherwise, you may compile a binary that crashes randomly. We also recommend that you use @code{-felide-contructors} and @code{-fno-rtti} along with @code{-fno-exceptions}. When in doubt, do the following: @example CFLAGS="-O3" CXX=gcc CXXFLAGS="-O3 -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static @end example On most systems this will give you a fast and stable binary. @c texi2html fails to split chapters if I use strong for all of this. If you run into problems, @strong{PLEASE ALWAYS USE @code{mysqlbug}} when posting questions to @email{mysql@@lists.mysql.com}. Even if the problem isn't a bug, @code{mysqlbug} gathers system information that will help others solve your problem. By not using @code{mysqlbug}, you lessen the likelihood of getting a solution to your problem! You will find @code{mysqlbug} in the @file{scripts} directory after you unpack the distribution. @xref{Bug reports}. @menu * Quick install:: Quick installation overview * Applying patches:: Applying patches * configure options:: Typical @code{configure} options * Installing source tree:: Installing from the Development Source Tree * Compilation problems:: Problems Compiling? * MIT-pthreads:: MIT-pthreads Notes @end menu @node Quick install, Applying patches, Installing source, Installing source @subsection Quick Installation Overview The basic commands you must execute to install a MySQL source distribution are: @example shell> groupadd mysql shell> useradd -g mysql mysql shell> gunzip < mysql-VERSION.tar.gz | tar -xvf - shell> cd mysql-VERSION shell> ./configure --prefix=/usr/local/mysql shell> make shell> make install shell> scripts/mysql_install_db shell> chown -R root /usr/local/mysql shell> chown -R mysql /usr/local/mysql/var shell> chgrp -R mysql /usr/local/mysql shell> cp support-files/my-medium.cnf /etc/my.cnf shell> /usr/local/mysql/bin/safe_mysqld --user=mysql & @end example If you want have support for InnoDB tables, you should edit the @code{/etc/my.cnf} file and remove the @code{#} character before the parameters that starts with @code{innodb_...}. @xref{Option files}. @xref{InnoDB start}. If you start from a source RPM, then do the following: @example shell> rpm --rebuild MySQL-VERSION.src.rpm @end example This will make a binary RPM that you can install. You can add new users using the @code{bin/mysql_setpermission} script if you install the @code{DBI} and @code{Msql-Mysql-modules} Perl modules. A more detailed description follows. To install a source distribution, follow the steps below, then proceed to @ref{Post-installation}, for post-installation initialization and testing: @enumerate @item Pick the directory under which you want to unpack the distribution, and move into it. @item Obtain a distribution file from one of the sites listed in @ref{Getting MySQL, , Getting MySQL}. @item If you are interested in using Berkeley DB tables with MySQL, you will need to obtain a patched version of the Berkeley DB source code. Please read the chapter on Berkeley DB tables before proceeding. @xref{BDB}. MySQL source distributions are provided as compressed @code{tar} archives and have names like @file{mysql-VERSION.tar.gz}, where @code{VERSION} is a number like @value{mysql_version}. @item Add a user and group for @code{mysqld} to run as: @example shell> groupadd mysql shell> useradd -g mysql mysql @end example These commands add the @code{mysql} group, and the @code{mysql} user. The syntax for @code{useradd} and @code{groupadd} may differ slightly on different versions of Unix. They may also be called @code{adduser} and @code{addgroup}. You may wish to call the user and group something else instead of @code{mysql}. @item Unpack the distribution into the current directory: @example shell> gunzip < /path/to/mysql-VERSION.tar.gz | tar xvf - @end example This command creates a directory named @file{mysql-VERSION}. @item Change into the top-level directory of the unpacked distribution: @example shell> cd mysql-VERSION @end example Note that currently you must configure and build MySQL from this top-level directory. You can not build it in a different directory. @item Configure the release and compile everything: @example shell> ./configure --prefix=/usr/local/mysql shell> make @end example When you run @code{configure}, you might want to specify some options. Run @code{./configure --help} for a list of options. @ref{configure options, , @code{configure} options}, discusses some of the more useful options. If @code{configure} fails, and you are going to send mail to @email{mysql@@lists.mysql.com} to ask for assistance, please include any lines from @file{config.log} that you think can help solve the problem. Also include the last couple of lines of output from @code{configure} if @code{configure} aborts. Post the bug report using the @code{mysqlbug} script. @xref{Bug reports}. If the compile fails, see @ref{Compilation problems}, for help with a number of common problems. @item Install everything: @example shell> make install @end example You might need to run this command as @code{root}. @item Create the MySQL grant tables (necessary only if you haven't installed MySQL before): @example shell> scripts/mysql_install_db @end example Note that MySQL versions older than Version 3.22.10 started the MySQL server when you run @code{mysql_install_db}. This is no longer true! @item Change ownership of binaries to @code{root} and ownership of the data directory to the user that you will run @code{mysqld} as: @example shell> chown -R root /usr/local/mysql shell> chown -R mysql /usr/local/mysql/var shell> chgrp -R mysql /usr/local/mysql @end example The first command changes the @code{owner} attribute of the files to the @code{root} user, the second one changes the @code{owner} attribute of the data directory to the @code{mysql} user, and the third one changes the @code{group} attribute to the @code{mysql} group. @item If you want to install support for the Perl @code{DBI}/@code{DBD} interface, see @ref{Perl support}. @item If you would like MySQL to start automatically when you boot your machine, you can copy @code{support-files/mysql.server} to the location where your system has its startup files. More information can be found in the @code{support-files/mysql.server} script itself and in @ref{Automatic start}. @end enumerate After everything has been installed, you should initialize and test your distribution: @example shell> /usr/local/mysql/bin/safe_mysqld --user=mysql & @end example If that command fails immediately with @code{mysqld daemon ended} then you can find some information in the file @file{mysql-data-directory/'hostname'.err}. The likely reason is that you already have another @code{mysqld} server running. @xref{Multiple servers}. @xref{Post-installation}. @node Applying patches, configure options, Quick install, Installing source @subsection Applying Patches @cindex patches, applying @cindex applying, patches Sometimes patches appear on the mailing list or are placed in the @uref{http://www.mysql.com/Downloads/Patches, patches area} of the MySQL Web site. To apply a patch from the mailing list, save the message in which the patch appears in a file, change into the top-level directory of your MySQL source tree, and run these commands: @example shell> patch -p1 < patch-file-name shell> rm config.cache shell> make clean @end example Patches from the FTP site are distributed as plain text files or as files compressed with @code{gzip}. Apply a plain patch as shown above for mailing list patches. To apply a compressed patch, change into the top-level directory of your MySQL source tree and run these commands: @example shell> gunzip < patch-file-name.gz | patch -p1 shell> rm config.cache shell> make clean @end example After applying a patch, follow the instructions for a normal source install, beginning with the @code{./configure} step. After running the @code{make install} step, restart your MySQL server. You may need to bring down any currently running server before you run @code{make install}. (Use @code{mysqladmin shutdown} to do this.) Some systems do not allow you to install a new version of a program if it replaces the version that is currently executing. @node configure options, Installing source tree, Applying patches, Installing source @subsection Typical @code{configure} Options @findex without-server option @cindex @code{configure} script @cindex options, configure @cindex configuration options @cindex log files @cindex files, log The @code{configure} script gives you a great deal of control over how you configure your MySQL distribution. Typically you do this using options on the @code{configure} command line. You can also affect @code{configure} using certain environment variables. @xref{Environment variables}. For a list of options supported by @code{configure}, run this command: @example shell> ./configure --help @end example Some of the more commonly-used @code{configure} options are described below: @itemize @bullet @item To compile just the MySQL client libraries and client programs and not the server, use the @code{--without-server} option: @example shell> ./configure --without-server @end example @c FIX incorrect.. If you don't have a C++ compiler, @code{mysql} will not compile (it is the one client program that requires C++). In this case, you can remove the code in @code{configure} that tests for the C++ compiler and then run @code{./configure} with the @code{--without-server} option. The compile step will still try to build @code{mysql}, but you can ignore any warnings about @file{mysql.cc}. (If @code{make} stops, try @code{make -k} to tell it to continue with the rest of the build even if errors occur.) @item If you don't want your log files and database directories located under @file{/usr/local/var}, use a @code{configure} command, something like one of these: @example shell> ./configure --prefix=/usr/local/mysql shell> ./configure --prefix=/usr/local \ --localstatedir=/usr/local/mysql/data @end example The first command changes the installation prefix so that everything is installed under @file{/usr/local/mysql} rather than the default of @file{/usr/local}. The second command preserves the default installation prefix, but overrides the default location for database directories (normally @file{/usr/local/var}) and changes it to @code{/usr/local/mysql/data}. @cindex changing socket location @cindex socket location, changing @cindex mysql.sock, changing location of @item If you are using Unix and you want the MySQL socket located somewhere other than the default location (normally in the directory @file{/tmp} or @file{/var/run}) use a @code{configure} command like this: @example shell> ./configure --with-unix-socket-path=/usr/local/mysql/tmp/mysql.sock @end example Note that the given file must be an absolute pathname! You can also later change the location @file{mysql.sock} by using the MySQL option files. @xref{Problems with mysql.sock}. @cindex compiling, statically @cindex statically, compiling @item If you want to compile statically linked programs (for example, to make a binary distribution, to get more speed, or to work around problems with some RedHat Linux distributions), run @code{configure} like this: @example shell> ./configure --with-client-ldflags=-all-static \ --with-mysqld-ldflags=-all-static @end example @tindex @code{CC} environment variable @tindex environment variable, @code{CC} @tindex @code{CXX} environment variable @tindex environment variable, @code{CXX} @cindex @code{gcc} @cindex C++ compiler, @code{gcc} @cindex compiler, C++ @code{gcc} @item If you are using @code{gcc} and don't have @code{libg++} or @code{libstdc++} installed, you can tell @code{configure} to use @code{gcc} as your C++ compiler: @example shell> CC=gcc CXX=gcc ./configure @end example When you use @code{gcc} as your C++ compiler, it will not attempt to link in @code{libg++} or @code{libstdc++}. Here is some common environment variables to set depending on the compiler you are using: @tindex CXXFLAGS environment variable @tindex environment variable, CXXFLAGS @multitable @columnfractions .20 .80 @item gcc 2.7.2.1 @tab CC=gcc CXX=gcc CXXFLAGS="-O3 -felide-constructors" @item egcs 1.0.3a @tab CC=gcc CXX=gcc CXXFLAGS="-O3 -felide-constructors -fno-exceptions -fno-rtti" @item gcc 2.95.2 @tab CFLAGS="-O3 -mpentiumpro" CXX=gcc CXXFLAGS="-O3 -mpentiumpro -felide-constructors -fno-exceptions -fno-rtti" @item pgcc 2.90.29 or newer @tab CFLAGS="-O3 -mpentiumpro -mstack-align-double" CXX=gcc CXXFLAGS="-O3 -mpentiumpro -mstack-align-double -felide-constructors -fno-exceptions -fno-rtti" @end multitable In most cases you can get a reasonably optimal MySQL binary by using the options from the above and adding the following options to the configure line: @example --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static @end example The full configure line would in other words be something like the following for all recent gcc versions: @example CFLAGS="-O3 -mpentiumpro" CXX=gcc CXXFLAGS="-O3 -mpentiumpro -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static @end example The binaries we provide on the MySQL Web site at @uref{http://www.mysql.com} are all compiled with full optimization and should be perfect for most users. @xref{MySQL binaries}. There are some things you can tweak to make an even faster binary, but this is only for advanced users. @xref{Compile and link options}. If the build fails and produces errors about your compiler or linker not being able to create the shared library @file{libmysqlclient.so.#} (@samp{#} is a version number), you can work around this problem by giving the @code{--disable-shared} option to @code{configure}. In this case, @code{configure} will not build a shared @code{libmysqlclient.so.#} library. @cindex default values, suppression @cindex suppression, default values @item You can configure MySQL not to use @code{DEFAULT} column values for non-@code{NULL} columns (that is, columns that are not allowed to be @code{NULL}). This causes @code{INSERT} statements to generate an error unless you explicitly specify values for all columns that require a non-@code{NULL} value. To suppress use of default values, run @code{configure} like this: @tindex CXXFLAGS environment variable @tindex environment variable, CXXFLAGS @example shell> CXXFLAGS=-DDONT_USE_DEFAULT_FIELDS ./configure @end example @cindex character sets @findex configure option, --with-charset @findex configure option, --with-extra-charset @item By default, MySQL uses the ISO-8859-1 (Latin1) character set. To change the default set, use the @code{--with-charset} option: @example shell> ./configure --with-charset=CHARSET @end example @code{CHARSET} may be one of @code{big5}, @code{cp1251}, @code{cp1257}, @code{czech}, @code{danish}, @code{dec8}, @code{dos}, @code{euc_kr}, @code{gb2312}, @code{gbk}, @code{german1}, @code{hebrew}, @code{hp8}, @code{hungarian}, @code{koi8_ru}, @code{koi8_ukr}, @code{latin1}, @code{latin2}, @code{sjis}, @code{swe7}, @code{tis620}, @code{ujis}, @code{usa7}, or @code{win1251ukr}. @xref{Character sets}. If you want to convert characters between the server and the client, you should take a look at the @code{SET OPTION CHARACTER SET} command. @xref{SET OPTION, , @code{SET OPTION}}. @cindex @code{myisamchk} @strong{Warning:} If you change character sets after having created any tables, you will have to run @code{myisamchk -r -q} on every table. Your indexes may be sorted incorrectly otherwise. (This can happen if you install MySQL, create some tables, then reconfigure MySQL to use a different character set and reinstall it.) With the option @code{--with-extra-charset=LIST} you can define which additional character sets should be incompiled in the server. Here @code{LIST} is either a list of character set separated with space, @code{complex} to include all characters that can't be dynamically loaded or @code{all} to include all character sets into the binaries. @item To configure MySQL with debugging code, use the @code{--with-debug} option: @example shell> ./configure --with-debug @end example This causes a safe memory allocator to be included that can find some errors and that provides output about what is happening. @xref{Debugging server}. @item If your client programs are using threads, you need to also compile a thread-safe version of the MySQL client library with the @code{--enable-thread-safe-client} configure options. This will create a @code{libmysqlclient_r} library with which you should link your threaded applications. @xref{Threaded clients}. @item Options that pertain to particular systems can be found in the system-specific section of this manual. @xref{Operating System Specific Notes}. @end itemize @node Installing source tree, Compilation problems, configure options, Installing source @subsection Installing from the Development Source Tree @cindex development source tree @cindex BitKeeper tree @cindex cvs tree @strong{CAUTION:} You should read this section only if you are interested in helping us test our new code. If you just want to get MySQL up and running on your system, you should use a standard release distribution (either a source or binary distribution will do). To obtain our most recent development source tree, use these instructions: @enumerate @item Download @strong{BitKeeper} from @uref{http://www.bitmover.com/cgi-bin/download.cgi}. You will need @strong{Bitkeeper} 2.0 or newer to access our repository. @item Follow the instructions to install it. @item After @strong{BitKeeper} is installed, use this command if you want to clone the MySQL 3.23 branch: @example shell> bk clone bk://work.mysql.com:7000 mysql @end example To clone the 4.0 branch, use this command instead: @example shell> bk clone bk://work.mysql.com:7001 mysql-4.0 @end example The initial download of the source tree may take a while, depending on the speed of your connection; be patient. @item You will need GNU @code{autoconf}, @code{automake}, @code{libtool}, and @code{m4} to run the next set of commands. If you get some strange error during this stage, check that you really have @code{libtool} installed! @example shell> cd mysql shell> bk -r edit shell> aclocal; autoheader; autoconf; automake; shell> ./configure # Add your favorite options here shell> make @end example A collection of our standard configure scripts is located in the @file{BUILD/} subdirectory. If you are lazy, you can use @file{BUILD/compile-pentium-debug}. To compile on a different architecture, modify the script removing flags that are Pentium-specific. @item When the build is done, run @code{make install}. Be careful with this on a production machine; the command may overwrite your live release installation. If you have another installation of MySQL, we recommand that you run @code{./configure} with different values for the @code{prefix}, @code{tcp-port}, and @code{unix-socket-path} options than those used for your production server. @item Play hard with your new installation and try to make the new features crash. Start by running @code{make test}. @xref{MySQL test suite}. @item If you have gotten to the @code{make} stage and the distribution does not compile, please report it to @email{bugs@@lists.mysql.com}. If you have installed the latest versions of the required GNU tools, and they crash trying to process our configuration files, please report that also. However, if you execute @code{aclocal} and get a @code{command not found} error or a similar problem, do not report it. Instead, make sure all the necessary tools are installed and that your @code{PATH} variable is set correctly so your shell can find them. @item After the initial @code{bk clone} operation to get the source tree, you should run @code{bk pull} periodically to get the updates. @item You can examine the change history for the tree with all the diffs by using @code{bk sccstool}. If you see some funny diffs or code that you have a question about, do not hesitate to send e-mail to @email{internals@@lists.mysql.com}. Also, if you think you have a better idea on how to do something, send an email to the same address with a patch. @code{bk diffs} will produce a patch for you after you have made changes to the source. If you do not have the time to code your idea, just send a description. @item @strong{BitKeeper} has a nice help utility that you can access via @code{bk helptool}. @end enumerate @node Compilation problems, MIT-pthreads, Installing source tree, Installing source @subsection Problems Compiling? @cindex compiling, problems @cindex problems, compiling @cindex reconfiguring @cindex @code{config.cache} file @cindex files, @code{config.cache} All MySQL programs compile cleanly for us with no warnings on Solaris using @code{gcc}. On other systems, warnings may occur due to differences in system include files. See @ref{MIT-pthreads} for warnings that may occur when using MIT-pthreads. For other problems, check the list below. The solution to many problems involves reconfiguring. If you do need to reconfigure, take note of the following: @cindex running @code{configure} after prior invocation @cindex @code{configure}, running after prior invocation @cindex reconfiguring @cindex config.cache @itemize @bullet @item If @code{configure} is run after it already has been run, it may use information that was gathered during its previous invocation. This information is stored in @file{config.cache}. When @code{configure} starts up, it looks for that file and reads its contents if it exists, on the assumption that the information is still correct. That assumption is invalid when you reconfigure. @item Each time you run @code{configure}, you must run @code{make} again to recompile. However, you may want to remove old object files from previous builds first, because they were compiled using different configuration options. @end itemize To prevent old configuration information or object files from being used, run these commands before rerunning @code{configure}: @example shell> rm config.cache shell> make clean @end example Alternatively, you can run @code{make distclean}. The list below describes some of the problems compiling MySQL that have been found to occur most often: @itemize @bullet @item @cindex @code{cc1plus} problems @cindex @code{fatal signal 11} @cindex @code{sql_yacc.cc} problems @cindex internal compiler errors @cindex virtual memory, problems while compiling @cindex configure option, --with-low-memory If you get errors when compiling @file{sql_yacc.cc}, such as the ones shown below, you have probably run out of memory or swap space: @example Internal compiler error: program cc1plus got fatal signal 11 or Out of virtual memory or Virtual memory exhausted @end example The problem is that @code{gcc} requires huge amounts of memory to compile @file{sql_yacc.cc} with inline functions. Try running @code{configure} with the @code{--with-low-memory} option: @example shell> ./configure --with-low-memory @end example This option causes @code{-fno-inline} to be added to the compile line if you are using @code{gcc} and @code{-O0} if you are using something else. You should try the @code{--with-low-memory} option even if you have so much memory and swap space that you think you can't possibly have run out. This problem has been observed to occur even on systems with generous hardware configurations, and the @code{--with-low-memory} option usually fixes it. @item By default, @code{configure} picks @code{c++} as the compiler name and GNU @code{c++} links with @code{-lg++}. If you are using @code{gcc}, that behavior can cause problems during configuration such as this: @cindex C++ compiler cannot create executables @example configure: error: installation or configuration problem: C++ compiler cannot create executables. @end example @tindex CXX environment variable @tindex Environment variable, CXX You might also observe problems during compilation related to @code{g++}, @code{libg++}, or @code{libstdc++}. One cause of these problems is that you may not have @code{g++}, or you may have @code{g++} but not @code{libg++}, or @code{libstdc++}. Take a look at the @file{config.log} file. It should contain the exact reason why your c++ compiler didn't work! To work around these problems, you can use @code{gcc} as your C++ compiler. Try setting the environment variable @code{CXX} to @code{"gcc -O3"}. For example: @tindex CXX environment variable @tindex Environment variables, CXX @example shell> CXX="gcc -O3" ./configure @end example This works because @code{gcc} compiles C++ sources as well as @code{g++} does, but does not link in @code{libg++} or @code{libstdc++} by default. Another way to fix these problems, of course, is to install @code{g++}, @code{libg++} and @code{libstdc++}. @item If your compile fails with errors, such as any of the following, you must upgrade your version of @code{make} to GNU @code{make}: @example making all in mit-pthreads make: Fatal error in reader: Makefile, line 18: Badly formed macro assignment or make: file `Makefile' line 18: Must be a separator (: or pthread.h: No such file or directory @end example @cindex Solaris troubleshooting @cindex FreeBSD troubleshooting @cindex troubleshooting, Solaris @cindex troubleshooting, FreeBSD Solaris and FreeBSD are known to have troublesome @code{make} programs. GNU @code{make} Version 3.75 is known to work. @tindex CC environment variable @tindex environment variable, CC @tindex CXX environment variable @tindex environment variable, CXX @tindex CFLAGS environment variable @tindex environment variable, CFLAGS @tindex CXXFLAGS environment variable @tindex environment variable, CXXFLAGS @item If you want to define flags to be used by your C or C++ compilers, do so by adding the flags to the @code{CFLAGS} and @code{CXXFLAGS} environment variables. You can also specify the compiler names this way using @code{CC} and @code{CXX}. For example: @example shell> CC=gcc shell> CFLAGS=-O3 shell> CXX=gcc shell> CXXFLAGS=-O3 shell> export CC CFLAGS CXX CXXFLAGS @end example See @ref{MySQL binaries}, for a list of flag definitions that have been found to be useful on various systems. @item If you get an error message like this, you need to upgrade your @code{gcc} compiler: @example client/libmysql.c:273: parse error before `__attribute__' @end example @code{gcc} 2.8.1 is known to work, but we recommend using @code{gcc} 2.95.2 or @code{egcs} 1.0.3a instead. @item If you get errors such as those shown below when compiling @code{mysqld}, @code{configure} didn't correctly detect the type of the last argument to @code{accept()}, @code{getsockname()}, or @code{getpeername()}: @example cxx: Error: mysqld.cc, line 645: In this statement, the referenced type of the pointer value "&length" is "unsigned long", which is not compatible with "int". new_sock = accept(sock, (struct sockaddr *)&cAddr, &length); @end example To fix this, edit the @file{config.h} file (which is generated by @code{configure}). Look for these lines: @example /* Define as the base type of the last arg to accept */ #define SOCKET_SIZE_TYPE XXX @end example Change @code{XXX} to @code{size_t} or @code{int}, depending on your operating system. (Note that you will have to do this each time you run @code{configure}, because @code{configure} regenerates @file{config.h}.) @item The @file{sql_yacc.cc} file is generated from @file{sql_yacc.yy}. Normally the build process doesn't need to create @file{sql_yacc.cc}, because MySQL comes with an already-generated copy. However, if you do need to re-create it, you might encounter this error: @example "sql_yacc.yy", line xxx fatal: default action causes potential... @end example This is a sign that your version of @code{yacc} is deficient. You probably need to install @code{bison} (the GNU version of @code{yacc}) and use that instead. @item If you need to debug @code{mysqld} or a MySQL client, run @code{configure} with the @code{--with-debug} option, then recompile and link your clients with the new client library. @xref{Debugging client}. @end itemize @node MIT-pthreads, , Compilation problems, Installing source @subsection MIT-pthreads Notes @cindex MIT-pthreads @cindex thread support, non-native This section describes some of the issues involved in using MIT-pthreads. Note that on Linux you should NOT use MIT-pthreads but install LinuxThreads! @xref{Linux}. If your system does not provide native thread support, you will need to build MySQL using the MIT-pthreads package. This includes older FreeBSD systems, SunOS 4.x, Solaris 2.4 and earlier, and some others. @xref{Which OS}. @itemize @bullet @item On most systems, you can force MIT-pthreads to be used by running @code{configure} with the @code{--with-mit-threads} option: @example shell> ./configure --with-mit-threads @end example Building in a non-source directory is not supported when using MIT-pthreads, because we want to minimize our changes to this code. @item The checks that determine whether or not to use MIT-pthreads occur only during the part of the configuration process that deals with the server code. If you have configured the distribution using @code{--without-server} to build only the client code, clients will not know whether or not MIT-pthreads is being used and will use Unix socket connections by default. Because Unix sockets do not work under MIT-pthreads, this means you will need to use @code{-h} or @code{--host} when you run client programs. @item When MySQL is compiled using MIT-pthreads, system locking is disabled by default for performance reasons. You can tell the server to use system locking with the @code{--use-locking} option. @item Sometimes the pthread @code{bind()} command fails to bind to a socket without any error message (at least on Solaris). The result is that all connections to the server fail. For example: @example shell> mysqladmin version mysqladmin: connect to server at '' failed; error: 'Can't connect to mysql server on localhost (146)' @end example The solution to this is to kill the @code{mysqld} server and restart it. This has only happened to us when we have forced the server down and done a restart immediately. @item With MIT-pthreads, the @code{sleep()} system call isn't interruptible with @code{SIGINT} (break). This is only noticeable when you run @code{mysqladmin --sleep}. You must wait for the @code{sleep()} call to terminate before the interrupt is served and the process stops. @item When linking, you may receive warning messages like these (at least on Solaris); they can be ignored: @example ld: warning: symbol `_iob' has differing sizes: (file /my/local/pthreads/lib/libpthread.a(findfp.o) value=0x4; file /usr/lib/libc.so value=0x140); /my/local/pthreads/lib/libpthread.a(findfp.o) definition taken ld: warning: symbol `__iob' has differing sizes: (file /my/local/pthreads/lib/libpthread.a(findfp.o) value=0x4; file /usr/lib/libc.so value=0x140); /my/local/pthreads/lib/libpthread.a(findfp.o) definition taken @end example @item Some other warnings also can be ignored: @example implicit declaration of function `int strtoll(...)' implicit declaration of function `int strtoul(...)' @end example @item We haven't gotten @code{readline} to work with MIT-pthreads. (This isn't needed, but may be interesting for someone.) @end itemize @node Post-installation, Upgrade, Installing source, Installing @section Post-installation Setup and Testing @cindex post-installation, setup and testing @cindex testing, post-installation @cindex setup, post-installation @menu * mysql_install_db:: Problems running @code{mysql_install_db} * Starting server:: Problems starting the MySQL server * Automatic start:: Starting and stopping MySQL automatically @end menu Once you've installed MySQL (from either a binary or source distribution), you need to initialize the grant tables, start the server, and make sure that the server works okay. You may also wish to arrange for the server to be started and stopped automatically when your system starts up and shuts down. Normally you install the grant tables and start the server like this for installation from a source distribution: @cindex starting, the server @cindex server, starting @example shell> ./scripts/mysql_install_db shell> cd mysql_installation_directory shell> ./bin/safe_mysqld --user=mysql & @end example For a binary distribution (not RPM or pkg packages), do this: @example shell> cd mysql_installation_directory shell> ./bin/mysql_install_db shell> ./bin/safe_mysqld --user=mysql & @end example This creates the @code{mysql} database which will hold all database privileges, the @code{test} database which you can use to test MySQL and also privilege entries for the user that run @code{mysql_install_db} and a @code{root} user (without any passwords). This also starts the @code{mysqld} server. @code{mysql_install_db} will not overwrite any old privilege tables, so it should be safe to run in any circumstances. If you don't want to have the @code{test} database you can remove it with @code{mysqladmin -u root drop test}. Testing is most easily done from the top-level directory of the MySQL distribution. For a binary distribution, this is your installation directory (typically something like @file{/usr/local/mysql}). For a source distribution, this is the main directory of your MySQL source tree. @cindex testing, the server In the commands shown below in this section and in the following subsections, @code{BINDIR} is the path to the location in which programs like @code{mysqladmin} and @code{safe_mysqld} are installed. For a binary distribution, this is the @file{bin} directory within the distribution. For a source distribution, @code{BINDIR} is probably @file{/usr/local/bin}, unless you specified an installation directory other than @file{/usr/local} when you ran @code{configure}. @code{EXECDIR} is the location in which the @code{mysqld} server is installed. For a binary distribution, this is the same as @code{BINDIR}. For a source distribution, @code{EXECDIR} is probably @file{/usr/local/libexec}. Testing is described in detail below: @cindex testing, installation @enumerate @item If necessary, start the @code{mysqld} server and set up the initial MySQL grant tables containing the privileges that determine how users are allowed to connect to the server. This is normally done with the @code{mysql_install_db} script: @example shell> scripts/mysql_install_db @end example Typically, @code{mysql_install_db} needs to be run only the first time you install MySQL. Therefore, if you are upgrading an existing installation, you can skip this step. (However, @code{mysql_install_db} is quite safe to use and will not update any tables that already exist, so if you are unsure of what to do, you can always run @code{mysql_install_db}.) @code{mysql_install_db} creates six tables (@code{user}, @code{db}, @code{host}, @code{tables_priv}, @code{columns_priv}, and @code{func}) in the @code{mysql} database. A description of the initial privileges is given in @ref{Default privileges}. Briefly, these privileges allow the MySQL @code{root} user to do anything, and allow anybody to create or use databases with a name of @code{'test'} or starting with @code{'test_'}. If you don't set up the grant tables, the following error will appear in the log file when you start the server: @tindex host.frm, problems finding @example mysqld: Can't find file: 'host.frm' @end example The above may also happen with a binary MySQL distribution if you don't start MySQL by executing exactly @code{./bin/safe_mysqld}! @xref{safe_mysqld, , @code{safe_mysqld}}. You might need to run @code{mysql_install_db} as @code{root}. However, if you prefer, you can run the MySQL server as an unprivileged (non-@code{root}) user, provided that user can read and write files in the database directory. Instructions for running MySQL as an unprivileged user are given in @ref{Changing MySQL user, , Changing MySQL user}. If you have problems with @code{mysql_install_db}, see @ref{mysql_install_db, , @code{mysql_install_db}}. There are some alternatives to running the @code{mysql_install_db} script as it is provided in the MySQL distribution: @itemize @bullet @item You may want to edit @code{mysql_install_db} before running it, to change the initial privileges that are installed into the grant tables. This is useful if you want to install MySQL on a lot of machines with the same privileges. In this case you probably should need only to add a few extra @code{INSERT} statements to the @code{mysql.user} and @code{mysql.db} tables! @item If you want to change things in the grant tables after installing them, you can run @code{mysql_install_db}, then use @code{mysql -u root mysql} to connect to the grant tables as the MySQL @code{root} user and issue SQL statements to modify the grant tables directly. @item It is possible to re-create the grant tables completely after they have already been created. You might want to do this if you've already installed the tables but then want to re-create them after editing @code{mysql_install_db}. @end itemize For more information about these alternatives, see @ref{Default privileges}. @item Start the MySQL server like this: @example shell> cd mysql_installation_directory shell> bin/safe_mysqld & @end example If you have problems starting the server, see @ref{Starting server}. @item Use @code{mysqladmin} to verify that the server is running. The following commands provide a simple test to check that the server is up and responding to connections: @example shell> BINDIR/mysqladmin version shell> BINDIR/mysqladmin variables @end example The output from @code{mysqladmin version} varies slightly depending on your platform and version of MySQL, but should be similar to that shown below: @example shell> BINDIR/mysqladmin version mysqladmin Ver 8.14 Distrib 3.23.32, for linux on i586 Copyright (C) 2000 MySQL AB & MySQL Finland AB & TCX DataKonsult AB This software comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to modify and redistribute it under the GPL license Server version 3.23.32-debug Protocol version 10 Connection Localhost via Unix socket TCP port 3306 UNIX socket /tmp/mysql.sock Uptime: 16 sec Threads: 1 Questions: 9 Slow queries: 0 Opens: 7 Flush tables: 2 Open tables: 0 Queries per second avg: 0.000 Memory in use: 132K Max memory used: 16773K @end example To get a feeling for what else you can do with @code{BINDIR/mysqladmin}, invoke it with the @code{--help} option. @item Verify that you can shut down the server: @cindex server, shutdown @cindex shutting down, the server @example shell> BINDIR/mysqladmin -u root shutdown @end example @item Verify that you can restart the server. Do this using @code{safe_mysqld} or by invoking @code{mysqld} directly. For example: @cindex server, restart @cindex restarting, the server @example shell> BINDIR/safe_mysqld --log & @end example If @code{safe_mysqld} fails, try running it from the MySQL installation directory (if you are not already there). If that doesn't work, see @ref{Starting server}. @item Run some simple tests to verify that the server is working. The output should be similar to what is shown below: @example shell> BINDIR/mysqlshow +-----------+ | Databases | +-----------+ | mysql | +-----------+ shell> BINDIR/mysqlshow mysql Database: mysql +--------------+ | Tables | +--------------+ | columns_priv | | db | | func | | host | | tables_priv | | user | +--------------+ shell> BINDIR/mysql -e "select host,db,user from db" mysql +------+--------+------+ | host | db | user | +------+--------+------+ | % | test | | | % | test_% | | +------+--------+------+ @end example There is also a benchmark suite in the @file{sql-bench} directory (under the MySQL installation directory) that you can use to compare how MySQL performs on different platforms. The @file{sql-bench/Results} directory contains the results from many runs against different databases and platforms. To run all tests, execute these commands: @example shell> cd sql-bench shell> run-all-tests @end example If you don't have the @file{sql-bench} directory, you are probably using an RPM for a binary distribution. (Source distribution RPMs include the benchmark directory.) In this case, you must first install the benchmark suite before you can use it. Beginning with MySQL Version 3.22, there are benchmark RPM files named @file{mysql-bench-VERSION-i386.rpm} that contain benchmark code and data. If you have a source distribution, you can also run the tests in the @file{tests} subdirectory. For example, to run @file{auto_increment.tst}, do this: @example shell> BINDIR/mysql -vvf test < ./tests/auto_increment.tst @end example The expected results are shown in the @file{./tests/auto_increment.res} file. @end enumerate @node mysql_install_db, Starting server, Post-installation, Post-installation @subsection Problems Running @code{mysql_install_db} @cindex @code{mysql_install_db} script @cindex scripts, @code{mysql_install_db} The purpose of the @code{mysql_install_db} script is to generate new MySQL privilege tables. It will not affect any other data! It will also not do anything if you already have MySQL privilege tables installed! If you want to re-create your privilege tables, you should take down the @code{mysqld} server, if it's running, and then do something like: @example mv mysql-data-directory/mysql mysql-data-directory/mysql-old mysql_install_db @end example This section lists problems you might encounter when you run @code{mysql_install_db}: @table @strong @item @code{mysql_install_db} doesn't install the grant tables You may find that @code{mysql_install_db} fails to install the grant tables and terminates after displaying the following messages: @example starting mysqld daemon with databases from XXXXXX mysql daemon ended @end example In this case, you should examine the log file very carefully! The log should be located in the directory @file{XXXXXX} named by the error message, and should indicate why @code{mysqld} didn't start. If you don't understand what happened, include the log when you post a bug report using @code{mysqlbug}! @xref{Bug reports}. @item There is already a @code{mysqld} daemon running In this case, you probably don't have to run @code{mysql_install_db} at all. You have to run @code{mysql_install_db} only once, when you install MySQL the first time. @item Installing a second @code{mysqld} daemon doesn't work when one daemon is running This can happen when you already have an existing MySQL installation, but want to put a new installation in a different place (for example, for testing, or perhaps you simply want to run two installations at the same time). Generally the problem that occurs when you try to run the second server is that it tries to use the same socket and port as the old one. In this case you will get the error message: @code{Can't start server: Bind on TCP/IP port: Address already in use} or @code{Can't start server : Bind on unix socket...}. @xref{Installing many servers}. @item You don't have write access to @file{/tmp} @cindex write access, tmp @cindex temporary file, write access @cindex files, @code{tmp} If you don't have write access to create a socket file at the default place (in @file{/tmp}) or permission to create temporary files in @file{/tmp,} you will get an error when running @code{mysql_install_db} or when starting or using @code{mysqld}. You can specify a different socket and temporary directory as follows: @tindex TMPDIR environment variable @tindex MYSQL_UNIX_PORT environment variable @tindex Environment variable, TMPDIR @tindex Environment variable, MYSQL_UNIX_PORT @example shell> TMPDIR=/some_tmp_dir/ shell> MYSQL_UNIX_PORT=/some_tmp_dir/mysqld.sock shell> export TMPDIR MYSQL_UNIX_PORT @end example @xref{Problems with mysql.sock}. @file{some_tmp_dir} should be the path to some directory for which you have write permission. @xref{Environment variables}. After this you should be able to run @code{mysql_install_db} and start the server with these commands: @example shell> scripts/mysql_install_db shell> BINDIR/safe_mysqld & @end example @item @code{mysqld} crashes immediately If you are running RedHat Version 5.0 with a version of @code{glibc} older than 2.0.7-5, you should make sure you have installed all @code{glibc} patches! There is a lot of information about this in the MySQL mail archives. Links to the mail archives are available online at @uref{http://www.mysql.com/documentation/}. Also, see @ref{Linux}. You can also start @code{mysqld} manually using the @code{--skip-grant-tables} option and add the privilege information yourself using @code{mysql}: @example shell> BINDIR/safe_mysqld --skip-grant-tables & shell> BINDIR/mysql -u root mysql @end example From @code{mysql}, manually execute the SQL commands in @code{mysql_install_db}. Make sure you run @code{mysqladmin flush-privileges} or @code{mysqladmin reload} afterward to tell the server to reload the grant tables. @end table @node Starting server, Automatic start, mysql_install_db, Post-installation @subsection Problems Starting the MySQL Server @cindex server, starting problems @cindex problems, starting the server If you are going to use tables that support transactions (BDB, InnoDB), you should first create a my.cnf file and set startup options for the table types you plan to use. @xref{Table types}. Generally, you start the @code{mysqld} server in one of three ways: @itemize @bullet @item By invoking @code{mysql.server}. This script is used primarily at system startup and shutdown, and is described more fully in @ref{Automatic start}. @item By invoking @code{safe_mysqld}, which tries to determine the proper options for @code{mysqld} and then runs it with those options. @xref{safe_mysqld, , @code{safe_mysqld}}. @item On NT you should install @code{mysqld} as a service as follows: @example bin\mysqld-nt --install # Install MySQL as a service @end example You can now start/stop @code{mysqld} as follows: @example NET START mysql NET STOP mysql @end example Note that in this case you can't use any other options for @code{mysqld}! You can remove the service as follows: @example bin\mysqld-nt --remove # remove MySQL as a service @end example @item By invoking @code{mysqld} directly. @end itemize When the @code{mysqld} daemon starts up, it changes directory to the data directory. This is where it expects to write log files and the pid (process ID) file, and where it expects to find databases. The data directory location is hardwired in when the distribution is compiled. However, if @code{mysqld} expects to find the data directory somewhere other than where it really is on your system, it will not work properly. If you have problems with incorrect paths, you can find out what options @code{mysqld} allows and what the default path settings are by invoking @code{mysqld} with the @code{--help} option. You can override the defaults by specifying the correct pathnames as command-line arguments to @code{mysqld}. (These options can be used with @code{safe_mysqld} as well.) Normally you should need to tell @code{mysqld} only the base directory under which MySQL is installed. You can do this with the @code{--basedir} option. You can also use @code{--help} to check the effect of changing path options (note that @code{--help} @emph{must} be the final option of the @code{mysqld} command). For example: @example shell> EXECDIR/mysqld --basedir=/usr/local --help @end example Once you determine the path settings you want, start the server without the @code{--help} option. Whichever method you use to start the server, if it fails to start up correctly, check the log file to see if you can find out why. Log files are located in the data directory (typically @file{/usr/local/mysql/data} for a binary distribution, @file{/usr/local/var} for a source distribution, @file{\mysql\data\mysql.err} on Windows.) Look in the data directory for files with names of the form @file{host_name.err} and @file{host_name.log} where @code{host_name} is the name of your server host. Then check the last few lines of these files: @example shell> tail host_name.err shell> tail host_name.log @end example If you find something like the following in the log file: @example 000729 14:50:10 bdb: Recovery function for LSN 1 27595 failed 000729 14:50:10 bdb: warning: ./test/t1.db: No such file or directory 000729 14:50:10 Can't init databases @end example This means that you didn't start @code{mysqld} with @code{--bdb-no-recover} and Berkeley DB found something wrong with its log files when it tried to recover your databases. To be able to continue, you should move away the old Berkeley DB log file from the database directory to some other place, where you can later examine these. The log files are named @file{log.0000000001}, where the number will increase over time. If you are running @code{mysqld} with BDB table support and @code{mysqld} core dumps at start this could be because of some problems with the BDB recover log. In this case you can try starting @code{mysqld} with @code{--bdb-no-recover}. If this helps, then you should remove all @file{log.*} files from the data directory and try starting @code{mysqld} again. If you get the following error, it means that some other program (or another @code{mysqld} server) is already using the TCP/IP port or socket @code{mysqld} is trying to use: @example Can't start server: Bind on TCP/IP port: Address already in use or Can't start server : Bind on unix socket... @end example Use @code{ps} to make sure that you don't have another @code{mysqld} server running. If you can't find another server running, you can try to execute the command @code{telnet your-host-name tcp-ip-port-number} and press @code{RETURN} a couple of times. If you don't get an error message like @code{telnet: Unable to connect to remote host: Connection refused}, something is using the TCP/IP port @code{mysqld} is trying to use. See @ref{mysql_install_db} and @ref{Multiple servers}. If @code{mysqld} is currently running, you can find out what path settings it is using by executing this command: @example shell> mysqladmin variables @end example or @example shell> mysqladmin -h 'your-host-name' variables @end example If @code{safe_mysqld} starts the server but you can't connect to it, you should make sure you have an entry in @file{/etc/hosts} that looks like this: @example 127.0.0.1 localhost @end example This problem occurs only on systems that don't have a working thread library and for which MySQL must be configured to use MIT-pthreads. If you can't get @code{mysqld} to start you can try to make a trace file to find the problem. @xref{Making trace files}. If you are using InnoDB tables, refer to the InnoDB-specific startup options. @xref{InnoDB start}. If you are using BDB (Berkeley DB) tables, you should familiarize yourself with the different BDB specific startup options. @xref{BDB start}. @node Automatic start, , Starting server, Post-installation @subsection Starting and Stopping MySQL Automatically @cindex starting, the server automatically @cindex stopping, the server @cindex server, starting and stopping The @code{mysql.server} and @code{safe_mysqld} scripts can be used to start the server automatically at system startup time. @code{mysql.server} can also be used to stop the server. The @code{mysql.server} script can be used to start or stop the server by invoking it with @code{start} or @code{stop} arguments: @example shell> mysql.server start shell> mysql.server stop @end example @code{mysql.server} can be found in the @file{share/mysql} directory under the MySQL installation directory or in the @file{support-files} directory of the MySQL source tree. Before @code{mysql.server} starts the server, it changes directory to the MySQL installation directory, then invokes @code{safe_mysqld}. You might need to edit @code{mysql.server} if you have a binary distribution that you've installed in a non-standard location. Modify it to @code{cd} into the proper directory before it runs @code{safe_mysqld}. If you want the server to run as some specific user, add an appropriate @code{user} line to the @file{/etc/my.cnf} file, as shown later in this section. @code{mysql.server stop} brings down the server by sending a signal to it. You can take down the server manually by executing @code{mysqladmin shutdown}. You might want to add these start and stop commands to the appropriate places in your @file{/etc/rc*} files when you start using MySQL for production applications. Note that if you modify @code{mysql.server}, then upgrade MySQL sometime, your modified version will be overwritten, so you should make a copy of your edited version that you can reinstall. If your system uses @file{/etc/rc.local} to start external scripts, you should append the following to it: @example /bin/sh -c 'cd /usr/local/mysql ; ./bin/safe_mysqld --user=mysql &' @end example @cindex changing socket location You can also add options for @code{mysql.server} in a global @file{/etc/my.cnf} file. A typical @file{/etc/my.cnf} file might look like this: @example [mysqld] datadir=/usr/local/mysql/var socket=/var/tmp/mysql.sock port=3306 user=mysql [mysql.server] basedir=/usr/local/mysql @end example The @code{mysql.server} script understands the following options: @code{datadir}, @code{basedir}, and @code{pid-file}. The following table shows which option groups each of the startup scripts read from option files: @multitable @columnfractions .20 .80 @item @strong{Script} @tab @strong{Option groups} @item @code{mysqld} @tab @code{mysqld} and @code{server} @item @code{mysql.server} @tab @code{mysql.server}, @code{mysqld}, and @code{server} @item @code{safe_mysqld} @tab @code{mysql.server}, @code{mysqld}, and @code{server} @end multitable @xref{Option files}. @node Upgrade, Operating System Specific Notes, Post-installation, Installing @section Upgrading/Downgrading MySQL @cindex upgrading @cindex downgrading You can always move the MySQL form and data files between different versions on the same architecture as long as you have the same base version of MySQL. The current base version is 3. If you change the character set when running MySQL (which may also change the sort order), you must run @code{myisamchk -r -q} on all tables. Otherwise your indexes may not be ordered correctly. If you are afraid of new versions, you can always rename your old @code{mysqld} to something like @code{mysqld}-'old-version-number'. If your new @code{mysqld} then does something unexpected, you can simply shut it down and restart with your old @code{mysqld}! When you do an upgrade you should also back up your old databases, of course. If after an upgrade, you experience problems with recompiled client programs, like @code{Commands out of sync} or unexpected core dumps, you probably have used an old header or library file when compiling your programs. In this case you should check the date for your @file{mysql.h} file and @file{libmysqlclient.a} library to verify that they are from the new MySQL distribution. If not, please recompile your programs! If you get some problems that the new @code{mysqld} server doesn't want to start or that you can't connect without a password, check that you don't have some old @file{my.cnf} file from your old installation! You can check this with: @code{program-name --print-defaults}. If this outputs anything other than the program name, you have an active @code{my.cnf} file that will affect things! It is a good idea to rebuild and reinstall the @code{Msql-Mysql-modules} distribution whenever you install a new release of MySQL, particularly if you notice symptoms such as all your @code{DBI} scripts dumping core after you upgrade MySQL. @menu * Upgrading-from-3.22:: Upgrading from a 3.22 version to 3.23 * Upgrading-from-3.21:: Upgrading from a 3.21 version to 3.22 * Upgrading-from-3.20:: Upgrading from a 3.20 version to 3.21 * Upgrading-to-arch:: Upgrading to another architecture @end menu @node Upgrading-from-3.22, Upgrading-from-3.21, Upgrade, Upgrade @subsection Upgrading From Version 3.22 to Version 3.23 @cindex compatibility, between MySQL versions @cindex upgrading, 3.22 to 3.23 MySQL Version 3.23 supports tables of the new @code{MyISAM} type and the old @code{ISAM} type. You don't have to convert your old tables to use these with Version 3.23. By default, all new tables will be created with type @code{MyISAM} (unless you start @code{mysqld} with the @code{--default-table-type=isam} option). You can change an @code{ISAM} table to a @code{MyISAM} table with @code{ALTER TABLE table_name TYPE=MyISAM} or the Perl script @code{mysql_convert_table_format}. Version 3.22 and 3.21 clients will work without any problems with a Version 3.23 server. The following lists tell what you have to watch out for when upgrading to Version 3.23: @itemize @bullet @item All tables that uses the @code{tis620} character set must be fixed with @code{myisamchk -r} or @code{REPAIR TABLE}. @item If you do a @code{DROP DATABASE} on a symbolic linked database, both the link and the original database is deleted. (This didn't happen in 3.22 because configure didn't detect the @code{readlink} system call). @item @code{OPTIMIZE TABLE} now only works for @strong{MyISAM} tables. For other table types, you can use @code{ALTER TABLE} to optimize the table. During @code{OPTIMIZE TABLE} the table is now locked from other threads. @item The MySQL client @code{mysql} is now by default started with the option @code{--no-named-commands (-g)}. This option can be disabled with @code{--enable-named-commands (-G)}. This may cause incompatibility problems in some cases, for example in SQL scripts that use named commands without a semicolon! Long format commands still work from the first line. @item If you are using the @code{german} character sort order, you must repair all your tables with @code{isamchk -r}, as we have made some changes in the sort order! @item The default return type of @code{IF} will now depend on both arguments and not only the first argument. @item @code{AUTO_INCREMENT} will not work with negative numbers. The reason for this is that negative numbers caused problems when wrapping from -1 to 0. @code{AUTO_INCREMENT} is now for MyISAM tables handled at a lower level and is much faster than before. For MyISAM tables old numbers are also not reused anymore, even if you delete some rows from the table. @item @code{CASE}, @code{DELAYED}, @code{ELSE}, @code{END}, @code{FULLTEXT}, @code{INNER}, @code{RIGHT}, @code{THEN} and @code{WHEN} are now reserved words. @item @code{FLOAT(X)} is now a true floating-point type and not a value with a fixed number of decimals. @item When declaring @code{DECIMAL(length,dec)} the length argument no longer includes a place for the sign or the decimal point. @item A @code{TIME} string must now be of one of the following formats: @code{[[[DAYS] [H]H:]MM:]SS[.fraction]} or @code{[[[[[H]H]H]H]MM]SS[.fraction]} @item @code{LIKE} now compares strings using the same character comparison rules as @code{'='}. If you require the old behavior, you can compile MySQL with the @code{CXXFLAGS=-DLIKE_CMP_TOUPPER} flag. @item @code{REGEXP} is now case insensitive for normal (not binary) strings. @item When you check/repair tables you should use @code{CHECK TABLE} or @code{myisamchk} for @code{MyISAM} tables (@code{.MYI}) and @code{isamchk} for ISAM (@code{.ISM}) tables. @item If you want your @code{mysqldump} files to be compatible between MySQL Version 3.22 and Version 3.23, you should not use the @code{--opt} or @code{--full} option to @code{mysqldump}. @item Check all your calls to @code{DATE_FORMAT()} to make sure there is a @samp{%} before each format character. (Later MySQL Version 3.22 did allow this syntax.) @item @code{mysql_fetch_fields_direct} is now a function (it was a macro) and it returns a pointer to a @code{MYSQL_FIELD} instead of a @code{MYSQL_FIELD}. @item @code{mysql_num_fields()} can no longer be used on a @code{MYSQL*} object (it's now a function that takes @code{MYSQL_RES*} as an argument. You should now use @code{mysql_field_count()} instead. @item In MySQL Version 3.22, the output of @code{SELECT DISTINCT ...} was almost always sorted. In Version 3.23, you must use @code{GROUP BY} or @code{ORDER BY} to obtain sorted output. @item @code{SUM()} now returns @code{NULL}, instead of 0, if there is no matching rows. This is according to ANSI SQL. @item An @code{AND} or @code{OR} with @code{NULL} values will now return @code{NULL} instead of 0. This mostly affects queries that use @code{NOT} on an @code{AND/OR} expression as @code{NOT NULL} = @code{NULL}. @code{LPAD()} and @code{RPAD()} will shorten the result string if it's longer than the length argument. @end itemize @node Upgrading-from-3.21, Upgrading-from-3.20, Upgrading-from-3.22, Upgrade @subsection Upgrading from Version 3.21 to Version 3.22 @cindex compatibility, between MySQL versions @cindex upgrading, 3.21 to 3.22 Nothing that affects compatibility has changed between Version 3.21 and 3.22. The only pitfall is that new tables that are created with @code{DATE} type columns will use the new way to store the date. You can't access these new fields from an old version of @code{mysqld}. After installing MySQL Version 3.22, you should start the new server and then run the @code{mysql_fix_privilege_tables} script. This will add the new privileges that you need to use the @code{GRANT} command. If you forget this, you will get @code{Access denied} when you try to use @code{ALTER TABLE}, @code{CREATE INDEX}, or @code{DROP INDEX}. If your MySQL root user requires a password, you should give this as an argument to @code{mysql_fix_privilege_tables}. The C API interface to @code{mysql_real_connect()} has changed. If you have an old client program that calls this function, you must place a @code{0} for the new @code{db} argument (or recode the client to send the @code{db} element for faster connections). You must also call @code{mysql_init()} before calling @code{mysql_real_connect()}! This change was done to allow the new @code{mysql_options()} function to save options in the @code{MYSQL} handler structure. The @code{mysqld} variable @code{key_buffer} has changed names to @code{key_buffer_size}, but you can still use the old name in your startup files. @node Upgrading-from-3.20, Upgrading-to-arch, Upgrading-from-3.21, Upgrade @subsection Upgrading from Version 3.20 to Version 3.21 @cindex upgrading, 3.20 to 3.21 If you are running a version older than Version 3.20.28 and want to switch to Version 3.21, you need to do the following: You can start the @code{mysqld} Version 3.21 server with @code{safe_mysqld --old-protocol} to use it with clients from a Version 3.20 distribution. In this case, the new client function @code{mysql_errno()} will not return any server error, only @code{CR_UNKNOWN_ERROR} (but it works for client errors), and the server uses the old @code{password()} checking rather than the new one. If you are @strong{NOT} using the @code{--old-protocol} option to @code{mysqld}, you will need to make the following changes: @itemize @bullet @item All client code must be recompiled. If you are using ODBC, you must get the new @strong{MyODBC} 2.x driver. @item The script @code{scripts/add_long_password} must be run to convert the @code{Password} field in the @code{mysql.user} table to @code{CHAR(16)}. @item All passwords must be reassigned in the @code{mysql.user} table (to get 62-bit rather than 31-bit passwords). @item The table format hasn't changed, so you don't have to convert any tables. @end itemize MySQL Version 3.20.28 and above can handle the new @code{user} table format without affecting clients. If you have a MySQL version earlier than Version 3.20.28, passwords will no longer work with it if you convert the @code{user} table. So to be safe, you should first upgrade to at least Version 3.20.28 and then upgrade to Version 3.21. @cindex Protocol mismatch The new client code works with a 3.20.x @code{mysqld} server, so if you experience problems with 3.21.x, you can use the old 3.20.x server without having to recompile the clients again. If you are not using the @code{--old-protocol} option to @code{mysqld}, old clients will issue the error message: @example ERROR: Protocol mismatch. Server Version = 10 Client Version = 9 @end example The new Perl @code{DBI}/@code{DBD} interface also supports the old @code{mysqlperl} interface. The only change you have to make if you use @code{mysqlperl} is to change the arguments to the @code{connect()} function. The new arguments are: @code{host}, @code{database}, @code{user}, @code{password} (the @code{user} and @code{password} arguments have changed places). @xref{Perl DBI Class, , Perl @code{DBI} Class}. The following changes may affect queries in old applications: @itemize @bullet @item @code{HAVING} must now be specified before any @code{ORDER BY} clause. @item The parameters to @code{LOCATE()} have been swapped. @item There are some new reserved words. The most notable are @code{DATE}, @code{TIME}, and @code{TIMESTAMP}. @end itemize @node Upgrading-to-arch, , Upgrading-from-3.20, Upgrade @subsection Upgrading to Another Architecture @cindex upgrading, different architecture If you are using MySQL Version 3.23, you can copy the @code{.frm}, @code{.MYI}, and @code{.MYD} files between different architectures that support the same floating-point format. (MySQL takes care of any byte swapping issues.) The MySQL @code{ISAM} data and index files (@file{.ISD} and @file{*.ISM}, respectively) are architecture-dependent and in some cases OS-dependent. If you want to move your applications to another machine that has a different architecture or OS than your current machine, you should not try to move a database by simply copying the files to the other machine. Use @code{mysqldump} instead. By default, @code{mysqldump} will create a file full of SQL statements. You can then transfer the file to the other machine and feed it as input to the @code{mysql} client. Try @code{mysqldump --help} to see what options are available. If you are moving the data to a newer version of MySQL, you should use @code{mysqldump --opt} with the newer version to get a fast, compact dump. The easiest (although not the fastest) way to move a database between two machines is to run the following commands on the machine on which the database is located: @example shell> mysqladmin -h 'other hostname' create db_name shell> mysqldump --opt db_name \ | mysql -h 'other hostname' db_name @end example If you want to copy a database from a remote machine over a slow network, you can use: @example shell> mysqladmin create db_name shell> mysqldump -h 'other hostname' --opt --compress db_name \ | mysql db_name @end example You can also store the result in a file, then transfer the file to the target machine and load the file into the database there. For example, you can dump a database to a file on the source machine like this: @example shell> mysqldump --quick db_name | gzip > db_name.contents.gz @end example (The file created in this example is compressed.) Transfer the file containing the database contents to the target machine and run these commands there: @example shell> mysqladmin create db_name shell> gunzip < db_name.contents.gz | mysql db_name @end example @cindex @code{mysqldump} @cindex @code{mysqlimport} You can also use @code{mysqldump} and @code{mysqlimport} to accomplish the database transfer. For big tables, this is much faster than simply using @code{mysqldump}. In the commands shown below, @code{DUMPDIR} represents the full pathname of the directory you use to store the output from @code{mysqldump}. First, create the directory for the output files and dump the database: @example shell> mkdir DUMPDIR shell> mysqldump --tab=DUMPDIR db_name @end example Then transfer the files in the @code{DUMPDIR} directory to some corresponding directory on the target machine and load the files into MySQL there: @example shell> mysqladmin create db_name # create database shell> cat DUMPDIR/*.sql | mysql db_name # create tables in database shell> mysqlimport db_name DUMPDIR/*.txt # load data into tables @end example Also, don't forget to copy the @code{mysql} database, because that's where the grant tables (@code{user}, @code{db}, @code{host}) are stored. You may have to run commands as the MySQL @code{root} user on the new machine until you have the @code{mysql} database in place. After you import the @code{mysql} database on the new machine, execute @code{mysqladmin flush-privileges} so that the server reloads the grant table information. @node Operating System Specific Notes, , Upgrade, Installing @section Operating System Specific Notes @menu * Linux:: Linux Notes (All Linux Versions) * Windows:: Windows Notes * Solaris:: Solaris Notes * BSD Notes:: BSD Notes * Mac OS X:: Mac OS X Notes * Other Unix Notes:: Other Unix Notes * OS/2:: OS/2 Notes * BeOS:: BeOS Notes * Novell Netware:: Novell Netware Notes @end menu @node Linux, Windows, Operating System Specific Notes, Operating System Specific Notes @subsection Linux Notes (All Linux Versions) @menu * Binary notes-Linux:: Linux Notes for Binary Distributions * Linux-x86:: Linux x86 Notes * Linux-SPARC:: Linux SPARC Notes * Linux-Alpha:: Linux Alpha Notes * Linux-PowerPC:: Linux PowerPC Notes * Linux-MIPS:: Linux MIPS Notes * Linux-IA64:: Linux IA64 Notes @end menu The notes below regarding @strong{glibc} apply only to the situation when you build MySQL yourself. If you are running Linux on an x86 machine, in most cases it is much better for you to just use our binary. We link our binaries against the best patched version of @strong{glibc} we can come up with and with the best compiler options, in an attempt to make it suitable for a high-load server. So if you read the text below, and are in doubt about what you should do, try our binary first to see if it meets your needs, and worry about your own build only after you have discovered that our binary is not good enough. In that case, we would appreciate a note about it, so we can build a better binary next time. For a typical user, even for setups with a lot of concurrent connections and/or tables exceeding 2GB limit, our binary in most cases is the best choice. MySQL uses LinuxThreads on Linux. If you are using an old Linux version that doesn't have @code{glibc2}, you must install LinuxThreads before trying to compile MySQL. You can get LinuxThreads at @uref{http://www.mysql.com/Downloads/Linux}. @strong{NOTE:} We have seen some strange problems with Linux 2.2.14 and MySQL on SMP systems; If you have a SMP system, we recommend you to upgrade to Linux 2.4 ASAP! Your system will be faster and more stable by doing this! Note that @code{glibc} versions before and including Version 2.1.1 have a fatal bug in @code{pthread_mutex_timedwait} handling, which is used when you do @code{INSERT DELAYED}. We recommend you to not use @code{INSERT DELAYED} before upgrading glibc. If you plan to have 1000+ concurrent connections, you will need to make some changes to LinuxThreads, recompile it, and relink MySQL against the new @file{libpthread.a}. Increase @code{PTHREAD_THREADS_MAX} in @file{sysdeps/unix/sysv/linux/bits/local_lim.h} to 4096 and decrease @code{STACK_SIZE} in @file{linuxthreads/internals.h} to 256 KB. The paths are relative to the root of @code{glibc} Note that MySQL will not be stable with around 600-1000 connections if @code{STACK_SIZE} is the default of 2 MB. If you have a problem with that MySQL can't open enough files, or connections, it may be that you haven't configured Linux to handle enough files. In Linux 2.2 and forwards, you can check the number of allocated file handlers by doing: @example cat /proc/sys/fs/file-max cat /proc/sys/fs/dquot-max cat /proc/sys/fs/super-max @end example If you have more than 16M of memory, you should add something like the following in your boot script (@file{/etc/rc/boot.local} on SuSE): @example echo 65536 > /proc/sys/fs/file-max echo 8192 > /proc/sys/fs/dquot-max echo 1024 > /proc/sys/fs/super-max @end example You can also run the above from the command line as root, but in this case your old limits will be used next time your computer reboots. You should also add /etc/my.cnf: @example [safe_mysqld] open-files-limit=8192 @end example The above should allow MySQL to create up to 8192 connections + files. The @code{STACK_SIZE} constant in LinuxThreads controls the spacing of thread stacks in the address space. It needs to be large enough so that there will be plenty of room for the stack of each individual thread, but small enough to keep the stack of some thread from running into the global @code{mysqld} data. Unfortunately, the Linux implementation of @code{mmap()}, as we have experimentally discovered, will successfully unmap an already mapped region if you ask it to map out an address already in use, zeroing out the data on the entire page, instead of returning an error. So, the safety of @code{mysqld} or any other threaded application depends on the "gentleman" behavior of the code that creates threads. The user must take measures to make sure the number of running threads at any time is sufficiently low for thread stacks to stay away from the global heap. With @code{mysqld}, you should enforce this "gentleman" behavior by setting a reasonable value for the @code{max_connections} variable. If you build MySQL yourself and do not want to mess with patching LinuxThreads, you should set @code{max_connections} to a value no higher than 500. It should be even less if you have a large key buffer, large heap tables, or some other things that make @code{mysqld} allocate a lot of memory or if you are running a 2.2 kernel with a 2GB patch. If you are using our binary or RPM version 3.23.25 or later, you can safely set @code{max_connections} at 1500, assuming no large key buffer or heap tables with lots of data. The more you reduce @code{STACK_SIZE} in LinuxThreads the more threads you can safely create. We recommend the values between 128K and 256K. If you use a lot of concurrent connections, you may suffer from a "feature" in the 2.2 kernel that penalizes a process for forking or cloning a child in an attempt to prevent a fork bomb attack. This will cause MySQL not to scale well as you increase the number of concurrent clients. On single CPU systems, we have seen this manifested in a very slow thread creation, which means it may take a long time to connect to MySQL (as long as 1 minute), and it may take just as long to shut it down. On multiple CPU systems, we have observed a gradual drop in query speed as the number of clients increases. In the process of trying to find a solution, we have received a kernel patch from one of our users, who claimed it made a lot of difference for his site. The patch is available here (@uref{http://www.mysql.com/Downloads/Patches/linux-fork.patch}). We have now done rather extensive testing of this patch on both development and production systems. It has significantly improved @code{MySQL} performance without causing any problems and we now recommend it to our users who are still running high-load servers on 2.2 kernels. This issue has been fixed in the 2.4 kernel, so if you are not satisfied with the current performance of your system, rather than patching your 2.2 kernel, it might be easier to just upgrade to 2.4, which will also give you a nice SMP boost in addition to fixing this fairness bug. We have tested MySQL on the 2.4 kernel on a 2 CPU machine and found MySQL scales MUCH better - there was virtually no slowdown on query throughput all the way up to 1000 clients, and MySQL scaling factor ( computed as the ratio of maximum throughput to the throughput with one client) was 180%. We have observed similar results on a 4-CPU system - virtually no slowdown as the number of clients was increased up to 1000, and 300% scaling factor. So for a high-load SMP server we would definitely recommend the 2.4 kernel at this point. We have discovered that it is essential to run @code{mysqld} process with the highest possible priority on the 2.4 kernel to achieve maximum performance. This can be done by adding @code{renice -20 $$} command to @code{safe_mysqld}. In our testing on a 4-CPU machine, increasing the priority gave 60% increase in throughput with 400 clients. We are currently also trying to collect more info on how well @code{MySQL} performs on 2.4 kernel on 4-way and 8-way systems. If you have access such a system and have done some benchmarks, please send a mail to @email{docs@@mysql.com} with the results - we will include them in the manual. There is another issue that greatly hurts MySQL performance, especially on SMP systems. The implementation of mutex in LinuxThreads in @strong{glibc-2.1} is very bad for programs with many threads that only hold the mutex for a short time. On an SMP system, ironic as it is, if you link MySQL against unmodified @strong{LinuxThreads}, removing processors from the machine improves MySQL performance in many cases. We have made a patch available for @strong{glibc 2.1.3}, @uref{http://www.mysql.com/Downloads/Linux/linuxthreads-2.1-patch,linuxthreads-2.1-patch} to correct this behavior. With @strong{glibc-2.2.2} MySQL version 3.23.36 will use the adaptive mutex, which is much better than even the patched one in @strong{glibc-2.1.3}. Be warned, however, that under some conditions, the current mutex code in @strong{glibc-2.2.2} overspins, which hurts MySQL performance. The chance of this condition can be reduced by renicing @code{mysqld} process to the highest priority. We have also been able to correct the overspin behavior with a patch, available @uref{http://www.mysql.com/Downloads/Linux/linuxthreads-2.2.2.patch,here}. It combines the correction of overspin, maximum number of threads, and stack spacing all in one. You will need to apply it in the @code{linuxthreads} directory with @code{patch -p0 </tmp/linuxthreads-2.2.2.patch}. We hope it will be included in some form in to the future releases of @code{glibc-2.2}. In any case, if you link against @code{glibc-2.2.2} you still need to correct @code{STACK_SIZE} and @code{PTHREAD_THREADS_MAX}. We hope that the defaults will be corrected to some more acceptable values for high-load MySQL setup in the future, so that your own build can be reduced to @code{./configure; make; make install}. We recommend that you use the above patches to build a special static version of @code{libpthread.a} and use it only for statically linking against @code{MySQL}. We know that the patches are safe for @code{MySQL} and significantly improve its performance, but we cannot say anything about other applications. If you link other applications against the patched version of the library, or build a patched shared version and install it on your system, you are doing it at your own risk with regard to other applications that depend on @code{LinuxThreads}. If you experience any strange problems during the installation of MySQL, or with some common utilties hanging, it is very likely that they are either library or compiler related. If this is the case, using our binary will resolve them. One known problem with the binary distribution is that with older Linux systems that use @code{libc} (like RedHat 4.x or Slackware), you will get some non-fatal problems with hostname resolution. @xref{Binary notes-Linux}. When using LinuxThreads you will see a minimum of three processes running. These are in fact threads. There will be one thread for the LinuxThreads manager, one thread to handle connections, and one thread to handle alarms and signals. Note that the Linux kernel and the LinuxThread library can by default only have 1024 threads. This means that you can only have up to 1021 connections to MySQL on an unpatched system. The page @uref{http://www.volano.com/linuxnotes.html} contains information how to go around this limit. If you see a dead @code{mysqld} daemon process with @code{ps}, this usually means that you have found a bug in MySQL or you have a corrupted table. @xref{Crashing}. To get a core dump on Linux if @code{mysqld} dies with a SIGSEGV signal, you can start @code{mysqld} with the @code{--core-file} option. Note that you also probably need to raise the @code{core file size} by adding @code{ulimit -c 1000000} to @code{safe_mysqld} or starting @code{safe_mysqld} with @code{--core-file-sizes=1000000}. @xref{safe_mysqld, , @code{safe_mysqld}}. If you are linking your own MySQL client and get the error: @example ld.so.1: ./my: fatal: libmysqlclient.so.4: open failed: No such file or directory @end example When executing them, the problem can be avoided by one of the following methods: @itemize @bullet @item Link the client with the following flag (instead of @code{-Lpath}): @code{-Wl,r/path-libmysqlclient.so}. @item Copy @code{libmysqclient.so} to @file{/usr/lib}. @tindex LD_RUN_PATH environment variable @tindex environment variable, LD_RUN_PATH @item Add the pathname of the directory where @code{libmysqlclient.so} is located to the @code{LD_RUN_PATH} environment variable before running your client. @end itemize If you are using the Fujitsu compiler @code{(fcc / FCC)} you will have some problems compiling MySQL because the Linux header files are very @code{gcc} oriented. The following @code{configure} line should work with @code{fcc/FCC}: @example CC=fcc CFLAGS="-O -K fast -K lib -K omitfp -Kpreex -D_GNU_SOURCE -DCONST=const -DNO_STRTOLL_PROTO" CXX=FCC CXXFLAGS="-O -K fast -K lib -K omitfp -K preex --no_exceptions --no_rtti -D_GNU_SOURCE -DCONST=const -Dalloca=__builtin_alloca -DNO_STRTOLL_PROTO '-D_EXTERN_INLINE=static __inline'" ./configure --prefix=/usr/local/mysql --enable-assembler --with-mysqld-ldflags=-all-static --disable-shared --with-low-memory @end example @node Binary notes-Linux, Linux-x86, Linux, Linux @subsubsection Linux Notes for Binary Distributions @cindex binary distributions, on Linux @cindex Linux, binary distribution MySQL needs at least Linux Version 2.0. The binary release is linked with @code{-static}, which means you do not normally need to worry about which version of the system libraries you have. You need not install LinuxThreads, either. A program linked with @code{-static} is slightly bigger than a dynamically linked program but also slightly faster (3-5%). One problem, however, is that you can't use user-definable functions (UDFs) with a statically linked program. If you are going to write or use UDF functions (this is something only for C or C++ programmers), you must compile MySQL yourself, using dynamic linking. If you are using a @code{libc}-based system (instead of a @code{glibc2} system), you will probably get some problems with hostname resolving and @code{getpwnam()} with the binary release. (This is because @code{glibc} unfortunately depends on some external libraries to resolve hostnames and @code{getpwent()}, even when compiled with @code{-static}). In this case you probably get the following error message when you run @code{mysql_install_db}: @example Sorry, the host 'xxxx' could not be looked up @end example or the following error when you try to run @code{mysqld} with the @code{--user} option: @example getpwnam: No such file or directory @end example You can solve this problem in one of the following ways: @itemize @bullet @item Get a MySQL source distribution (an RPM or the @code{tar.gz} distribution) and install this instead. @item Execute @code{mysql_install_db --force}; This will not execute the @code{resolveip} test in @code{mysql_install_db}. The downside is that you can't use host names in the grant tables; you must use IP numbers instead (except for @code{localhost}). If you are using an old MySQL release that doesn't support @code{--force}, you have to remove the @code{resolveip} test in @code{mysql_install} with an editor. @item Start @code{mysqld} with @code{su} instead of using @code{--user}. @end itemize The Linux-Intel binary and RPM releases of MySQL are configured for the highest possible speed. We are always trying to use the fastest stable compiler available. MySQL Perl support requires Version Perl 5.004_03 or newer. On some Linux 2.2 versions, you may get the error @code{Resource temporarily unavailable} when you do a lot of new connections to a @code{mysqld} server over TCP/IP. The problem is that Linux has a delay between when you close a TCP/IP socket and until this is actually freed by the system. As there is only room for a finite number of TCP/IP slots, you will get the above error if you try to do too many new TCP/IP connections during a small time, like when you run the MySQL @file{test-connect} benchmark over TCP/IP. We have mailed about this problem a couple of times to different Linux mailing lists but have never been able to resolve this properly. The only known 'fix' to this problem is to use persistent connections in your clients or use sockets, if you are running the database server and clients on the same machine. We hope that the @code{Linux 2.4} kernel will fix this problem in the future. @node Linux-x86, Linux-SPARC, Binary notes-Linux, Linux @subsubsection Linux x86 Notes MySQL requires @code{libc} Version 5.4.12 or newer. It's known to work with @code{libc} 5.4.46. @code{glibc} Version 2.0.6 and later should also work. There have been some problems with the @code{glibc} RPMs from RedHat, so if you have problems, check whether or not there are any updates! The @code{glibc} 2.0.7-19 and 2.0.7-29 RPMs are known to work. On some older Linux distributions, @code{configure} may produce an error like this: @example Syntax error in sched.h. Change _P to __P in the /usr/include/sched.h file. See the Installation chapter in the Reference Manual. @end example Just do what the error message says and add an extra underscore to the @code{_P} macro that has only one underscore, then try again. You may get some warnings when compiling; those shown below can be ignored: @example mysqld.cc -o objs-thread/mysqld.o mysqld.cc: In function `void init_signals()': mysqld.cc:315: warning: assignment of negative value `-1' to `long unsigned int' mysqld.cc: In function `void * signal_hand(void *)': mysqld.cc:346: warning: assignment of negative value `-1' to `long unsigned int' @end example In Debian GNU/Linux, if you want MySQL to start automatically when the system boots, do the following: @example shell> cp support-files/mysql.server /etc/init.d/mysql.server shell> /usr/sbin/update-rc.d mysql.server defaults 99 @end example @code{mysql.server} can be found in the @file{share/mysql} directory under the MySQL installation directory or in the @file{support-files} directory of the MySQL source tree. If @code{mysqld} always core dumps when it starts up, the problem may be that you have an old @file{/lib/libc.a}. Try renaming it, then remove @file{sql/mysqld} and do a new @code{make install} and try again. This problem has been reported on some Slackware installations. If you get the following error when linking @code{mysqld}, it means that your @file{libg++.a} is not installed correctly: @example /usr/lib/libc.a(putc.o): In function `_IO_putc': putc.o(.text+0x0): multiple definition of `_IO_putc' @end example You can avoid using @file{libg++.a} by running @code{configure} like this: @example shell> CXX=gcc ./configure @end example @node Linux-SPARC, Linux-Alpha, Linux-x86, Linux @subsubsection Linux SPARC Notes In some implementations, @code{readdir_r()} is broken. The symptom is that @code{SHOW DATABASES} always returns an empty set. This can be fixed by removing @code{HAVE_READDIR_R} from @file{config.h} after configuring and before compiling. Some problems will require patching your Linux installation. The patch can be found at @uref{http://www.mysql.com/Downloads/patches/Linux-sparc-2.0.30.diff}. This patch is against the Linux distribution @file{sparclinux-2.0.30.tar.gz} that is available at @code{vger.rutgers.edu} (a version of Linux that was never merged with the official 2.0.30). You must also install LinuxThreads Version 0.6 or newer. @node Linux-Alpha, Linux-PowerPC, Linux-SPARC, Linux @subsubsection Linux Alpha Notes MySQL Version 3.23.12 is the first MySQL version that is tested on Linux-Alpha. If you plan to use MySQL on Linux-Alpha, you should ensure that you have this version or newer. We have tested MySQL on Alpha with our benchmarks and test suite, and it appears to work nicely. The main thing we haven't yet had time to test is how things works with many concurrent users. When we compiled the standard MySQL binary we are using SuSE 6.4, kernel 2.2.13-SMP, Compaq C compiler (V6.2-504) and Compaq C++ compiler (V6.3-005) on a Comaq DS20 machine with an Alpha EV6 processor. You can find the above compilers at @uref{http://www.support.compaq.com/alpha-tools}). By using these compilers, instead of gcc, we get about 9-14 % better performance with MySQL. Note that the configure line optimized the binary for the current CPU; This means you can only use our binary if you have an Alpha EV6 processor. We also compile statically to avoid library problems. @example CC=ccc CFLAGS="-fast" CXX=cxx CXXFLAGS="-fast -noexceptions -nortti" ./configure --prefix=/usr/local/mysql --disable-shared --with-extra-charsets=complex --enable-thread-safe-client --with-mysqld-ldflags=-non_shared --with-client-ldflags=-non_shared @end example If you want to use egcs the following configure line worked for us: @example CFLAGS="-O3 -fomit-frame-pointer" CXX=gcc CXXFLAGS="-O3 -fomit-frame-pointer -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --disable-shared @end example Some known problems when running MySQL on Linux-Alpha: @itemize @bullet @item Debugging threaded applications like MySQL will not work with @code{gdb 4.18}. You should download and use gdb 5.0 instead! @item If you try linking @code{mysqld} statically when using @code{gcc}, the resulting image will core dump at start. In other words, @strong{DON'T} use @code{--with-mysqld-ldflags=-all-static} with @code{gcc}. @end itemize @node Linux-PowerPC, Linux-MIPS, Linux-Alpha, Linux @subsubsection Linux PowerPC Notes MySQL should work on MkLinux with the newest @code{glibc} package (tested with @code{glibc} 2.0.7). @node Linux-MIPS, Linux-IA64, Linux-PowerPC, Linux @subsubsection Linux MIPS Notes To get MySQL to work on Qube2, (Linux Mips), you need the newest @code{glibc} libraries (@code{glibc-2.0.7-29C2} is known to work). You must also use the @code{egcs} C++ compiler (@code{egcs-1.0.2-9}, @code{gcc 2.95.2} or newer). @node Linux-IA64, , Linux-MIPS, Linux @subsubsection Linux IA64 Notes To get MySQL to compile on Linux Ia64, we use the following compile line: Using @code{gcc-2.96}: @example CC=gcc CFLAGS="-O3 -fno-omit-frame-pointer" CXX=gcc CXXFLAGS="-O3 -fno-omit-frame-pointer -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql "--with-comment=Official MySQL binary" --with-extra-charsets=complex @end example On Ia64 the MySQL client binaries are using shared libraries. This means that if you install our binary distribution in some other place than @file{/usr/local/mysql} you need to either modify @file{/etc/ld.so.conf} or add the path to the directory where you have @file{libmysqlclient.so} to the @code{LD_LIBRARY_PATH} environment variable. @xref{Link errors}. @node Windows, Solaris, Linux, Operating System Specific Notes @subsection Windows Notes This section describes installation and use of MySQL on Windows. This information is also provided in the @file{README} file that comes with the MySQL Windows distribution. @menu * Win95 start:: Starting MySQL on Win95 / Win98 * NT start:: Starting MySQL on NT / Win2000 * Windows running:: Running MySQL on Windows * Windows and SSH:: Connecting to a remote MySQL from Windows with SSH * Windows symbolic links:: Splitting data across different disks under Win32 * Windows compiling:: Compiling MySQL clients on Windows. * Windows vs Unix:: MySQL-Windows compared to Unix MySQL @end menu @node Win95 start, NT start, Windows, Windows @subsubsection Starting MySQL on Windows 95 or Windows 98 MySQL uses TCP/IP to connect a client to a server. (This will allow any machine on your network to connect to your MySQL server.) Because of this, you must install TCP/IP on your machine before starting MySQL. You can find TCP/IP on your Windows CD-ROM. Note that if you are using an old Win95 release (for example OSR2), it's likely that you have an old Winsock package! MySQL requires Winsock 2! You can get the newest Winsock from @uref{http://www.microsoft.com/}. Win98 has the new Winsock 2 library, so the above doesn't apply for Win98. To start the @code{mysqld} server, you should start an MS-DOS window and type: @example C:\> C:\mysql\bin\mysqld @end example This will start @code{mysqld} in the background without a window. You can kill the MySQL server by executing: @example C:\> C:\mysql\bin\mysqladmin -u root shutdown @end example Note that Win95 and Win98 don't support creation of named pipes. On Win95 and Win98, you can only use named pipes to connect to a remote MySQL server running on a Windows NT server host. (The MySQL server must also support named pipes, of course. For example, using @code{mysqld-opt} under NT will not allow named pipe connections. You should use either @code{mysqld-nt} or @code{mysqld-max-nt}.) If @code{mysqld} doesn't start, please check the @file{\mysql\data\mysql.err} file to see if the server wrote any message there to indicate the cause of the problem. You can also try to start the server with @code{mysqld --standalone}; In this case, you may get some useful information on the screen that may help solve the problem. The last option is to start @code{mysqld} with @code{--standalone --debug}. In this case @code{mysqld} will write a log file @file{C:\mysqld.trace} that should contain the reason why @code{mysqld} doesn't start. @xref{Making trace files}. @node NT start, Windows running, Win95 start, Windows @subsubsection Starting MySQL on Windows NT or Windows 2000 The Win95/Win98 section also applies to MySQL on NT/Win2000, with the following differences: To get MySQL to work with TCP/IP on NT, you must install service pack 3 (or newer)! Note that everything in the following that applies for NT also applies for Win2000! For NT/Win2000, the server name is @code{mysqld-nt}. Normally you should install MySQL as a service on NT/Win2000: @example C:\> C:\mysql\bin\mysqld-nt --install @end example or @example C:\> C:\mysql\bin\mysqld-max-nt --install @end example (Under Windows NT, you can actually install any of the server binaries as a service, but only those having names that end with @code{-nt.exe} provide support for named pipes.) You can start and stop the MySQL service with these commands: @example C:\> NET START mysql C:\> NET STOP mysql @end example Note that in this case you can't use any other options for @code{mysqld-nt}! You can also run @code{mysqld-nt} as a stand-alone program on NT if you need to start @code{mysqld-nt} with any options! If you start @code{mysqld-nt} without options on NT, @code{mysqld-nt} tries to start itself as a service with the default service options. If you have stopped @code{mysqld-nt}, you have to start it with @code{NET START mysql}. The service is installed with the name @code{MySQL}. Once installed, it must be started using the Services Control Manager (SCM) Utility found in the Control Panel, or by using the @code{NET START MySQL} command. If any options are desired, they must be specified as ``Startup parameters'' in the SCM utility before you start the MySQL service. Once running, @code{mysqld-nt} can be stopped using @code{mysqladmin}, or from the SCM utility or by using the command @code{NET STOP MySQL}. If you use SCM to stop @code{mysqld-nt}, there is a strange message from SCM about @code{mysqld shutdown normally}. When run as a service, @code{mysqld-nt} has no access to a console and so no messages can be seen. On NT you can get the following service error messages: @multitable @columnfractions .3 .7 @item Permission Denied @tab Means that it cannot find @code{mysqld-nt.exe}. @item Cannot Register @tab Means that the path is incorrect. @item Failed to install service. @tab Means that the service is already installed or that the Service Control Manager is in bad state. @end multitable If you have problems installing @code{mysqld-nt} as a service, try starting it with the full path: @example C:\> C:\mysql\bin\mysqld-nt --install @end example If this doesn't work, you can get @code{mysqld-nt} to start properly by fixing the path in the registry! If you don't want to start @code{mysqld-nt} as a service, you can start it as follows: @example C:\> C:\mysql\bin\mysqld-nt --standalone @end example or @example C:\> C:\mysql\bin\mysqld --standalone --debug @end example The last version gives you a debug trace in @file{C:\mysqld.trace}. @xref{Making trace files}. @node Windows running, Windows and SSH, NT start, Windows @subsubsection Running MySQL on Windows @cindex TCP/IP @cindex named pipes MySQL supports TCP/IP on all Windows platforms and named pipes on NT. The default is to use named pipes for local connections on NT and TCP/IP for all other cases if the client has TCP/IP installed. The host name specifies which protocol is used: @multitable @columnfractions .3 .7 @strong{Host name} @tab @strong{Protocol} @item NULL (none) @tab On NT, try named pipes first; if that doesn't work, use TCP/IP. On Win95/Win98, TCP/IP is used. @item . @tab Named pipes @item localhost @tab TCP/IP to current host @item hostname @tab TCP/IP @end multitable You can force a MySQL client to use named pipes by specifying the @code{--pipe} option or by specifying @code{.} as the host name. Use the @code{--socket} option to specify the name of the pipe. Note that starting from 3.23.50, named pipes are only enabled if mysqld is started with with @code{--enable-named-pipe}. This is because some users have experienced problems shutting down the MySQL server when one uses named pipes. You can test whether or not MySQL is working by executing the following commands: @example C:\> C:\mysql\bin\mysqlshow C:\> C:\mysql\bin\mysqlshow -u root mysql C:\> C:\mysql\bin\mysqladmin version status proc C:\> C:\mysql\bin\mysql test @end example If @code{mysqld} is slow to answer to connections on Win95/Win98, there is probably a problem with your DNS. In this case, start @code{mysqld} with @code{--skip-name-resolve} and use only @code{localhost} and IP numbers in the MySQL grant tables. You can also avoid DNS when connecting to a @code{mysqld-nt} MySQL server running on NT by using the @code{--pipe} argument to specify use of named pipes. This works for most MySQL clients. There are two versions of the MySQL command-line tool: @multitable @columnfractions .25 .75 @item @code{mysql} @tab Compiled on native Windows, which offers very limited text editing capabilities. @item @code{mysqlc} @tab Compiled with the Cygnus GNU compiler and libraries, which offers @code{readline} editing. @end multitable If you want to use @code{mysqlc.exe}, you must copy @file{C:\mysql\lib\cygwinb19.dll} to your Windows system directory (@file{\windows\system} or similar place). The default privileges on Windows give all local users full privileges to all databases without specifying a password. To make MySQL more secure, you should set a password for all users and remove the row in the @code{mysql.user} table that has @code{Host='localhost'} and @code{User=''}. You should also add a password for the @code{root} user. The following example starts by removing the anonymous user that can be used by anyone to access the @code{test} database, then sets a @code{root} user password: @example C:\> C:\mysql\bin\mysql mysql mysql> DELETE FROM user WHERE Host='localhost' AND User=''; mysql> QUIT C:\> C:\mysql\bin\mysqladmin reload C:\> C:\mysql\bin\mysqladmin -u root password your_password @end example After you've set the password, if you want to take down the @code{mysqld} server, you can do so using this command: @example C:\> mysqladmin --user=root --password=your_password shutdown @end example If you are using the old shareware version of MySQL Version 3.21 under Windows, the above command will fail with an error: @code{parse error near 'SET OPTION password'}. The fix is in to upgrade to the current MySQL version, which is freely available. With the current MySQL versions you can easily add new users and change privileges with @code{GRANT} and @code{REVOKE} commands. @xref{GRANT}. @node Windows and SSH, Windows symbolic links, Windows running, Windows @subsubsection Connecting to a Remote MySQL from Windows with SSH @c FIX this is ugly, real ugly. @cindex SSH @cindex connecting, remotely with SSH Here is a note about how to connect to get a secure connection to remote MySQL server with SSH (by David Carlson @email{dcarlson@@mplcomm.com}): @itemize @bullet @item Install an SSH client on your Windows machine. As a user, the best non-free one I've found is from @code{SecureCRT} from @uref{http://www.vandyke.com/}. Another option is @code{f-secure} from @uref{http://www.f-secure.com/}. You can also find some free ones on @strong{Google} at @uref{http://directory.google.com/Top/Computers/Security/Products_and_Tools/Cryptography/SSH/Clients/Windows/}. @item Start your Windows SSH client. Set @code{Host_Name = yourmysqlserver_URL_or_IP}. Set @code{userid=your_userid} to log in to your server (probably not the same as your MySQL login/password. @item Set up port forwarding. Either do a remote forward (Set @code{local_port: 3306}, @code{remote_host: yourmysqlservername_or_ip}, @code{remote_port: 3306} ) or a local forward (Set @code{port: 3306}, @code{host: localhost}, @code{remote port: 3306}). @item Save everything, otherwise you'll have to redo it the next time. @item Log in to your server with SSH session you just created. @item On your Windows machine, start some ODBC application (such as Access). @item Create a new file in Windows and link to MySQL using the ODBC driver the same way you normally do, EXCEPT type in @code{localhost} for the MySQL host server --- not @code{yourmysqlservername}. @end itemize You should now have an ODBC connection to MySQL, encrypted using SSH. @node Windows symbolic links, Windows compiling, Windows and SSH, Windows @subsubsection Splitting Data Across Different Disks on Windows @cindex symbolic links @cindex using multiple disks to start data @cindex disks, splitting data across Beginning with MySQL Version 3.23.16, the @code{mysqld-max} and @code{mysql-max-nt} servers in the MySQL distribution are compiled with the @code{-DUSE_SYMDIR} option. This allows you to put a database on different disk by adding a symbolic link to it (in a manner similar to the way that symbolic links work on Unix). On Windows, you make a symbolic link to a database by creating a file that contains the path to the destination directory and saving this in the @file{mysql_data} directory under the filename @file{database.sym}. Note that the symbolic link will be used only if the directory @file{mysql_data_dir\database} doesn't exist. For example, if the MySQL data directory is @file{C:\mysql\data} and you want to have database @code{foo} located at @file{D:\data\foo}, you should create the file @file{C:\mysql\data\foo.sym} that contains the text @code{D:\data\foo\}. After that, all tables created in the database @code{foo} will be created in @file{D:\data\foo}. Note that because of the speed penalty you get when opening every table, we have not enabled this by default even if you have compiled MySQL with support for this. To enable symlinks you should put in your @code{my.cnf} or @code{my.ini} file the following entry: @example [mysqld] use-symbolic-links @end example In MySQL 4.0 we will enable symlinks by default. Then you should instead use the @code{skip-symlink} option if you want to disable this. @node Windows compiling, Windows vs Unix, Windows symbolic links, Windows @subsubsection Compiling MySQL Clients on Windows @cindex compiling, on Windows @cindex Windows, compiling on In your source files, you should include @file{windows.h} before you include @file{mysql.h}: @example #if defined(_WIN32) || defined(_WIN64) #include <windows.h> #endif #include <mysql.h> @end example You can either link your code with the dynamic @file{libmysql.lib} library, which is just a wrapper to load in @file{libmysql.dll} on demand, or link with the static @file{mysqlclient.lib} library. Note that as the mysqlclient libraries are compiled as threaded libraries, you should also compile your code to be multi-threaded! @node Windows vs Unix, , Windows compiling, Windows @subsubsection MySQL-Windows Compared to Unix MySQL @cindex Windows, versus Unix @cindex operating systems, Windows versus Unix MySQL-Windows has by now proven itself to be very stable. This version of MySQL has the same features as the corresponding Unix version with the following exceptions: @table @strong @item Win95 and threads Win95 leaks about 200 bytes of main memory for each thread creation. Each connection in MySQL creates a new thread, so you shouldn't run @code{mysqld} for an extended time on Win95 if your server handles many connections! WinNT and Win98 don't suffer from this bug. @item Concurrent reads MySQL depends on the @code{pread()} and @code{pwrite()} calls to be able to mix @code{INSERT} and @code{SELECT}. Currently we use mutexes to emulate @code{pread()}/@code{pwrite()}. We will, in the long run, replace the file level interface with a virtual interface so that we can use the @code{readfile()}/@code{writefile()} interface on NT to get more speed. The current implementation limits the number of open files MySQL can use to 1024, which means that you will not be able to run as many concurrent threads on NT as on Unix. @item Blocking read MySQL uses a blocking read for each connection. This means that: @itemize @bullet @item A connection will not be disconnected automatically after 8 hours, as happens with the Unix version of MySQL. @item If a connection hangs, it's impossible to break it without killing MySQL. @item @code{mysqladmin kill} will not work on a sleeping connection. @item @code{mysqladmin shutdown} can't abort as long as there are sleeping connections. @end itemize We plan to fix this problem when our Windows developers have figured out a nice workaround. @item UDF functions For the moment, MySQL-Windows does not support user-definable functions. @item @code{DROP DATABASE} You can't drop a database that is in use by some thread. @item Killing MySQL from the task manager You can't kill MySQL from the task manager or with the shutdown utility in Win95. You must take it down with @code{mysqladmin shutdown}. @item Case-insensitive names Filenames are case insensitive on Windows, so database and table names are also case insensitive in MySQL for Windows. The only restriction is that database and table names must be specified using the same case throughout a given statement. @xref{Name case sensitivity}. @item The @samp{\} directory character Pathname components in Win95 are separated by the @samp{\} character, which is also the escape character in MySQL. If you are using @code{LOAD DATA INFILE} or @code{SELECT ... INTO OUTFILE}, you must double the @samp{\} character: @example mysql> LOAD DATA INFILE "C:\\tmp\\skr.txt" INTO TABLE skr; mysql> SELECT * INTO OUTFILE 'C:\\tmp\\skr.txt' FROM skr; @end example Alternatively, use Unix style filenames with @samp{/} characters: @example mysql> LOAD DATA INFILE "C:/tmp/skr.txt" INTO TABLE skr; mysql> SELECT * INTO OUTFILE 'C:/tmp/skr.txt' FROM skr; @end example @item @code{Can't open named pipe} error If you use a MySQL 3.22 version on NT with the newest mysql-clients you will get the following error: @example error 2017: can't open named pipe to host: . pipe... @end example @tindex .my.cnf file This is because the release version of MySQL uses named pipes on NT by default. You can avoid this error by using the @code{--host=localhost} option to the new MySQL clients or create an option file @file{C:\my.cnf} that contains the following information: @example [client] host = localhost @end example Starting from 3.23.50, named pipes are only enabled if mysqld is started with @code{--enable-named-pipe}. @item @code{Access denied for user} error If you get the error @code{Access denied for user: 'some-user@@unknown' to database 'mysql'} when accessing a MySQL server on the same machine, this means that MySQL can't resolve your host name properly. To fix this, you should create a file @file{\windows\hosts} with the following information: @example 127.0.0.1 localhost @end example @item @code{ALTER TABLE} While you are executing an @code{ALTER TABLE} statement, the table is locked from usage by other threads. This has to do with the fact that on Windows, you can't delete a file that is in use by another threads. (In the future, we may find some way to work around this problem.) @item @code{DROP TABLE} on a table that is in use by a @code{MERGE} table will not work The @code{MERGE} handler does its table mapping hidden from MySQL. Because Windows doesn't allow you to drop files that are open, you first must flush all @code{MERGE} tables (with @code{FLUSH TABLES}) or drop the @code{MERGE} table before dropping the table. We will fix this at the same time we introduce @code{VIEW}s. @end table Here are some open issues for anyone who might want to help us with the Windows release: @cindex Windows, open issues @itemize @bullet @item Make a single-user @code{MYSQL.DLL} server. This should include everything in a standard MySQL server, except thread creation. This will make MySQL much easier to use in applications that don't need a true client/server and don't need to access the server from other hosts. @item Add some nice start and shutdown icons to the MySQL installation. @item Create a tool to manage registry entries for the MySQL startup options. The registry entry reading is already coded into @file{mysqld.cc}, but it should be recoded to be more parameter oriented. The tool should also be able to update the @file{C:\my.cnf} option file if the user prefers to use that instead of the registry. @item When registering @code{mysqld} as a service with @code{--install} (on NT) it would be nice if you could also add default options on the command line. For the moment, the workaround is to list the parameters in the @file{C:\my.cnf} file instead. @item It would be real nice to be able to kill @code{mysqld} from the task manager. For the moment, you must use @code{mysqladmin shutdown}. @item Port @code{readline} to Windows for use in the @code{mysql} command line tool. @item GUI versions of the standard MySQL clients (@code{mysql}, @code{mysqlshow}, @code{mysqladmin}, and @code{mysqldump}) would be nice. @item It would be nice if the socket read and write functions in @file{net.c} were interruptible. This would make it possible to kill open threads with @code{mysqladmin kill} on Windows. @item @code{mysqld} always starts in the "C" locale and not in the default locale. We would like to have @code{mysqld} use the current locale for the sort order. @item Implement UDF functions with @code{.DLL}s. @item Add macros to use the faster thread-safe increment/decrement methods provided by Windows. @end itemize Other Windows-specific issues are described in the @file{README} file that comes with the MySQL-Windows distribution. @node Solaris, BSD Notes, Windows, Operating System Specific Notes @subsection Solaris Notes @cindex Solaris installation problems @cindex problems, installing on Solaris @cindex tar, problems on Solaris @cindex errors, directory checksum @cindex checksum errors On Solaris, you may run into trouble even before you get the MySQL distribution unpacked! Solaris @code{tar} can't handle long file names, so you may see an error like this when you unpack MySQL: @example x mysql-3.22.12-beta/bench/Results/ATIS-mysql_odbc-NT_4.0-cmp-db2,informix,ms-sql,mysql,oracle,solid,sybase, 0 bytes, 0 tape blocks tar: directory checksum error @end example In this case, you must use GNU @code{tar} (@code{gtar}) to unpack the distribution. You can find a precompiled copy for Solaris at @uref{http://www.mysql.com/Downloads/}. Sun native threads work only on Solaris 2.5 and higher. For Version 2.4 and earlier, MySQL will automatically use MIT-pthreads. @xref{MIT-pthreads}. If you get the following error from configure: @example checking for restartable system calls... configure: error can not run test programs while cross compiling @end example This means that you have something wrong with your compiler installation! In this case you should upgrade your compiler to a newer version. You may also be able to solve this problem by inserting the following row into the @file{config.cache} file: @example ac_cv_sys_restartable_syscalls=$@{ac_cv_sys_restartable_syscalls='no'@} @end example If you are using Solaris on a SPARC, the recommended compiler is @code{gcc} 2.95.2. You can find this at @uref{http://gcc.gnu.org/}. Note that @code{egcs} 1.1.1 and @code{gcc} 2.8.1 don't work reliably on SPARC! The recommended @code{configure} line when using @code{gcc} 2.95.2 is: @example CC=gcc CFLAGS="-O3" \ CXX=gcc CXXFLAGS="-O3 -felide-constructors -fno-exceptions -fno-rtti" \ ./configure --prefix=/usr/local/mysql --with-low-memory --enable-assembler @end example If you have a ultra sparc, you can get 4 % more performance by adding "-mcpu=v8 -Wa,-xarch=v8plusa" to CFLAGS and CXXFLAGS. If you have Sun Workshop (Fortre) 5.3 (or newer) compiler, you can run @code{configure} like this: @example CC=cc CFLAGS="-Xa -fast -xO4 -native -xstrconst -mt" \ CXX=CC CXXFLAGS="-noex -xO4 -mt" \ ./configure --prefix=/usr/local/mysql --enable-assembler @end example In the MySQL benchmarks, we got a 6 % speedup on an Ultrasparc when using Sun Workshop 5.3 compared to using gcc with -mcpu flags. If you get a problem with @code{fdatasync} or @code{sched_yield}, you can fix this by adding @code{LIBS=-lrt} to the configure line The following paragraph is only relevant for older compilers than WorkShop 5.3: You may also have to edit the @code{configure} script to change this line: @example #if !defined(__STDC__) || __STDC__ != 1 @end example to this: @example #if !defined(__STDC__) @end example If you turn on @code{__STDC__} with the @code{-Xc} option, the Sun compiler can't compile with the Solaris @file{pthread.h} header file. This is a Sun bug (broken compiler or broken include file). If @code{mysqld} issues the error message shown below when you run it, you have tried to compile MySQL with the Sun compiler without enabling the multi-thread option (@code{-mt}): @example libc internal error: _rmutex_unlock: rmutex not held @end example Add @code{-mt} to @code{CFLAGS} and @code{CXXFLAGS} and try again. If you get the following error when compiling MySQL with @code{gcc}, it means that your @code{gcc} is not configured for your version of Solaris: @example shell> gcc -O3 -g -O2 -DDBUG_OFF -o thr_alarm ... ./thr_alarm.c: In function `signal_hand': ./thr_alarm.c:556: too many arguments to function `sigwait' @end example The proper thing to do in this case is to get the newest version of @code{gcc} and compile it with your current @code{gcc} compiler! At least for Solaris 2.5, almost all binary versions of @code{gcc} have old, unusable include files that will break all programs that use threads (and possibly other programs)! Solaris doesn't provide static versions of all system libraries (@code{libpthreads} and @code{libdl}), so you can't compile MySQL with @code{--static}. If you try to do so, you will get the error: @example ld: fatal: library -ldl: not found or undefined reference to `dlopen' or cannot find -lrt @end example If too many processes try to connect very rapidly to @code{mysqld}, you will see this error in the MySQL log: @example Error in accept: Protocol error @end example You might try starting the server with the @code{--set-variable back_log=50} option as a workaround for this. @xref{Command-line options}. If you are linking your own MySQL client, you might get the following error when you try to execute it: @example ld.so.1: ./my: fatal: libmysqlclient.so.#: open failed: No such file or directory @end example The problem can be avoided by one of the following methods: @itemize @bullet @item Link the client with the following flag (instead of @code{-Lpath}): @code{-Wl,r/full-path-to-libmysqlclient.so}. @item Copy @file{libmysqclient.so} to @file{/usr/lib}. @tindex LD_RUN_PATH environment variable @tindex Environment variable, LD_RUN_PATH @item Add the pathname of the directory where @file{libmysqlclient.so} is located to the @code{LD_RUN_PATH} environment variable before running your client. @end itemize When using the @code{--with-libwrap} configure option, you must also include the libraries that @file{libwrap.a} needs: @example --with-libwrap="/opt/NUtcpwrapper-7.6/lib/libwrap.a -lnsl -lsocket @end example If you have problems with configure trying to link with @code{-lz} and you don't have @code{zlib} installed, you have two options: @itemize @bullet @item If you want to be able to use the compressed communication protocol, you need to get and install zlib from ftp.gnu.org. @item Configure with @code{--with-named-z-libs=no}. @end itemize If you are using gcc and have problems with loading @code{UDF} functions into MySQL, try adding @code{-lgcc} to the link line for the @code{UDF} function. If you would like MySQL to start automatically, you can copy @file{support-files/mysql.server} to @file{/etc/init.d} and create a symbolic link to it named @file{/etc/rc3.d/S99mysql.server}. @menu * Solaris 2.7:: Solaris 2.7/2.8 Notes * Solaris x86:: Solaris x86 Notes @end menu @node Solaris 2.7, Solaris x86, Solaris, Solaris @subsubsection Solaris 2.7/2.8 Notes You can normally use a Solaris 2.6 binary on Solaris 2.7 and 2.8. Most of the Solaris 2.6 issues also apply for Solaris 2.7 and 2.8. Note that MySQL Version 3.23.4 and above should be able to autodetect new versions of Solaris and enable workarounds for the following problems! Solaris 2.7 / 2.8 has some bugs in the include files. You may see the following error when you use @code{gcc}: @example /usr/include/widec.h:42: warning: `getwc' redefined /usr/include/wchar.h:326: warning: this is the location of the previous definition @end example If this occurs, you can do the following to fix the problem: Copy @code{/usr/include/widec.h} to @code{.../lib/gcc-lib/os/gcc-version/include} and change line 41 from: @example #if !defined(lint) && !defined(__lint) to #if !defined(lint) && !defined(__lint) && !defined(getwc) @end example Alternatively, you can edit @file{/usr/include/widec.h} directly. Either way, after you make the fix, you should remove @file{config.cache} and run @code{configure} again! If you get errors like this when you run @code{make}, it's because @code{configure} didn't detect the @file{curses.h} file (probably because of the error in @file{/usr/include/widec.h}): @example In file included from mysql.cc:50: /usr/include/term.h:1060: syntax error before `,' /usr/include/term.h:1081: syntax error before `;' @end example The solution to this is to do one of the following: @itemize @bullet @item Configure with @code{CFLAGS=-DHAVE_CURSES_H CXXFLAGS=-DHAVE_CURSES_H ./configure}. @item Edit @file{/usr/include/widec.h} as indicted above and rerun configure. @item Remove the @code{#define HAVE_TERM} line from @file{config.h} file and run @code{make} again. @end itemize If you get a problem that your linker can't find @code{-lz} when linking your client program, the problem is probably that your @file{libz.so} file is installed in @file{/usr/local/lib}. You can fix this by one of the following methods: @itemize @bullet @item Add @file{/usr/local/lib} to @code{LD_LIBRARY_PATH}. @item Add a link to @file{libz.so} from @file{/lib}. @item If you are using Solaris 8, you can install the optional zlib from your Solaris 8 CD distribution. @item Configure MySQL with the @code{--with-named-z-libs=no} option. @end itemize @node Solaris x86, , Solaris 2.7, Solaris @subsubsection Solaris x86 Notes On Solaris 2.8 on x86, @code{mysqld} will core dump if you run 'strip' in. If you are using @code{gcc} or @code{egcs} on Solaris x86 and you experience problems with core dumps under load, you should use the following @code{configure} command: @example CC=gcc CFLAGS="-O3 -fomit-frame-pointer -DHAVE_CURSES_H" \ CXX=gcc \ CXXFLAGS="-O3 -fomit-frame-pointer -felide-constructors -fno-exceptions -fno-rtti -DHAVE_CURSES_H" \ ./configure --prefix=/usr/local/mysql @end example This will avoid problems with the @code{libstdc++} library and with C++ exceptions. If this doesn't help, you should compile a debug version and run it with a trace file or under @code{gdb}. @xref{Using gdb on mysqld}. @node BSD Notes, Mac OS X, Solaris, Operating System Specific Notes @subsection BSD Notes @menu * FreeBSD:: FreeBSD Notes * NetBSD:: NetBSD notes * OpenBSD:: OpenBSD Notes * OpenBSD 2.5:: OpenBSD 2.5 Notes * OpenBSD 2.8:: OpenBSD 2.8 Notes * BSDI:: BSD/OS Notes * BSDI2:: BSD/OS Version 2.x Notes * BSDI3:: BSD/OS Version 3.x Notes * BSDI4:: BSD/OS Version 4.x Notes @end menu @node FreeBSD, NetBSD, BSD Notes, BSD Notes @subsubsection FreeBSD Notes FreeBSD 3.x is recommended for running MySQL since the thread package is much more integrated. The easiest and therefor the preferred way to install is to use the mysql-server and mysql-client ports available on @uref{http://www.freebsd.org}. Using these gives you: @itemize @bullet @item A working MySQL with all optimizations known to work on your version of FreeBSD enabled. @item Automatic configuration and build. @item Startup scripts installed in /usr/local/etc/rc.d. @item Ability to see which files that are installed with pkg_info -L. And to remove them all with pkg_delete if you no longer want MySQL on that machine. @end itemize It is recommended you use MIT-pthreads on FreeBSD 2.x and native threads on Versions 3 and up. It is possible to run with native threads on some late 2.2.x versions but you may encounter problems shutting down @code{mysqld}. The MySQL @file{Makefile}s require GNU make (@code{gmake}) to work. If you want to compile MySQL you need to install GNU @code{make} first. Be sure to have your name resolver setup correct. Otherwise you may experience resolver delays or failures when connecting to @code{mysqld}. Make sure that the @code{localhost} entry in the @file{/etc/hosts} file is correct (otherwise you will have problems connecting to the database). The @file{/etc/hosts} file should start with a line: @example 127.0.0.1 localhost localhost.your.domain @end example If you notice that @code{configure} will use MIT-pthreads, you should read the MIT-pthreads notes. @xref{MIT-pthreads}. If you get an error from @code{make install} that it can't find @file{/usr/include/pthreads}, @code{configure} didn't detect that you need MIT-pthreads. This is fixed by executing these commands: @example shell> rm config.cache shell> ./configure --with-mit-threads @end example FreeBSD is also known to have a very low default file handle limit. @xref{Not enough file handles}. Uncomment the ulimit -n section in safe_mysqld or raise the limits for the @code{mysqld} user in /etc/login.conf (and rebuild it with cap_mkdb /etc/login.conf). Also be sure you set the appropriate class for this user in the password file if you are not using the default (use: chpass mysqld-user-name). @xref{safe_mysqld, , @code{safe_mysqld}}. If you get problems with the current date in MySQL, setting the @code{TZ} variable will probably help. @xref{Environment variables}. To get a secure and stable system you should only use FreeBSD kernels that are marked @code{-STABLE}. @node NetBSD, OpenBSD, FreeBSD, BSD Notes @subsubsection NetBSD notes To compile on NetBSD you need GNU @code{make}. Otherwise the compile will crash when @code{make} tries to run @code{lint} on C++ files. @node OpenBSD, OpenBSD 2.5, NetBSD, BSD Notes @subsubsection OpenBSD Notes @menu * OpenBSD 2.5:: OpenBSD 2.5 Notes * OpenBSD 2.8:: OpenBSD 2.8 Notes @end menu @node OpenBSD 2.5, OpenBSD 2.8, OpenBSD, BSD Notes @subsubsection OpenBSD 2.5 Notes On OpenBSD Version 2.5, you can compile MySQL with native threads with the following options: @example CFLAGS=-pthread CXXFLAGS=-pthread ./configure --with-mit-threads=no @end example @node OpenBSD 2.8, BSDI, OpenBSD 2.5, BSD Notes @subsubsection OpenBSD 2.8 Notes Our users have reported that OpenBSD 2.8 has a threading bug which causes problems with MySQL. The OpenBSD Developers have fixed the problem, but as of January 25th, 2001, it's only available in the ``-current'' branch. The symptoms of this threading bug are: slow response, high load, high CPU usage, and crashes. @node BSDI, BSDI2, OpenBSD 2.8, BSD Notes @subsubsection BSD/OS Notes @menu * BSDI2:: BSD/OS 2.x notes * BSDI3:: BSD/OS 3.x notes * BSDI4:: BSD/OS 4.x notes @end menu @node BSDI2, BSDI3, BSDI, BSD Notes @subsubsection BSD/OS Version 2.x Notes If you get the following error when compiling MySQL, your @code{ulimit} value for virtual memory is too low: @example item_func.h: In method `Item_func_ge::Item_func_ge(const Item_func_ge &)': item_func.h:28: virtual memory exhausted make[2]: *** [item_func.o] Error 1 @end example Try using @code{ulimit -v 80000} and run @code{make} again. If this doesn't work and you are using @code{bash}, try switching to @code{csh} or @code{sh}; some BSDI users have reported problems with @code{bash} and @code{ulimit}. If you are using @code{gcc}, you may also use have to use the @code{--with-low-memory} flag for @code{configure} to be able to compile @file{sql_yacc.cc}. If you get problems with the current date in MySQL, setting the @code{TZ} variable will probably help. @xref{Environment variables}. @node BSDI3, BSDI4, BSDI2, BSD Notes @subsubsection BSD/OS Version 3.x Notes Upgrade to BSD/OS Version 3.1. If that is not possible, install BSDIpatch M300-038. Use the following command when configuring MySQL: @example shell> env CXX=shlicc++ CC=shlicc2 \ ./configure \ --prefix=/usr/local/mysql \ --localstatedir=/var/mysql \ --without-perl \ --with-unix-socket-path=/var/mysql/mysql.sock @end example The following is also known to work: @example shell> env CC=gcc CXX=gcc CXXFLAGS=-O3 \ ./configure \ --prefix=/usr/local/mysql \ --with-unix-socket-path=/var/mysql/mysql.sock @end example You can change the directory locations if you wish, or just use the defaults by not specifying any locations. If you have problems with performance under heavy load, try using the @code{--skip-thread-priority} option to @code{mysqld}! This will run all threads with the same priority; on BSDI Version 3.1, this gives better performance (at least until BSDI fixes their thread scheduler). If you get the error @code{virtual memory exhausted} while compiling, you should try using @code{ulimit -v 80000} and run @code{make} again. If this doesn't work and you are using @code{bash}, try switching to @code{csh} or @code{sh}; some BSDI users have reported problems with @code{bash} and @code{ulimit}. @node BSDI4, , BSDI3, BSD Notes @subsubsection BSD/OS Version 4.x Notes BSDI Version 4.x has some thread-related bugs. If you want to use MySQL on this, you should install all thread-related patches. At least M400-023 should be installed. On some BSDI Version 4.x systems, you may get problems with shared libraries. The symptom is that you can't execute any client programs, for example, @code{mysqladmin}. In this case you need to reconfigure not to use shared libraries with the @code{--disable-shared} option to configure. Some customers have had problems on BSDI 4.0.1 that the @code{mysqld} binary after a while can't open tables. This is because some library/system related bug causes @code{mysqld} to change current directory without asking for this! The fix is to either upgrade to 3.23.34 or after running @code{configure} remove the line @code{#define HAVE_REALPATH} from @code{config.h} before running make. Note that the above means that you can't symbolic link a database directories to another database directory or symbolic link a table to another database on BSDI! (Making a symbolic link to another disk is ok). @node Mac OS X, Other Unix Notes, BSD Notes, Operating System Specific Notes @subsection Mac OS X Notes @menu * Mac OS X Public Beta:: Mac OS X Public Beta * Mac OS X Server:: Mac OS X Server @end menu @node Mac OS X Public Beta, Mac OS X Server, Mac OS X, Mac OS X @subsubsection Mac OS X Public Beta MySQL should work without any problems on Mac OS X Public Beta (Darwin). You don't need the pthread patches for this OS! @node Mac OS X Server, , Mac OS X Public Beta, Mac OS X @subsubsection Mac OS X Server Before trying to configure MySQL on Mac OS X server you must first install the pthread package from @uref{http://www.prnet.de/RegEx/mysql.html}. Our binary for Mac OS X is compiled on Rhapsody 5.5 with the following configure line: @example CC=gcc CFLAGS="-O2 -fomit-frame-pointer" CXX=gcc CXXFLAGS="-O2 -fomit-frame-pointer" ./configure --prefix=/usr/local/mysql "--with-comment=Official MySQL binary" --with-extra-charsets=complex --disable-shared @end example You might want to also add aliases to your shell's resource file to access @code{mysql} and @code{mysqladmin} from the command line: @example alias mysql '/usr/local/mysql/bin/mysql' alias mysqladmin '/usr/local/mysql/bin/mysqladmin' @end example @node Other Unix Notes, OS/2, Mac OS X, Operating System Specific Notes @subsection Other Unix Notes @menu * Binary notes-HP-UX:: HP-UX Notes for Binary Distributions * HP-UX 10.20:: HP-UX Version 10.20 Notes * HP-UX 11.x:: HP-UX Version 11.x Notes * IBM-AIX:: IBM-AIX notes * SunOS:: SunOS 4 Notes * Alpha-DEC-UNIX:: Alpha-DEC-UNIX Notes (Tru64) * Alpha-DEC-OSF1:: Alpha-DEC-OSF1 Notes * SGI-Irix:: SGI Irix Notes * SCO:: SCO Notes * SCO Unixware:: SCO Unixware Version 7.0 Notes @end menu @node Binary notes-HP-UX, HP-UX 10.20, Other Unix Notes, Other Unix Notes @subsubsection HP-UX Notes for Binary Distributions @cindex HP-UX, binary distribution @cindex binary distributions, on HP-UX Some of the binary distributions of MySQL for HP-UX is distributed as an HP depot file and as a tar file. To use the depot file you must be running at least HP-UX 10.x to have access to HP's software depot tools. The HP version of MySQL was compiled on an HP 9000/8xx server under HP-UX 10.20, and uses MIT-pthreads. It is known to work well under this configuration. MySQL Version 3.22.26 and newer can also be built with HP's native thread package. Other configurations that may work: @itemize @bullet @item HP 9000/7xx running HP-UX 10.20+ @item HP 9000/8xx running HP-UX 10.30 @end itemize The following configurations almost definitely won't work: @itemize @bullet @item HP 9000/7xx or 8xx running HP-UX 10.x where x < 2 @item HP 9000/7xx or 8xx running HP-UX 9.x @end itemize To install the distribution, use one of the commands below, where @code{/path/to/depot} is the full pathname of the depot file: @itemize @bullet @item To install everything, including the server, client and development tools: @example shell> /usr/sbin/swinstall -s /path/to/depot mysql.full @end example @item To install only the server: @example shell> /usr/sbin/swinstall -s /path/to/depot mysql.server @end example @item To install only the client package: @example shell> /usr/sbin/swinstall -s /path/to/depot mysql.client @end example @item To install only the development tools: @example shell> /usr/sbin/swinstall -s /path/to/depot mysql.developer @end example @end itemize The depot places binaries and libraries in @file{/opt/mysql} and data in @file{/var/opt/mysql}. The depot also creates the appropriate entries in @file{/etc/init.d} and @file{/etc/rc2.d} to start the server automatically at boot time. Obviously, this entails being @code{root} to install. To install the HP-UX tar.gz distribution, you must have a copy of GNU @code{tar}. @node HP-UX 10.20, HP-UX 11.x, Binary notes-HP-UX, Other Unix Notes @subsubsection HP-UX Version 10.20 Notes There are a couple of small problems when compiling MySQL on HP-UX. We recommend that you use @code{gcc} instead of the HP-UX native compiler, because @code{gcc} produces better code! We recommend using gcc 2.95 on HP-UX. Don't use high optimization flags (like -O6) as this may not be safe on HP-UX. Note that MIT-pthreads can't be compiled with the HP-UX compiler because it can't compile @code{.S} (assembler) files. The following configure line should work: @example CFLAGS="-DHPUX -I/opt/dce/include" CXXFLAGS="-DHPUX -I/opt/dce/include -felide-constructors -fno-exceptions -fno-rtti" CXX=gcc ./configure --with-pthread --with-named-thread-libs='-ldce' --prefix=/usr/local/mysql --disable-shared @end example If you are compiling @code{gcc} 2.95 yourself, you should NOT link it with the DCE libraries (@code{libdce.a} or @code{libcma.a}) if you want to compile MySQL with MIT-pthreads. If you mix the DCE and MIT-pthreads packages you will get a @code{mysqld} to which you cannot connect. Remove the DCE libraries while you compile @code{gcc} 2.95! @node HP-UX 11.x, IBM-AIX, HP-UX 10.20, Other Unix Notes @subsubsection HP-UX Version 11.x Notes For HP-UX Version 11.x we recommend MySQL Version 3.23.15 or later. Because of some critical bugs in the standard HP-UX libraries, you should install the following patches before trying to run MySQL on HP-UX 11.0: @example PHKL_22840 Streams cumulative PHNE_22397 ARPA cumulative @end example This will solve a problem that one gets @code{EWOULDBLOCK} from @code{recv()} and @code{EBADF} from @code{accept()} in threaded applications. If you are using @code{gcc} 2.95.1 on an unpatched HP-UX 11.x system, you will get the error: @example In file included from /usr/include/unistd.h:11, from ../include/global.h:125, from mysql_priv.h:15, from item.cc:19: /usr/include/sys/unistd.h:184: declaration of C function ... /usr/include/sys/pthread.h:440: previous declaration ... In file included from item.h:306, from mysql_priv.h:158, from item.cc:19: @end example The problem is that HP-UX doesn't define @code{pthreads_atfork()} consistently. It has conflicting prototypes in @file{/usr/include/sys/unistd.h}:184 and @file{/usr/include/sys/pthread.h}:440 (details below). One solution is to copy @file{/usr/include/sys/unistd.h} into @file{mysql/include} and edit @file{unistd.h} and change it to match the definition in @file{pthread.h}. Here's the diff: @example 183,184c183,184 < extern int pthread_atfork(void (*prepare)(), void (*parent)(), < void (*child)()); --- > extern int pthread_atfork(void (*prepare)(void), void (*parent)(void), > void (*child)(void)); @end example After this, the following configure line should work: @example CFLAGS="-fomit-frame-pointer -O3 -fpic" CXX=gcc CXXFLAGS="-felide-constructors -fno-exceptions -fno-rtti -O3" ./configure --prefix=/usr/local/mysql --disable-shared @end example Here is some information that a HP-UX Version 11.x user sent us about compiling MySQL with HP-UX:x compiler: @example Environment: proper compilers. setenv CC cc setenv CXX aCC flags setenv CFLAGS -D_REENTRANT setenv CXXFLAGS -D_REENTRANT setenv CPPFLAGS -D_REENTRANT % aCC -V aCC: HP ANSI C++ B3910B X.03.14.06 % cc -V /tmp/empty.c cpp.ansi: HP92453-01 A.11.02.00 HP C Preprocessor (ANSI) ccom: HP92453-01 A.11.01.00 HP C Compiler cc: "/tmp/empty.c", line 1: warning 501: Empty source file. configuration: ./configure --with-pthread \ --prefix=/source-control/mysql \ --with-named-thread-libs=-lpthread \ --with-low-memory added '#define _CTYPE_INCLUDED' to include/m_ctype.h. This symbol is the one defined in HP's /usr/include/ctype.h: /* Don't include std ctype.h when this is included */ #define _CTYPE_H #define __CTYPE_INCLUDED #define _CTYPE_INCLUDED #define _CTYPE_USING /* Don't put names in global namespace. */ @end example @itemize @bullet @item I had to use the compile-time flag @code{-D_REENTRANT} to get the compiler to recognize the prototype for @code{localtime_r}. Alternatively I could have supplied the prototype for @code{localtime_r}. But I wanted to catch other bugs without needing to run into them. I wasn't sure where I needed it, so I added it to all flags. @item The optimization flags used by MySQL (-O3) are not recognized by HP's compilers. I did not change the flags. @end itemize If you get the following error from @code{configure} @example checking for cc option to accept ANSI C... no configure: error: MySQL requires a ANSI C compiler (and a C++ compiler). Try gcc. See the Installation chapter in the Reference Manual. @end example Check that you don't have the path to the K&R compiler before the path to the HP-UX C and C++ compiler. @node IBM-AIX, SunOS, HP-UX 11.x, Other Unix Notes @subsubsection IBM-AIX notes @cindex problems, installing on IBM-AIX Automatic detection of @code{xlC} is missing from Autoconf, so a @code{configure} command something like this is needed when compiling MySQL (This example uses the IBM compiler): @example export CC="xlc_r -ma -O3 -qstrict -qoptimize=3 -qmaxmem=8192 " export CXX="xlC_r -ma -O3 -qstrict -qoptimize=3 -qmaxmem=8192" export CFLAGS="-I /usr/local/include" export LDFLAGS="-L /usr/local/lib" export CPPFLAGS=$CFLAGS export CXXFLAGS=$CFLAGS ./configure --prefix=/usr/local \ --localstatedir=/var/mysql \ --sysconfdir=/etc/mysql \ --sbindir='/usr/local/bin' \ --libexecdir='/usr/local/bin' \ --enable-thread-safe-client \ --enable-large-files @end example Above are the options used to compile the MySQL distribution that can be found at @uref{http://www-frec.bull.com/}. If you change the @code{-O3} to @code{-O2} in the above configure line, you must also remove the @code{-qstrict} option (this is a limitation in the IBM C compiler). If you are using @code{gcc} or @code{egcs} to compile MySQL, you @strong{MUST} use the @code{-fno-exceptions} flag, as the exception handling in @code{gcc}/@code{egcs} is not thread safe! (This is tested with @code{egcs} 1.1.). There are also some known problems with IBM's assembler, which may cause it to generate bad code when used with gcc. We recommend the following @code{configure} line with @code{egcs} and @code{gcc 2.95} on AIX: @example CC="gcc -pipe -mcpu=power -Wa,-many" \ CXX="gcc -pipe -mcpu=power -Wa,-many" \ CXXFLAGS="-felide-constructors -fno-exceptions -fno-rtti" \ ./configure --prefix=/usr/local/mysql --with-low-memory @end example The @code{-Wa,-many} is necessary for the compile to be successful. IBM is aware of this problem but is in to hurry to fix it because of the workaround available. We don't know if the @code{-fno-exceptions} is required with @code{gcc 2.95}, but as MySQL doesn't use exceptions and the above option generates faster code, we recommend that you should always use this option with @code{egcs / gcc}. If you get a problem with assembler code try changing the -mcpu=xxx to match your cpu. Typically power2, power, or powerpc may need to be used, alternatively you might need to use 604 or 604e. I'm not positive but I would think using "power" would likely be safe most of the time, even on a power2 machine. If you don't know what your cpu is then do a "uname -m", this will give you back a string that looks like "000514676700", with a format of xxyyyyyymmss where xx and ss are always 0's, yyyyyy is a unique system id and mm is the id of the CPU Planar. A chart of these values can be found at @uref{http://www.rs6000.ibm.com/doc_link/en_US/a_doc_lib/cmds/aixcmds5/uname.htm}. This will give you a machine type and a machine model you can use to determine what type of cpu you have. If you have problems with signals (MySQL dies unexpectedly under high load) you may have found an OS bug with threads and signals. In this case you can tell MySQL not to use signals by configuring with: @example shell> CFLAGS=-DDONT_USE_THR_ALARM CXX=gcc \ CXXFLAGS="-felide-constructors -fno-exceptions -fno-rtti -DDONT_USE_THR_ALARM" \ ./configure --prefix=/usr/local/mysql --with-debug --with-low-memory @end example This doesn't affect the performance of MySQL, but has the side effect that you can't kill clients that are ``sleeping'' on a connection with @code{mysqladmin kill} or @code{mysqladmin shutdown}. Instead, the client will die when it issues its next command. On some versions of AIX, linking with @code{libbind.a} makes @code{getservbyname} core dump. This is an AIX bug and should be reported to IBM. For AIX 4.2.1 and gcc you have to do the following changes. After configuring, edit @file{config.h} and @file{include/my_config.h} and change the line that says @example #define HAVE_SNPRINTF 1 @end example to @example #undef HAVE_SNPRINTF @end example And finally, in @file{mysqld.cc} you need to add a prototype for initgoups. @example #ifdef _AIX41 extern "C" int initgroups(const char *,int); #endif @end example @node SunOS, Alpha-DEC-UNIX, IBM-AIX, Other Unix Notes @subsubsection SunOS 4 Notes On SunOS 4, MIT-pthreads is needed to compile MySQL, which in turn means you will need GNU @code{make}. Some SunOS 4 systems have problems with dynamic libraries and @code{libtool}. You can use the following @code{configure} line to avoid this problem: @example shell> ./configure --disable-shared --with-mysqld-ldflags=-all-static @end example When compiling @code{readline}, you may get warnings about duplicate defines. These may be ignored. When compiling @code{mysqld}, there will be some @code{implicit declaration of function} warnings. These may be ignored. @node Alpha-DEC-UNIX, Alpha-DEC-OSF1, SunOS, Other Unix Notes @subsubsection Alpha-DEC-UNIX Notes (Tru64) If you are using egcs 1.1.2 on Digital Unix, you should upgrade to gcc 2.95.2, as egcs on DEC has some serious bugs! When compiling threaded programs under Digital Unix, the documentation recommends using the @code{-pthread} option for @code{cc} and @code{cxx} and the libraries @code{-lmach -lexc} (in addition to @code{-lpthread}). You should run @code{configure} something like this: @example CC="cc -pthread" CXX="cxx -pthread -O" \ ./configure --with-named-thread-libs="-lpthread -lmach -lexc -lc" @end example When compiling @code{mysqld}, you may see a couple of warnings like this: @example mysqld.cc: In function void handle_connections()': mysqld.cc:626: passing long unsigned int *' as argument 3 of accept(int,sockadddr *, int *)' @end example You can safely ignore these warnings. They occur because @code{configure} can detect only errors, not warnings. If you start the server directly from the command line, you may have problems with it dying when you log out. (When you log out, your outstanding processes receive a @code{SIGHUP} signal.) If so, try starting the server like this: @example shell> nohup mysqld [options] & @end example @code{nohup} causes the command following it to ignore any @code{SIGHUP} signal sent from the terminal. Alternatively, start the server by running @code{safe_mysqld}, which invokes @code{mysqld} using @code{nohup} for you. @xref{safe_mysqld, , @code{safe_mysqld}}. If you get a problem when compiling mysys/get_opt.c, just remove the line #define _NO_PROTO from the start of that file! If you are using Compac's CC compiler, the following configure line should work: @example CC="cc -pthread" CFLAGS="-O4 -ansi_alias -ansi_args -fast -inline speed all -arch host" CXX="cxx -pthread" CXXFLAGS="-O4 -ansi_alias -ansi_args -fast -inline speed all -arch host" export CC CFLAGS CXX CXXFLAGS ./configure \ --prefix=/usr/local/mysql \ --with-low-memory \ --enable-large-files \ --enable-shared=yes \ --with-named-thread-libs="-lpthread -lmach -lexc -lc" gnumake @end example If you get a problem with libtool, when compiling with shared libraries as above, when linking @code{mysql}, you should be able to get around this by issuing: @example cd mysql /bin/sh ../libtool --mode=link cxx -pthread -O3 -DDBUG_OFF \ -O4 -ansi_alias -ansi_args -fast -inline speed \ -speculate all \ -arch host -DUNDEF_HAVE_GETHOSTBYNAME_R \ -o mysql mysql.o readline.o sql_string.o completion_hash.o \ ../readline/libreadline.a -lcurses \ ../libmysql/.libs/libmysqlclient.so -lm cd .. gnumake gnumake install scripts/mysql_install_db @end example @node Alpha-DEC-OSF1, SGI-Irix, Alpha-DEC-UNIX, Other Unix Notes @subsubsection Alpha-DEC-OSF1 Notes If you have problems compiling and have DEC @code{CC} and @code{gcc} installed, try running @code{configure} like this: @example CC=cc CFLAGS=-O CXX=gcc CXXFLAGS=-O3 \ ./configure --prefix=/usr/local/mysql @end example If you get problems with the @file{c_asm.h} file, you can create and use a 'dummy' @file{c_asm.h} file with: @example touch include/c_asm.h CC=gcc CFLAGS=-I./include \ CXX=gcc CXXFLAGS=-O3 \ ./configure --prefix=/usr/local/mysql @end example Note that the following problems with the @code{ld} program can be fixed by downloading the latest DEC (Compaq) patch kit from: @uref{http://ftp.support.compaq.com/public/unix/}. On OSF1 V4.0D and compiler "DEC C V5.6-071 on Digital Unix V4.0 (Rev. 878)" the compiler had some strange behavior (undefined @code{asm} symbols). @code{/bin/ld} also appears to be broken (problems with @code{_exit undefined} errors occuring while linking @code{mysqld}). On this system, we have managed to compile MySQL with the following @code{configure} line, after replacing @code{/bin/ld} with the version from OSF 4.0C: @example CC=gcc CXX=gcc CXXFLAGS=-O3 ./configure --prefix=/usr/local/mysql @end example With the Digital compiler "C++ V6.1-029", the following should work: @example CC=cc -pthread CFLAGS=-O4 -ansi_alias -ansi_args -fast -inline speed -speculate all -arch host CXX=cxx -pthread CXXFLAGS=-O4 -ansi_alias -ansi_args -fast -inline speed -speculate all -arch host -noexceptions -nortti export CC CFLAGS CXX CXXFLAGS ./configure --prefix=/usr/mysql/mysql --with-mysqld-ldflags=-all-static --disable-shared --with-named-thread-libs="-lmach -lexc -lc" @end example In some versions of OSF1, the @code{alloca()} function is broken. Fix this by removing the line in @file{config.h} that defines @code{'HAVE_ALLOCA'}. The @code{alloca()} function also may have an incorrect prototype in @code{/usr/include/alloca.h}. This warning resulting from this can be ignored. @code{configure} will use the following thread libraries automatically: @code{--with-named-thread-libs="-lpthread -lmach -lexc -lc"}. When using @code{gcc}, you can also try running @code{configure} like this: @example shell> CFLAGS=-D_PTHREAD_USE_D4 CXX=gcc CXXFLAGS=-O3 ./configure .... @end example If you have problems with signals (MySQL dies unexpectedly under high load), you may have found an OS bug with threads and signals. In this case you can tell MySQL not to use signals by configuring with: @example shell> CFLAGS=-DDONT_USE_THR_ALARM \ CXXFLAGS=-DDONT_USE_THR_ALARM \ ./configure ... @end example This doesn't affect the performance of MySQL, but has the side effect that you can't kill clients that are ``sleeping'' on a connection with @code{mysqladmin kill} or @code{mysqladmin shutdown}. Instead, the client will die when it issues its next command. With @code{gcc} 2.95.2, you will probably run into the following compile error: @example sql_acl.cc:1456: Internal compiler error in `scan_region', at except.c:2566 Please submit a full bug report. @end example To fix this you should change to the @code{sql} directory and do a ``cut and paste'' of the last @code{gcc} line, but change @code{-O3} to @code{-O0} (or add @code{-O0} immediately after @code{gcc} if you don't have any @code{-O} option on your compile line.) After this is done you can just change back to the top-level directly and run @code{make} again. @node SGI-Irix, SCO, Alpha-DEC-OSF1, Other Unix Notes @subsubsection SGI Irix Notes If you are using Irix Version 6.5.3 or newer @code{mysqld} will only be able to create threads if you run it as a user with @code{CAP_SCHED_MGT} privileges (like @code{root}) or give the @code{mysqld} server this privilege with the following shell command: @example shell> chcap "CAP_SCHED_MGT+epi" /opt/mysql/libexec/mysqld @end example You may have to undefine some things in @file{config.h} after running @code{configure} and before compiling. In some Irix implementations, the @code{alloca()} function is broken. If the @code{mysqld} server dies on some @code{SELECT} statements, remove the lines from @file{config.h} that define @code{HAVE_ALLOC} and @code{HAVE_ALLOCA_H}. If @code{mysqladmin create} doesn't work, remove the line from @file{config.h} that defines @code{HAVE_READDIR_R}. You may have to remove the @code{HAVE_TERM_H} line as well. SGI recommends that you install all of the patches on this page as a set: http://support.sgi.com/surfzone/patches/patchset/6.2_indigo.rps.html At the very minimum, you should install the latest kernel rollup, the latest @code{rld} rollup, and the latest @code{libc} rollup. You definitely need all the POSIX patches on this page, for pthreads support: @uref{http://support.sgi.com/surfzone/patches/patchset/6.2_posix.rps.html} If you get the something like the following error when compiling @file{mysql.cc}: @example "/usr/include/curses.h", line 82: error(1084): invalid combination of type @end example Type the following in the top-level directory of your MySQL source tree: @example shell> extra/replace bool curses_bool < /usr/include/curses.h > include/curses.h shell> make @end example There have also been reports of scheduling problems. If only one thread is running, things go slow. Avoid this by starting another client. This may lead to a 2-to-10-fold increase in execution speed thereafter for the other thread. This is a poorly understood problem with Irix threads; you may have to improvise to find solutions until this can be fixed. If you are compiling with @code{gcc}, you can use the following @code{configure} command: @example CC=gcc CXX=gcc CXXFLAGS=-O3 \ ./configure --prefix=/usr/local/mysql --enable-thread-safe-client --with-named-thread-libs=-lpthread @end example On Irix 6.5.11 with native Irix C and C++ compilers ver. 7.3.1.2, the following is reported to work @example CC=cc CXX=CC CFLAGS='-O3 -n32 -TARG:platform=IP22 -I/usr/local/include \ -L/usr/local/lib' CXXFLAGS='-O3 -n32 -TARG:platform=IP22 \ -I/usr/local/include -L/usr/local/lib' ./configure --prefix=/usr/local/mysql \ --with-berkeley-db --with-innodb \ --with-libwrap=/usr/local --with-named-curses-libs=/usr/local/lib/libncurses.a @end example @node SCO, SCO Unixware, SGI-Irix, Other Unix Notes @subsubsection SCO Notes The current port is tested only on a ``sco3.2v5.0.4'' and ``sco3.2v5.0.5'' system. There has also been a lot of progress on a port to ``sco 3.2v4.2''. For the moment the recommended compiler on OpenServer is gcc 2.95.2. With this you should be able to compile MySQL with just: @example CC=gcc CXX=gcc ./configure ... (options) @end example @enumerate @item For OpenServer 5.0.X you need to use GDS in Skunkware 95 (95q4c). This is necessary because GNU @code{gcc} 2.7.2 in Skunkware 97 does not have GNU @code{as}. You can also use @code{egcs} 1.1.2 or newer @uref{http://www.egcs.com/}. If you are using @code{egcs} 1.1.2 you have to execute the following command: @example shell> cp -p /usr/include/pthread/stdtypes.h /usr/local/lib/gcc-lib/i386-pc-sco3.2v5.0.5/egcs-2.91.66/include/pthread/ @end example @item You need the port of GCC 2.5.x for this product and the Development system. They are required on this version of SCO Unix. You cannot just use the GCC Dev system. @item You should get the FSU Pthreads package and install it first. This can be found at @uref{http://www.cs.wustl.edu/~schmidt/ACE_wrappers/FSU-threads.tar.gz}. You can also get a precompiled package from @uref{http://www.mysql.com/Downloads/SCO/FSU-threads-3.5c.tar.gz}. @item FSU Pthreads can be compiled with SCO Unix 4.2 with tcpip. Or OpenServer 3.0 or Open Desktop 3.0 (OS 3.0 ODT 3.0), with the SCO Development System installed using a good port of GCC 2.5.x ODT or OS 3.0 you will need a good port of GCC 2.5.x There are a lot of problems without a good port. The port for this product requires the SCO Unix Development system. Without it, you are missing the libraries and the linker that is needed. @item To build FSU Pthreads on your system, do the following: @enumerate a @item Run @code{./configure} in the @file{threads/src} directory and select the SCO OpenServer option. This command copies @file{Makefile.SCO5} to @file{Makefile}. @item Run @code{make}. @item To install in the default @file{/usr/include} directory, login as root, then @code{cd} to the @file{thread/src} directory, and run @code{make install}. @end enumerate @item Remember to use GNU @code{make} when making MySQL. @item If you don't start @code{safe_mysqld} as root, you probably will get only the default 110 open files per process. @code{mysqld} will write a note about this in the log file. @item With SCO 3.2V5.0.5, you should use FSU Pthreads version 3.5c or newer. You should also use gcc 2.95.2 or newer! The following @code{configure} command should work: @example shell> ./configure --prefix=/usr/local/mysql --disable-shared @end example @item With SCO 3.2V4.2, you should use FSU Pthreads version 3.5c or newer. The following @code{configure} command should work: @example shell> CFLAGS="-D_XOPEN_XPG4" CXX=gcc CXXFLAGS="-D_XOPEN_XPG4" \ ./configure \ --prefix=/usr/local/mysql \ --with-named-thread-libs="-lgthreads -lsocket -lgen -lgthreads" \ --with-named-curses-libs="-lcurses" @end example You may get some problems with some include files. In this case, you can find new SCO-specific include files at @uref{http://www.mysql.com/Downloads/SCO/SCO-3.2v4.2-includes.tar.gz}. You should unpack this file in the @file{include} directory of your MySQL source tree. @end enumerate SCO development notes: @itemize @bullet @item MySQL should automatically detect FSU Pthreads and link @code{mysqld} with @code{-lgthreads -lsocket -lgthreads}. @item The SCO development libraries are re-entrant in FSU Pthreads. SCO claims that its libraries' functions are re-entrant, so they must be reentrant with FSU Pthreads. FSU Pthreads on OpenServer tries to use the SCO scheme to make re-entrant libraries. @item FSU Pthreads (at least the version at @uref{http://www.mysql.com/}) comes linked with GNU @code{malloc}. If you encounter problems with memory usage, make sure that @file{gmalloc.o} is included in @file{libgthreads.a} and @file{libgthreads.so}. @item In FSU Pthreads, the following system calls are pthreads-aware: @code{read()}, @code{write()}, @code{getmsg()}, @code{connect()}, @code{accept()}, @code{select()}, and @code{wait()}. @item The CSSA-2001-SCO.35.2 (the patch is listed in custom as erg711905-dscr_remap security patch (ver 2.0.0) breaks FSU threads and makes mysqld instable. You have to remove this one if you want to run mysqld on an OpenServer 5.0.6 machine. @end itemize If you want to install DBI on SCO, you have to edit the @file{Makefile} in DBI-xxx and each subdirectory. Note that the following assumes gcc 2.95.2 or newer: @example OLD: NEW: CC = cc CC = gcc CCCDLFLAGS = -KPIC -W1,-Bexport CCCDLFLAGS = -fpic CCDLFLAGS = -wl,-Bexport CCDLFLAGS = LD = ld LD = gcc -G -fpic LDDLFLAGS = -G -L/usr/local/lib LDDLFLAGS = -L/usr/local/lib LDFLAGS = -belf -L/usr/local/lib LDFLAGS = -L/usr/local/lib LD = ld LD = gcc -G -fpic OPTIMISE = -Od OPTIMISE = -O1 OLD: CCCFLAGS = -belf -dy -w0 -U M_XENIX -DPERL_SCO5 -I/usr/local/include NEW: CCFLAGS = -U M_XENIX -DPERL_SCO5 -I/usr/local/include @end example This is because the Perl dynaloader will not load the @code{DBI} modules if they were compiled with @code{icc} or @code{cc}. Perl works best when compiled with @code{cc}. @node SCO Unixware, , SCO, Other Unix Notes @subsubsection SCO Unixware Version 7.0 Notes You must use a version of MySQL at least as recent as Version 3.22.13 because that version fixes some portability problems under Unixware. We have been able to compile MySQL with the following @code{configure} command on Unixware Version 7.0.1: @example CC=cc CXX=CC ./configure --prefix=/usr/local/mysql @end example If you want to use @code{gcc}, you must use @code{gcc} 2.95.2 or newer. @menu * OS/2:: OS/2 Notes @end menu @node OS/2, BeOS, Other Unix Notes, Operating System Specific Notes @subsection OS/2 Notes MySQL uses quite a few open files. Because of this, you should add something like the following to your @file{CONFIG.SYS} file: @example SET EMXOPT=-c -n -h1024 @end example If you don't do this, you will probably run into the following error: @example File 'xxxx' not found (Errcode: 24) @end example When using MySQL with OS/2 Warp 3, FixPack 29 or above is required. With OS/2 Warp 4, FixPack 4 or above is required. This is a requirement of the Pthreads library. MySQL must be installed in a partition that supports long filenames such as HPFS, FAT32, etc. The @file{INSTALL.CMD} script must be run from OS/2's own @file{CMD.EXE} and may not work with replacement shells such as @file{4OS2.EXE}. The @file{scripts/mysql-install-db} script has been renamed. It is now called @file{install.cmd} and is a REXX script, which will set up the default MySQL security settings and create the WorkPlace Shell icons for MySQL. Dynamic module support is compiled in but not fully tested. Dynamic modules should be compiled using the Pthreads run-time library. @example gcc -Zdll -Zmt -Zcrtdll=pthrdrtl -I../include -I../regex -I.. \ -o example udf_example.cc -L../lib -lmysqlclient udf_example.def mv example.dll example.udf @end example @strong{Note:} Due to limitations in OS/2, UDF module name stems must not exceed 8 characters. Modules are stored in the @file{/mysql2/udf} directory; the @code{safe-mysqld.cmd} script will put this directory in the @code{BEGINLIBPATH} environment variable. When using UDF modules, specified extensions are ignored --- it is assumed to be @file{.udf}. For example, in Unix, the shared module might be named @file{example.so} and you would load a function from it like this: @example mysql> CREATE FUNCTION metaphon RETURNS STRING SONAME "example.so"; @end example Is OS/2, the module would be named @file{example.udf}, but you would not specify the module extension: @example mysql> CREATE FUNCTION metaphon RETURNS STRING SONAME "example"; @end example @node BeOS, Novell Netware, OS/2, Operating System Specific Notes @subsection BeOS Notes We are really interested in getting MySQL to work on BeOS, but unfortunately we don't have any person who knows BeOS or has time to do a port. We are interested in finding someone to do a port, and we will help them with any technical questions they may have while doing the port. We have previously talked with some BeOS developers that have said that MySQL is 80% ported to BeOS, but we haven't heard from them in a while. @node Novell Netware, , BeOS, Operating System Specific Notes @subsection Novell Netware Notes We are really interested in getting MySQL to work on Netware, but unfortunately we don't have any person who knows Netware or has time to do a port. We are interested in finding someone to do a port, and we will help them with any technical questions they may have while doing the port. @node Tutorial, MySQL Database Administration, Installing, Top @chapter Introduction to MySQL: A MySQL Tutorial @cindex tutorial @cindex terminal monitor, defined @cindex monitor, terminal @cindex options, provided by MySQL @menu * Connecting-disconnecting:: Connecting to and disconnecting from the server * Entering queries:: Entering queries * Database use:: Creating and using a database * Getting information:: Getting information about databases and tables * Examples:: Examples * Batch mode:: Using @code{mysql} in batch mode * Twin:: Queries from twin project * Apache:: Using MySQL with Apache @end menu This chapter provides a tutorial introduction to MySQL by showing how to use the @code{mysql} client program to create and use a simple database. @code{mysql} (sometimes referred to as the ``terminal monitor'' or just ``monitor'') is an interactive program that allows you to connect to a MySQL server, run queries, and view the results. @code{mysql} may also be used in batch mode: you place your queries in a file beforehand, then tell @code{mysql} to execute the contents of the file. Both ways of using @code{mysql} are covered here. To see a list of options provided by @code{mysql}, invoke it with the @code{--help} option: @example shell> mysql --help @end example This chapter assumes that @code{mysql} is installed on your machine and that a MySQL server is available to which you can connect. If this is not true, contact your MySQL administrator. (If @emph{you} are the administrator, you will need to consult other sections of this manual.) This chapter describes the entire process of setting up and using a database. If you are interested only in accessing an already-existing database, you may want to skip over the sections that describe how to create the database and the tables it contains. Because this chapter is tutorial in nature, many details are necessarily left out. Consult the relevant sections of the manual for more information on the topics covered here. @node Connecting-disconnecting, Entering queries, Tutorial, Tutorial @section Connecting to and Disconnecting from the Server @cindex connecting, to the server @cindex disconnecting, from the server @cindex server, connecting @cindex server, disconnecting To connect to the server, you'll usually need to provide a MySQL user name when you invoke @code{mysql} and, most likely, a password. If the server runs on a machine other than the one where you log in, you'll also need to specify a hostname. Contact your administrator to find out what connection parameters you should use to connect (that is, what host, user name, and password to use). Once you know the proper parameters, you should be able to connect like this: @example shell> mysql -h host -u user -p Enter password: ******** @end example The @code{********} represents your password; enter it when @code{mysql} displays the @code{Enter password:} prompt. If that works, you should see some introductory information followed by a @code{mysql>} prompt: @example shell> mysql -h host -u user -p Enter password: ******** Welcome to the MySQL monitor. Commands end with ; or \g. Your MySQL connection id is 459 to server version: 3.22.20a-log Type 'help' for help. mysql> @end example The prompt tells you that @code{mysql} is ready for you to enter commands. Some MySQL installations allow users to connect as the anonymous (unnamed) user to the server running on the local host. If this is the case on your machine, you should be able to connect to that server by invoking @code{mysql} without any options: @example shell> mysql @end example After you have connected successfully, you can disconnect any time by typing @code{QUIT} at the @code{mysql>} prompt: @example mysql> QUIT Bye @end example You can also disconnect by pressing Control-D. Most examples in the following sections assume you are connected to the server. They indicate this by the @code{mysql>} prompt. @node Entering queries, Database use, Connecting-disconnecting, Tutorial @section Entering Queries @cindex running, queries @cindex queries, entering @cindex entering, queries Make sure you are connected to the server, as discussed in the previous section. Doing so will not in itself select any database to work with, but that's okay. At this point, it's more important to find out a little about how to issue queries than to jump right in creating tables, loading data into them, and retrieving data from them. This section describes the basic principles of entering commands, using several queries you can try out to familiarize yourself with how @code{mysql} works. Here's a simple command that asks the server to tell you its version number and the current date. Type it in as shown below following the @code{mysql>} prompt and hit the RETURN key: @example mysql> SELECT VERSION(), CURRENT_DATE; +--------------+--------------+ | version() | CURRENT_DATE | +--------------+--------------+ | 3.22.20a-log | 1999-03-19 | +--------------+--------------+ 1 row in set (0.01 sec) mysql> @end example This query illustrates several things about @code{mysql}: @itemize @bullet @item A command normally consists of a SQL statement followed by a semicolon. (There are some exceptions where a semicolon is not needed. @code{QUIT}, mentioned earlier, is one of them. We'll get to others later.) @item When you issue a command, @code{mysql} sends it to the server for execution and displays the results, then prints another @code{mysql>} to indicate that it is ready for another command. @item @code{mysql} displays query output as a table (rows and columns). The first row contains labels for the columns. The rows following are the query results. Normally, column labels are the names of the columns you fetch from database tables. If you're retrieving the value of an expression rather than a table column (as in the example just shown), @code{mysql} labels the column using the expression itself. @item @code{mysql} shows how many rows were returned and how long the query took to execute, which gives you a rough idea of server performance. These values are imprecise because they represent wall clock time (not CPU or machine time), and because they are affected by factors such as server load and network latency. (For brevity, the ``rows in set'' line is not shown in the remaining examples in this chapter.) @end itemize Keywords may be entered in any lettercase. The following queries are equivalent: @example mysql> SELECT VERSION(), CURRENT_DATE; mysql> select version(), current_date; mysql> SeLeCt vErSiOn(), current_DATE; @end example Here's another query. It demonstrates that you can use @code{mysql} as a simple calculator: @example mysql> SELECT SIN(PI()/4), (4+1)*5; +-------------+---------+ | SIN(PI()/4) | (4+1)*5 | +-------------+---------+ | 0.707107 | 25 | +-------------+---------+ @end example The commands shown thus far have been relatively short, single-line statements. You can even enter multiple statements on a single line. Just end each one with a semicolon: @example mysql> SELECT VERSION(); SELECT NOW(); +--------------+ | version() | +--------------+ | 3.22.20a-log | +--------------+ +---------------------+ | NOW() | +---------------------+ | 1999-03-19 00:15:33 | +---------------------+ @end example A command need not be given all on a single line, so lengthy commands that require several lines are not a problem. @code{mysql} determines where your statement ends by looking for the terminating semicolon, not by looking for the end of the input line. (In other words, @code{mysql} accepts free-format input: it collects input lines but does not execute them until it sees the semicolon.) Here's a simple multiple-line statement: @example mysql> SELECT -> USER() -> , -> CURRENT_DATE; +--------------------+--------------+ | USER() | CURRENT_DATE | +--------------------+--------------+ | joesmith@@localhost | 1999-03-18 | +--------------------+--------------+ @end example In this example, notice how the prompt changes from @code{mysql>} to @code{->} after you enter the first line of a multiple-line query. This is how @code{mysql} indicates that it hasn't seen a complete statement and is waiting for the rest. The prompt is your friend, because it provides valuable feedback. If you use that feedback, you will always be aware of what @code{mysql} is waiting for. If you decide you don't want to execute a command that you are in the process of entering, cancel it by typing @code{\c}: @example mysql> SELECT -> USER() -> \c mysql> @end example Here, too, notice the prompt. It switches back to @code{mysql>} after you type @code{\c}, providing feedback to indicate that @code{mysql} is ready for a new command. The following table shows each of the prompts you may see and summarizes what they mean about the state that @code{mysql} is in: @cindex prompts, meanings @multitable @columnfractions .10 .9 @item @strong{Prompt} @tab @strong{Meaning} @item @code{mysql>} @tab Ready for new command. @item @code{@ @ @ @ ->} @tab Waiting for next line of multiple-line command. @item @code{@ @ @ @ '>} @tab Waiting for next line, collecting a string that begins with a single quote (@samp{'}). @item @code{@ @ @ @ ">} @tab Waiting for next line, collecting a string that begins with a double quote (@samp{"}). @end multitable Multiple-line statements commonly occur by accident when you intend to issue a command on a single line, but forget the terminating semicolon. In this case, @code{mysql} waits for more input: @example mysql> SELECT USER() -> @end example If this happens to you (you think you've entered a statement but the only response is a @code{->} prompt), most likely @code{mysql} is waiting for the semicolon. If you don't notice what the prompt is telling you, you might sit there for a while before realizing what you need to do. Enter a semicolon to complete the statement, and @code{mysql} will execute it: @example mysql> SELECT USER() -> ; +--------------------+ | USER() | +--------------------+ | joesmith@@localhost | +--------------------+ @end example The @code{'>} and @code{">} prompts occur during string collection. In MySQL, you can write strings surrounded by either @samp{'} or @samp{"} characters (for example, @code{'hello'} or @code{"goodbye"}), and @code{mysql} lets you enter strings that span multiple lines. When you see a @code{'>} or @code{">} prompt, it means that you've entered a line containing a string that begins with a @samp{'} or @samp{"} quote character, but have not yet entered the matching quote that terminates the string. That's fine if you really are entering a multiple-line string, but how likely is that? Not very. More often, the @code{'>} and @code{">} prompts indicate that you've inadvertantly left out a quote character. For example: @example mysql> SELECT * FROM my_table WHERE name = "Smith AND age < 30; "> @end example If you enter this @code{SELECT} statement, then hit RETURN and wait for the result, nothing will happen. Instead of wondering why this query takes so long, notice the clue provided by the @code{">} prompt. It tells you that @code{mysql} expects to see the rest of an unterminated string. (Do you see the error in the statement? The string @code{"Smith} is missing the second quote.) At this point, what do you do? The simplest thing is to cancel the command. However, you cannot just type @code{\c} in this case, because @code{mysql} interprets it as part of the string that it is collecting! Instead, enter the closing quote character (so @code{mysql} knows you've finished the string), then type @code{\c}: @example mysql> SELECT * FROM my_table WHERE name = "Smith AND age < 30; "> "\c mysql> @end example The prompt changes back to @code{mysql>}, indicating that @code{mysql} is ready for a new command. It's important to know what the @code{'>} and @code{">} prompts signify, because if you mistakenly enter an unterminated string, any further lines you type will appear to be ignored by @code{mysql} --- including a line containing @code{QUIT}! This can be quite confusing, especially if you don't know that you need to supply the terminating quote before you can cancel the current command. @node Database use, Getting information, Entering queries, Tutorial @section Creating and Using a Database @cindex databases, creating @cindex databases, using @cindex creating, databases @menu * Creating database:: Creating a database * Creating tables:: Creating a table * Loading tables:: Loading data into a table * Retrieving data:: Retrieving information from a table @end menu Now that you know how to enter commands, it's time to access a database. Suppose you have several pets in your home (your menagerie) and you'd like to keep track of various types of information about them. You can do so by creating tables to hold your data and loading them with the desired information. Then you can answer different sorts of questions about your animals by retrieving data from the tables. This section shows you how to: @itemize @bullet @item Create a database @item Create a table @item Load data into the table @item Retrieve data from the table in various ways @item Use multiple tables @end itemize The menagerie database will be simple (deliberately), but it is not difficult to think of real-world situations in which a similar type of database might be used. For example, a database like this could be used by a farmer to keep track of livestock, or by a veterinarian to keep track of patient records. A menagerie distribution containing some of the queries and sample data used in the following sections can be obtained from the MySQL Web site. It's available in either @uref{http://www.mysql.com/Downloads/Contrib/Examples/menagerie.tar.gz,compressed @code{tar} format} or @uref{http://www.mysql.com/Downloads/Contrib/Examples/menagerie.zip,Zip format}. Use the @code{SHOW} statement to find out what databases currently exist on the server: @example mysql> SHOW DATABASES; +----------+ | Database | +----------+ | mysql | | test | | tmp | +----------+ @end example The list of databases is probably different on your machine, but the @code{mysql} and @code{test} databases are likely to be among them. The @code{mysql} database is required because it describes user access privileges. The @code{test} database is often provided as a workspace for users to try things out. If the @code{test} database exists, try to access it: @example mysql> USE test Database changed @end example Note that @code{USE}, like @code{QUIT}, does not require a semicolon. (You can terminate such statements with a semicolon if you like; it does no harm.) The @code{USE} statement is special in another way, too: it must be given on a single line. You can use the @code{test} database (if you have access to it) for the examples that follow, but anything you create in that database can be removed by anyone else with access to it. For this reason, you should probably ask your MySQL administrator for permission to use a database of your own. Suppose you want to call yours @code{menagerie}. The administrator needs to execute a command like this: @example mysql> GRANT ALL ON menagerie.* TO your_mysql_name; @end example where @code{your_mysql_name} is the MySQL user name assigned to you. @node Creating database, Creating tables, Database use, Database use @subsection Creating and Selecting a Database @cindex selecting, databases @cindex databases, selecting If the administrator creates your database for you when setting up your permissions, you can begin using it. Otherwise, you need to create it yourself: @example mysql> CREATE DATABASE menagerie; @end example Under Unix, database names are case sensitive (unlike SQL keywords), so you must always refer to your database as @code{menagerie}, not as @code{Menagerie}, @code{MENAGERIE}, or some other variant. This is also true for table names. (Under Windows, this restriction does not apply, although you must refer to databases and tables using the same lettercase throughout a given query.) Creating a database does not select it for use; you must do that explicitly. To make @code{menagerie} the current database, use this command: @example mysql> USE menagerie Database changed @end example Your database needs to be created only once, but you must select it for use each time you begin a @code{mysql} session. You can do this by issuing a @code{USE} statement as shown above. Alternatively, you can select the database on the command line when you invoke @code{mysql}. Just specify its name after any connection parameters that you might need to provide. For example: @example shell> mysql -h host -u user -p menagerie Enter password: ******** @end example Note that @code{menagerie} is not your password on the command just shown. If you want to supply your password on the command line after the @code{-p} option, you must do so with no intervening space (for example, as @code{-pmypassword}, not as @code{-p mypassword}). However, putting your password on the command line is not recommended, because doing so exposes it to snooping by other users logged in on your machine. @node Creating tables, Loading tables, Creating database, Database use @subsection Creating a Table @cindex tables, creating @cindex creating, tables Creating the database is the easy part, but at this point it's empty, as @code{SHOW TABLES} will tell you: @example mysql> SHOW TABLES; Empty set (0.00 sec) @end example The harder part is deciding what the structure of your database should be: what tables you will need and what columns will be in each of them. You'll want a table that contains a record for each of your pets. This can be called the @code{pet} table, and it should contain, as a bare minimum, each animal's name. Because the name by itself is not very interesting, the table should contain other information. For example, if more than one person in your family keeps pets, you might want to list each animal's owner. You might also want to record some basic descriptive information such as species and sex. How about age? That might be of interest, but it's not a good thing to store in a database. Age changes as time passes, which means you'd have to update your records often. Instead, it's better to store a fixed value such as date of birth. Then, whenever you need age, you can calculate it as the difference between the current date and the birth date. MySQL provides functions for doing date arithmetic, so this is not difficult. Storing birth date rather than age has other advantages, too: @itemize @bullet @item You can use the database for tasks such as generating reminders for upcoming pet birthdays. (If you think this type of query is somewhat silly, note that it is the same question you might ask in the context of a business database to identify clients to whom you'll soon need to send out birthday greetings, for that computer-assisted personal touch.) @item You can calculate age in relation to dates other than the current date. For example, if you store death date in the database, you can easily calculate how old a pet was when it died. @end itemize You can probably think of other types of information that would be useful in the @code{pet} table, but the ones identified so far are sufficient for now: name, owner, species, sex, birth, and death. Use a @code{CREATE TABLE} statement to specify the layout of your table: @example mysql> CREATE TABLE pet (name VARCHAR(20), owner VARCHAR(20), -> species VARCHAR(20), sex CHAR(1), birth DATE, death DATE); @end example @code{VARCHAR} is a good choice for the @code{name}, @code{owner}, and @code{species} columns because the column values will vary in length. The lengths of those columns need not all be the same, and need not be @code{20}. You can pick any length from @code{1} to @code{255}, whatever seems most reasonable to you. (If you make a poor choice and it turns out later that you need a longer field, MySQL provides an @code{ALTER TABLE} statement.) Animal sex can be represented in a variety of ways, for example, @code{"m"} and @code{"f"}, or perhaps @code{"male"} and @code{"female"}. It's simplest to use the single characters @code{"m"} and @code{"f"}. The use of the @code{DATE} data type for the @code{birth} and @code{death} columns is a fairly obvious choice. Now that you have created a table, @code{SHOW TABLES} should produce some output: @example mysql> SHOW TABLES; +---------------------+ | Tables in menagerie | +---------------------+ | pet | +---------------------+ @end example To verify that your table was created the way you expected, use a @code{DESCRIBE} statement: @example mysql> DESCRIBE pet; +---------+-------------+------+-----+---------+-------+ | Field | Type | Null | Key | Default | Extra | +---------+-------------+------+-----+---------+-------+ | name | varchar(20) | YES | | NULL | | | owner | varchar(20) | YES | | NULL | | | species | varchar(20) | YES | | NULL | | | sex | char(1) | YES | | NULL | | | birth | date | YES | | NULL | | | death | date | YES | | NULL | | +---------+-------------+------+-----+---------+-------+ @end example You can use @code{DESCRIBE} any time, for example, if you forget the names of the columns in your table or what types they are. @node Loading tables, Retrieving data, Creating tables, Database use @subsection Loading Data into a Table @cindex loading, tables @cindex tables, loading data @cindex data, loading into tables After creating your table, you need to populate it. The @code{LOAD DATA} and @code{INSERT} statements are useful for this. Suppose your pet records can be described as shown below. (Observe that MySQL expects dates in @code{YYYY-MM-DD} format; this may be different than what you are used to.) @multitable @columnfractions .16 .16 .16 .16 .16 .16 @item @strong{name} @tab @strong{owner} @tab @strong{species} @tab @strong{sex} @tab @strong{birth} @tab @strong{death} @item Fluffy @tab Harold @tab cat @tab f @tab 1993-02-04 @tab @item Claws @tab Gwen @tab cat @tab m @tab 1994-03-17 @tab @item Buffy @tab Harold @tab dog @tab f @tab 1989-05-13 @tab @item Fang @tab Benny @tab dog @tab m @tab 1990-08-27 @tab @item Bowser @tab Diane @tab dog @tab m @tab 1998-08-31 @tab 1995-07-29 @item Chirpy @tab Gwen @tab bird @tab f @tab 1998-09-11 @tab @item Whistler @tab Gwen @tab bird @tab @tab 1997-12-09 @tab @item Slim @tab Benny @tab snake @tab m @tab 1996-04-29 @tab @end multitable Because you are beginning with an empty table, an easy way to populate it is to create a text file containing a row for each of your animals, then load the contents of the file into the table with a single statement. You could create a text file @file{pet.txt} containing one record per line, with values separated by tabs, and given in the order in which the columns were listed in the @code{CREATE TABLE} statement. For missing values (such as unknown sexes or death dates for animals that are still living), you can use @code{NULL} values. To represent these in your text file, use @code{\N}. For example, the record for Whistler the bird would look like this (where the whitespace between values is a single tab character): @multitable @columnfractions .15 .15 .15 .15 .25 .15 @item @code{Whistler} @tab @code{Gwen} @tab @code{bird} @tab @code{\N} @tab @code{1997-12-09} @tab @code{\N} @end multitable To load the text file @file{pet.txt} into the @code{pet} table, use this command: @example mysql> LOAD DATA LOCAL INFILE "pet.txt" INTO TABLE pet; @end example You can specify the column value separator and end of line marker explicitly in the @code{LOAD DATA} statement if you wish, but the defaults are tab and linefeed. These are sufficient for the statement to read the file @file{pet.txt} properly. When you want to add new records one at a time, the @code{INSERT} statement is useful. In its simplest form, you supply values for each column, in the order in which the columns were listed in the @code{CREATE TABLE} statement. Suppose Diane gets a new hamster named Puffball. You could add a new record using an @code{INSERT} statement like this: @example mysql> INSERT INTO pet -> VALUES ('Puffball','Diane','hamster','f','1999-03-30',NULL); @end example Note that string and date values are specified as quoted strings here. Also, with @code{INSERT}, you can insert @code{NULL} directly to represent a missing value. You do not use @code{\N} like you do with @code{LOAD DATA}. From this example, you should be able to see that there would be a lot more typing involved to load your records initially using several @code{INSERT} statements rather than a single @code{LOAD DATA} statement. @node Retrieving data, , Loading tables, Database use @subsection Retrieving Information from a Table @cindex data, retrieving @cindex tables, retrieving data @cindex retrieving, data from tables @cindex unloading, tables @menu * Selecting all:: Selecting all data * Selecting rows:: Selecting particular rows * Selecting columns:: Selecting particular columns * Sorting rows:: Sorting rows * Date calculations:: Date calculations * Working with NULL:: Working with @code{NULL} values * Pattern matching:: Pattern matching * Counting rows:: Counting rows * Multiple tables:: Using More Than one Table @end menu The @code{SELECT} statement is used to pull information from a table. The general form of the statement is: @example SELECT what_to_select FROM which_table WHERE conditions_to_satisfy @end example @code{what_to_select} indicates what you want to see. This can be a list of columns, or @code{*} to indicate ``all columns.'' @code{which_table} indicates the table from which you want to retrieve data. The @code{WHERE} clause is optional. If it's present, @code{conditions_to_satisfy} specifies conditions that rows must satisfy to qualify for retrieval. @node Selecting all, Selecting rows, Retrieving data, Retrieving data @subsubsection Selecting All Data The simplest form of @code{SELECT} retrieves everything from a table: @example mysql> SELECT * FROM pet; +----------+--------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +----------+--------+---------+------+------------+------------+ | Fluffy | Harold | cat | f | 1993-02-04 | NULL | | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | | Fang | Benny | dog | m | 1990-08-27 | NULL | | Bowser | Diane | dog | m | 1998-08-31 | 1995-07-29 | | Chirpy | Gwen | bird | f | 1998-09-11 | NULL | | Whistler | Gwen | bird | NULL | 1997-12-09 | NULL | | Slim | Benny | snake | m | 1996-04-29 | NULL | | Puffball | Diane | hamster | f | 1999-03-30 | NULL | +----------+--------+---------+------+------------+------------+ @end example This form of @code{SELECT} is useful if you want to review your entire table, for instance, after you've just loaded it with your initial dataset. As it happens, the output just shown reveals an error in your data file: Bowser appears to have been born after he died! Consulting your original pedigree papers, you find that the correct birth year is 1989, not 1998. There are are least a couple of ways to fix this: @itemize @bullet @item Edit the file @file{pet.txt} to correct the error, then empty the table and reload it using @code{DELETE} and @code{LOAD DATA}: @example mysql> SET AUTOCOMMIT=1; # Used for quick re-create of the table mysql> DELETE FROM pet; mysql> LOAD DATA LOCAL INFILE "pet.txt" INTO TABLE pet; @end example However, if you do this, you must also re-enter the record for Puffball. @item Fix only the erroneous record with an @code{UPDATE} statement: @example mysql> UPDATE pet SET birth = "1989-08-31" WHERE name = "Bowser"; @end example @end itemize As shown above, it is easy to retrieve an entire table. But typically you don't want to do that, particularly when the table becomes large. Instead, you're usually more interested in answering a particular question, in which case you specify some constraints on the information you want. Let's look at some selection queries in terms of questions about your pets that they answer. @node Selecting rows, Selecting columns, Selecting all, Retrieving data @subsubsection Selecting Particular Rows @cindex rows, selecting @cindex tables, selecting rows You can select only particular rows from your table. For example, if you want to verify the change that you made to Bowser's birth date, select Bowser's record like this: @example mysql> SELECT * FROM pet WHERE name = "Bowser"; +--------+-------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +--------+-------+---------+------+------------+------------+ | Bowser | Diane | dog | m | 1989-08-31 | 1995-07-29 | +--------+-------+---------+------+------------+------------+ @end example The output confirms that the year is correctly recorded now as 1989, not 1998. String comparisons are normally case insensitive, so you can specify the name as @code{"bowser"}, @code{"BOWSER"}, etc. The query result will be the same. You can specify conditions on any column, not just @code{name}. For example, if you want to know which animals were born after 1998, test the @code{birth} column: @example mysql> SELECT * FROM pet WHERE birth >= "1998-1-1"; +----------+-------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +----------+-------+---------+------+------------+-------+ | Chirpy | Gwen | bird | f | 1998-09-11 | NULL | | Puffball | Diane | hamster | f | 1999-03-30 | NULL | +----------+-------+---------+------+------------+-------+ @end example You can combine conditions, for example, to locate female dogs: @example mysql> SELECT * FROM pet WHERE species = "dog" AND sex = "f"; +-------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +-------+--------+---------+------+------------+-------+ | Buffy | Harold | dog | f | 1989-05-13 | NULL | +-------+--------+---------+------+------------+-------+ @end example The preceding query uses the @code{AND} logical operator. There is also an @code{OR} operator: @example mysql> SELECT * FROM pet WHERE species = "snake" OR species = "bird"; +----------+-------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +----------+-------+---------+------+------------+-------+ | Chirpy | Gwen | bird | f | 1998-09-11 | NULL | | Whistler | Gwen | bird | NULL | 1997-12-09 | NULL | | Slim | Benny | snake | m | 1996-04-29 | NULL | +----------+-------+---------+------+------------+-------+ @end example @code{AND} and @code{OR} may be intermixed. If you do that, it's a good idea to use parentheses to indicate how conditions should be grouped: @example mysql> SELECT * FROM pet WHERE (species = "cat" AND sex = "m") -> OR (species = "dog" AND sex = "f"); +-------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +-------+--------+---------+------+------------+-------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +-------+--------+---------+------+------------+-------+ @end example @node Selecting columns, Sorting rows, Selecting rows, Retrieving data @subsubsection Selecting Particular Columns @cindex columns, selecting @cindex tables, selecting columns If you don't want to see entire rows from your table, just name the columns in which you're interested, separated by commas. For example, if you want to know when your animals were born, select the @code{name} and @code{birth} columns: @example mysql> SELECT name, birth FROM pet; +----------+------------+ | name | birth | +----------+------------+ | Fluffy | 1993-02-04 | | Claws | 1994-03-17 | | Buffy | 1989-05-13 | | Fang | 1990-08-27 | | Bowser | 1989-08-31 | | Chirpy | 1998-09-11 | | Whistler | 1997-12-09 | | Slim | 1996-04-29 | | Puffball | 1999-03-30 | +----------+------------+ @end example To find out who owns pets, use this query: @example mysql> SELECT owner FROM pet; +--------+ | owner | +--------+ | Harold | | Gwen | | Harold | | Benny | | Diane | | Gwen | | Gwen | | Benny | | Diane | +--------+ @end example @findex DISTINCT However, notice that the query simply retrieves the @code{owner} field from each record, and some of them appear more than once. To minimize the output, retrieve each unique output record just once by adding the keyword @code{DISTINCT}: @example mysql> SELECT DISTINCT owner FROM pet; +--------+ | owner | +--------+ | Benny | | Diane | | Gwen | | Harold | +--------+ @end example You can use a @code{WHERE} clause to combine row selection with column selection. For example, to get birth dates for dogs and cats only, use this query: @example mysql> SELECT name, species, birth FROM pet -> WHERE species = "dog" OR species = "cat"; +--------+---------+------------+ | name | species | birth | +--------+---------+------------+ | Fluffy | cat | 1993-02-04 | | Claws | cat | 1994-03-17 | | Buffy | dog | 1989-05-13 | | Fang | dog | 1990-08-27 | | Bowser | dog | 1989-08-31 | +--------+---------+------------+ @end example @node Sorting rows, Date calculations, Selecting columns, Retrieving data @subsubsection Sorting Rows @cindex rows, sorting @cindex sorting, table rows @cindex sorting, data @cindex tables, sorting rows You may have noticed in the preceding examples that the result rows are displayed in no particular order. However, it's often easier to examine query output when the rows are sorted in some meaningful way. To sort a result, use an @code{ORDER BY} clause. Here are animal birthdays, sorted by date: @example mysql> SELECT name, birth FROM pet ORDER BY birth; +----------+------------+ | name | birth | +----------+------------+ | Buffy | 1989-05-13 | | Bowser | 1989-08-31 | | Fang | 1990-08-27 | | Fluffy | 1993-02-04 | | Claws | 1994-03-17 | | Slim | 1996-04-29 | | Whistler | 1997-12-09 | | Chirpy | 1998-09-11 | | Puffball | 1999-03-30 | +----------+------------+ @end example To sort in reverse order, add the @code{DESC} (descending) keyword to the name of the column you are sorting by: @example mysql> SELECT name, birth FROM pet ORDER BY birth DESC; +----------+------------+ | name | birth | +----------+------------+ | Puffball | 1999-03-30 | | Chirpy | 1998-09-11 | | Whistler | 1997-12-09 | | Slim | 1996-04-29 | | Claws | 1994-03-17 | | Fluffy | 1993-02-04 | | Fang | 1990-08-27 | | Bowser | 1989-08-31 | | Buffy | 1989-05-13 | +----------+------------+ @end example You can sort on multiple columns. For example, to sort by type of animal, then by birth date within animal type with youngest animals first, use the following query: @example mysql> SELECT name, species, birth FROM pet ORDER BY species, birth DESC; +----------+---------+------------+ | name | species | birth | +----------+---------+------------+ | Chirpy | bird | 1998-09-11 | | Whistler | bird | 1997-12-09 | | Claws | cat | 1994-03-17 | | Fluffy | cat | 1993-02-04 | | Fang | dog | 1990-08-27 | | Bowser | dog | 1989-08-31 | | Buffy | dog | 1989-05-13 | | Puffball | hamster | 1999-03-30 | | Slim | snake | 1996-04-29 | +----------+---------+------------+ @end example Note that the @code{DESC} keyword applies only to the column name immediately preceding it (@code{birth}); @code{species} values are still sorted in ascending order. @node Date calculations, Working with NULL, Sorting rows, Retrieving data @subsubsection Date Calculations @cindex date calculations @cindex calculating, dates @cindex extracting, dates @cindex age, calculating MySQL provides several functions that you can use to perform calculations on dates, for example, to calculate ages or extract parts of dates. To determine how many years old each of your pets is, compute the difference in the year part of the current date and the birth date, then subtract one if the current date occurs earlier in the calendar year than the birth date. The following query shows, for each pet, the birth date, the current date, and the age in years. @example mysql> SELECT name, birth, CURRENT_DATE, -> (YEAR(CURRENT_DATE)-YEAR(birth)) -> - (RIGHT(CURRENT_DATE,5)<RIGHT(birth,5)) -> AS age -> FROM pet; +----------+------------+--------------+------+ | name | birth | CURRENT_DATE | age | +----------+------------+--------------+------+ | Fluffy | 1993-02-04 | 2001-08-29 | 8 | | Claws | 1994-03-17 | 2001-08-29 | 7 | | Buffy | 1989-05-13 | 2001-08-29 | 12 | | Fang | 1990-08-27 | 2001-08-29 | 11 | | Bowser | 1989-08-31 | 2001-08-29 | 11 | | Chirpy | 1998-09-11 | 2001-08-29 | 2 | | Whistler | 1997-12-09 | 2001-08-29 | 3 | | Slim | 1996-04-29 | 2001-08-29 | 5 | | Puffball | 1999-03-30 | 2001-08-29 | 2 | +----------+------------+--------------+------+ @end example Here, @code{YEAR()} pulls out the year part of a date and @code{RIGHT()} pulls off the rightmost five characters that represent the @code{MM-DD} (calendar year) part of the date. The part of the expression that compares the @code{MM-DD} values evaluates to 1 or 0, which adjusts the year difference down a year if @code{CURRENT_DATE} occurs earlier in the year than @code{birth}. The full expression is somewhat ungainly, so an alias (@code{age}) is used to make the output column label more meaningful. The query works, but the result could be scanned more easily if the rows were presented in some order. This can be done by adding an @code{ORDER BY name} clause to sort the output by name: @example mysql> SELECT name, birth, CURRENT_DATE, -> (YEAR(CURRENT_DATE)-YEAR(birth)) -> - (RIGHT(CURRENT_DATE,5)<RIGHT(birth,5)) -> AS age -> FROM pet ORDER BY name; +----------+------------+--------------+------+ | name | birth | CURRENT_DATE | age | +----------+------------+--------------+------+ | Bowser | 1989-08-31 | 2001-08-29 | 11 | | Buffy | 1989-05-13 | 2001-08-29 | 12 | | Chirpy | 1998-09-11 | 2001-08-29 | 2 | | Claws | 1994-03-17 | 2001-08-29 | 7 | | Fang | 1990-08-27 | 2001-08-29 | 11 | | Fluffy | 1993-02-04 | 2001-08-29 | 8 | | Puffball | 1999-03-30 | 2001-08-29 | 2 | | Slim | 1996-04-29 | 2001-08-29 | 5 | | Whistler | 1997-12-09 | 2001-08-29 | 3 | +----------+------------+--------------+------+ @end example To sort the output by @code{age} rather than @code{name}, just use a different @code{ORDER BY} clause: @example mysql> SELECT name, birth, CURRENT_DATE, -> (YEAR(CURRENT_DATE)-YEAR(birth)) -> - (RIGHT(CURRENT_DATE,5)<RIGHT(birth,5)) -> AS age -> FROM pet ORDER BY age; +----------+------------+--------------+------+ | name | birth | CURRENT_DATE | age | +----------+------------+--------------+------+ | Chirpy | 1998-09-11 | 2001-08-29 | 2 | | Puffball | 1999-03-30 | 2001-08-29 | 2 | | Whistler | 1997-12-09 | 2001-08-29 | 3 | | Slim | 1996-04-29 | 2001-08-29 | 5 | | Claws | 1994-03-17 | 2001-08-29 | 7 | | Fluffy | 1993-02-04 | 2001-08-29 | 8 | | Fang | 1990-08-27 | 2001-08-29 | 11 | | Bowser | 1989-08-31 | 2001-08-29 | 11 | | Buffy | 1989-05-13 | 2001-08-29 | 12 | +----------+------------+--------------+------+ @end example A similar query can be used to determine age at death for animals that have died. You determine which animals these are by checking whether or not the @code{death} value is @code{NULL}. Then, for those with non-@code{NULL} values, compute the difference between the @code{death} and @code{birth} values: @example mysql> SELECT name, birth, death, -> (YEAR(death)-YEAR(birth)) - (RIGHT(death,5)<RIGHT(birth,5)) -> AS age -> FROM pet WHERE death IS NOT NULL ORDER BY age; +--------+------------+------------+------+ | name | birth | death | age | +--------+------------+------------+------+ | Bowser | 1989-08-31 | 1995-07-29 | 5 | +--------+------------+------------+------+ @end example The query uses @code{death IS NOT NULL} rather than @code{death != NULL} because @code{NULL} is a special value. This is explained later. @xref{Working with NULL, , Working with @code{NULL}}. What if you want to know which animals have birthdays next month? For this type of calculation, year and day are irrelevant; you simply want to extract the month part of the @code{birth} column. MySQL provides several date-part extraction functions, such as @code{YEAR()}, @code{MONTH()}, and @code{DAYOFMONTH()}. @code{MONTH()} is the appropriate function here. To see how it works, run a simple query that displays the value of both @code{birth} and @code{MONTH(birth)}: @example mysql> SELECT name, birth, MONTH(birth) FROM pet; +----------+------------+--------------+ | name | birth | MONTH(birth) | +----------+------------+--------------+ | Fluffy | 1993-02-04 | 2 | | Claws | 1994-03-17 | 3 | | Buffy | 1989-05-13 | 5 | | Fang | 1990-08-27 | 8 | | Bowser | 1989-08-31 | 8 | | Chirpy | 1998-09-11 | 9 | | Whistler | 1997-12-09 | 12 | | Slim | 1996-04-29 | 4 | | Puffball | 1999-03-30 | 3 | +----------+------------+--------------+ @end example Finding animals with birthdays in the upcoming month is easy, too. Suppose the current month is April. Then the month value is @code{4} and you look for animals born in May (month 5) like this: @example mysql> SELECT name, birth FROM pet WHERE MONTH(birth) = 5; +-------+------------+ | name | birth | +-------+------------+ | Buffy | 1989-05-13 | +-------+------------+ @end example There is a small complication if the current month is December, of course. You don't just add one to the month number (@code{12}) and look for animals born in month 13, because there is no such month. Instead, you look for animals born in January (month 1). You can even write the query so that it works no matter what the current month is. That way you don't have to use a particular month number in the query. @code{DATE_ADD()} allows you to add a time interval to a given date. If you add a month to the value of @code{NOW()}, then extract the month part with @code{MONTH()}, the result produces the month in which to look for birthdays: @example mysql> SELECT name, birth FROM pet -> WHERE MONTH(birth) = MONTH(DATE_ADD(NOW(), INTERVAL 1 MONTH)); @end example A different way to accomplish the same task is to add @code{1} to get the next month after the current one (after using the modulo function (@code{MOD}) to wrap around the month value to @code{0} if it is currently @code{12}): @example mysql> SELECT name, birth FROM pet -> WHERE MONTH(birth) = MOD(MONTH(NOW()), 12) + 1; @end example Note that @code{MONTH} returns a number between 1 and 12. And @code{MOD(something,12)} returns a number between 0 and 11. So the addition has to be after the @code{MOD()}, otherwise we would go from November (11) to January (1). @node Working with NULL, Pattern matching, Date calculations, Retrieving data @subsubsection Working with @code{NULL} Values @findex NULL @cindex NULL value The @code{NULL} value can be surprising until you get used to it. Conceptually, @code{NULL} means missing value or unknown value and it is treated somewhat differently than other values. To test for @code{NULL}, you cannot use the arithmetic comparison operators such as @code{=}, @code{<}, or @code{!=}. To demonstrate this for yourself, try the following query: @example mysql> SELECT 1 = NULL, 1 != NULL, 1 < NULL, 1 > NULL; +----------+-----------+----------+----------+ | 1 = NULL | 1 != NULL | 1 < NULL | 1 > NULL | +----------+-----------+----------+----------+ | NULL | NULL | NULL | NULL | +----------+-----------+----------+----------+ @end example Clearly you get no meaningful results from these comparisons. Use the @code{IS NULL} and @code{IS NOT NULL} operators instead: @example mysql> SELECT 1 IS NULL, 1 IS NOT NULL; +-----------+---------------+ | 1 IS NULL | 1 IS NOT NULL | +-----------+---------------+ | 0 | 1 | +-----------+---------------+ @end example In MySQL, 0 or @code{NULL} means false and anything else means true. The default truth value from a boolean operation is 1. This special treatment of @code{NULL} is why, in the previous section, it was necessary to determine which animals are no longer alive using @code{death IS NOT NULL} instead of @code{death != NULL}. @node Pattern matching, Counting rows, Working with NULL, Retrieving data @subsubsection Pattern Matching @cindex pattern matching @cindex matching, patterns @cindex expressions, extended MySQL provides standard SQL pattern matching as well as a form of pattern matching based on extended regular expressions similar to those used by Unix utilities such as @code{vi}, @code{grep}, and @code{sed}. SQL pattern matching allows you to use @samp{_} to match any single character and @samp{%} to match an arbitrary number of characters (including zero characters). In MySQL, SQL patterns are case insensitive by default. Some examples are shown below. Note that you do not use @code{=} or @code{!=} when you use SQL patterns; use the @code{LIKE} or @code{NOT LIKE} comparison operators instead. To find names beginning with @samp{b}: @example mysql> SELECT * FROM pet WHERE name LIKE "b%"; +--------+--------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +--------+--------+---------+------+------------+------------+ | Buffy | Harold | dog | f | 1989-05-13 | NULL | | Bowser | Diane | dog | m | 1989-08-31 | 1995-07-29 | +--------+--------+---------+------+------------+------------+ @end example To find names ending with @samp{fy}: @example mysql> SELECT * FROM pet WHERE name LIKE "%fy"; +--------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +--------+--------+---------+------+------------+-------+ | Fluffy | Harold | cat | f | 1993-02-04 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +--------+--------+---------+------+------------+-------+ @end example To find names containing a @samp{w}: @example mysql> SELECT * FROM pet WHERE name LIKE "%w%"; +----------+-------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +----------+-------+---------+------+------------+------------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Bowser | Diane | dog | m | 1989-08-31 | 1995-07-29 | | Whistler | Gwen | bird | NULL | 1997-12-09 | NULL | +----------+-------+---------+------+------------+------------+ @end example To find names containing exactly five characters, use the @samp{_} pattern character: @example mysql> SELECT * FROM pet WHERE name LIKE "_____"; +-------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +-------+--------+---------+------+------------+-------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +-------+--------+---------+------+------------+-------+ @end example The other type of pattern matching provided by MySQL uses extended regular expressions. When you test for a match for this type of pattern, use the @code{REGEXP} and @code{NOT REGEXP} operators (or @code{RLIKE} and @code{NOT RLIKE}, which are synonyms). Some characteristics of extended regular expressions are: @itemize @bullet @item @samp{.} matches any single character. @item A character class @samp{[...]} matches any character within the brackets. For example, @samp{[abc]} matches @samp{a}, @samp{b}, or @samp{c}. To name a range of characters, use a dash. @samp{[a-z]} matches any lowercase letter, whereas @samp{[0-9]} matches any digit. @item @samp{*} matches zero or more instances of the thing preceding it. For example, @samp{x*} matches any number of @samp{x} characters, @samp{[0-9]*} matches any number of digits, and @samp{.*} matches any number of anything. @item Regular expressions are case sensitive, but you can use a character class to match both lettercases if you wish. For example, @samp{[aA]} matches lowercase or uppercase @samp{a} and @samp{[a-zA-Z]} matches any letter in either case. @item The pattern matches if it occurs anywhere in the value being tested. (SQL patterns match only if they match the entire value.) @item To anchor a pattern so that it must match the beginning or end of the value being tested, use @samp{^} at the beginning or @samp{$} at the end of the pattern. @end itemize To demonstrate how extended regular expressions work, the @code{LIKE} queries shown above are rewritten below to use @code{REGEXP}. To find names beginning with @samp{b}, use @samp{^} to match the beginning of the name: @example mysql> SELECT * FROM pet WHERE name REGEXP "^b"; +--------+--------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +--------+--------+---------+------+------------+------------+ | Buffy | Harold | dog | f | 1989-05-13 | NULL | | Bowser | Diane | dog | m | 1989-08-31 | 1995-07-29 | +--------+--------+---------+------+------------+------------+ @end example Prior to MySQL Version 3.23.4, @code{REGEXP} is case sensitive, and the previous query will return no rows. To match either lowercase or uppercase @samp{b}, use this query instead: @example mysql> SELECT * FROM pet WHERE name REGEXP "^[bB]"; @end example From MySQL 3.23.4 on, to force a @code{REGEXP} comparison to be case sensitive, use the @code{BINARY} keyword to make one of the strings a binary string. This query will match only lowercase @samp{b} at the beginning of a name: @example mysql> SELECT * FROM pet WHERE name REGEXP BINARY "^b"; @end example To find names ending with @samp{fy}, use @samp{$} to match the end of the name: @example mysql> SELECT * FROM pet WHERE name REGEXP "fy$"; +--------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +--------+--------+---------+------+------------+-------+ | Fluffy | Harold | cat | f | 1993-02-04 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +--------+--------+---------+------+------------+-------+ @end example To find names containing a lowercase or uppercase @samp{w}, use this query: @example mysql> SELECT * FROM pet WHERE name REGEXP "w"; +----------+-------+---------+------+------------+------------+ | name | owner | species | sex | birth | death | +----------+-------+---------+------+------------+------------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Bowser | Diane | dog | m | 1989-08-31 | 1995-07-29 | | Whistler | Gwen | bird | NULL | 1997-12-09 | NULL | +----------+-------+---------+------+------------+------------+ @end example Because a regular expression pattern matches if it occurs anywhere in the value, it is not necessary in the previous query to put a wild card on either side of the pattern to get it to match the entire value like it would be if you used a SQL pattern. To find names containing exactly five characters, use @samp{^} and @samp{$} to match the beginning and end of the name, and five instances of @samp{.} in between: @example mysql> SELECT * FROM pet WHERE name REGEXP "^.....$"; +-------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +-------+--------+---------+------+------------+-------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +-------+--------+---------+------+------------+-------+ @end example You could also write the previous query using the @samp{@{n@}} ``repeat-@code{n}-times'' operator: @example mysql> SELECT * FROM pet WHERE name REGEXP "^.@{5@}$"; +-------+--------+---------+------+------------+-------+ | name | owner | species | sex | birth | death | +-------+--------+---------+------+------------+-------+ | Claws | Gwen | cat | m | 1994-03-17 | NULL | | Buffy | Harold | dog | f | 1989-05-13 | NULL | +-------+--------+---------+------+------------+-------+ @end example @node Counting rows, Multiple tables, Pattern matching, Retrieving data @subsubsection Counting Rows @cindex rows, counting @cindex tables, counting rows @cindex counting, table rows Databases are often used to answer the question, ``How often does a certain type of data occur in a table?'' For example, you might want to know how many pets you have, or how many pets each owner has, or you might want to perform various kinds of censuses on your animals. Counting the total number of animals you have is the same question as ``How many rows are in the @code{pet} table?'' because there is one record per pet. The @code{COUNT()} function counts the number of non-@code{NULL} results, so the query to count your animals looks like this: @example mysql> SELECT COUNT(*) FROM pet; +----------+ | COUNT(*) | +----------+ | 9 | +----------+ @end example Earlier, you retrieved the names of the people who owned pets. You can use @code{COUNT()} if you want to find out how many pets each owner has: @example mysql> SELECT owner, COUNT(*) FROM pet GROUP BY owner; +--------+----------+ | owner | COUNT(*) | +--------+----------+ | Benny | 2 | | Diane | 2 | | Gwen | 3 | | Harold | 2 | +--------+----------+ @end example Note the use of @code{GROUP BY} to group together all records for each @code{owner}. Without it, all you get is an error message: @example mysql> SELECT owner, COUNT(owner) FROM pet; ERROR 1140 at line 1: Mixing of GROUP columns (MIN(),MAX(),COUNT()...) with no GROUP columns is illegal if there is no GROUP BY clause @end example @code{COUNT()} and @code{GROUP BY} are useful for characterizing your data in various ways. The following examples show different ways to perform animal census operations. Number of animals per species: @example mysql> SELECT species, COUNT(*) FROM pet GROUP BY species; +---------+----------+ | species | COUNT(*) | +---------+----------+ | bird | 2 | | cat | 2 | | dog | 3 | | hamster | 1 | | snake | 1 | +---------+----------+ @end example Number of animals per sex: @example mysql> SELECT sex, COUNT(*) FROM pet GROUP BY sex; +------+----------+ | sex | COUNT(*) | +------+----------+ | NULL | 1 | | f | 4 | | m | 4 | +------+----------+ @end example (In this output, @code{NULL} indicates sex unknown.) Number of animals per combination of species and sex: @example mysql> SELECT species, sex, COUNT(*) FROM pet GROUP BY species, sex; +---------+------+----------+ | species | sex | COUNT(*) | +---------+------+----------+ | bird | NULL | 1 | | bird | f | 1 | | cat | f | 1 | | cat | m | 1 | | dog | f | 1 | | dog | m | 2 | | hamster | f | 1 | | snake | m | 1 | +---------+------+----------+ @end example You need not retrieve an entire table when you use @code{COUNT()}. For example, the previous query, when performed just on dogs and cats, looks like this: @example mysql> SELECT species, sex, COUNT(*) FROM pet -> WHERE species = "dog" OR species = "cat" -> GROUP BY species, sex; +---------+------+----------+ | species | sex | COUNT(*) | +---------+------+----------+ | cat | f | 1 | | cat | m | 1 | | dog | f | 1 | | dog | m | 2 | +---------+------+----------+ @end example Or, if you wanted the number of animals per sex only for known-sex animals: @example mysql> SELECT species, sex, COUNT(*) FROM pet -> WHERE sex IS NOT NULL -> GROUP BY species, sex; +---------+------+----------+ | species | sex | COUNT(*) | +---------+------+----------+ | bird | f | 1 | | cat | f | 1 | | cat | m | 1 | | dog | f | 1 | | dog | m | 2 | | hamster | f | 1 | | snake | m | 1 | +---------+------+----------+ @end example @node Multiple tables, , Counting rows, Retrieving data @subsubsection Using More Than one Table @cindex tables, multiple The @code{pet} table keeps track of which pets you have. If you want to record other information about them, such as events in their lives like visits to the vet or when litters are born, you need another table. What should this table look like? It needs: @itemize @bullet @item To contain the pet name so you know which animal each event pertains to. @item A date so you know when the event occurred. @item A field to describe the event. @item An event type field, if you want to be able to categorize events. @end itemize Given these considerations, the @code{CREATE TABLE} statement for the @code{event} table might look like this: @example mysql> CREATE TABLE event (name VARCHAR(20), date DATE, -> type VARCHAR(15), remark VARCHAR(255)); @end example As with the @code{pet} table, it's easiest to load the initial records by creating a tab-delimited text file containing the information: @multitable @columnfractions .15 .15 .15 .55 @item Fluffy @tab 1995-05-15 @tab litter @tab 4 kittens, 3 female, 1 male @item Buffy @tab 1993-06-23 @tab litter @tab 5 puppies, 2 female, 3 male @item Buffy @tab 1994-06-19 @tab litter @tab 3 puppies, 3 female @item Chirpy @tab 1999-03-21 @tab vet @tab needed beak straightened @item Slim @tab 1997-08-03 @tab vet @tab broken rib @item Bowser @tab 1991-10-12 @tab kennel @item Fang @tab 1991-10-12 @tab kennel @item Fang @tab 1998-08-28 @tab birthday @tab Gave him a new chew toy @item Claws @tab 1998-03-17 @tab birthday @tab Gave him a new flea collar @item Whistler @tab 1998-12-09 @tab birthday @tab First birthday @end multitable Load the records like this: @example mysql> LOAD DATA LOCAL INFILE "event.txt" INTO TABLE event; @end example Based on what you've learned from the queries you've run on the @code{pet} table, you should be able to perform retrievals on the records in the @code{event} table; the principles are the same. But when is the @code{event} table by itself insufficient to answer questions you might ask? Suppose you want to find out the ages of each pet when they had their litters. The @code{event} table indicates when this occurred, but to calculate the age of the mother, you need her birth date. Because that is stored in the @code{pet} table, you need both tables for the query: @example mysql> SELECT pet.name, (TO_DAYS(date) - TO_DAYS(birth))/365 AS age, remark -> FROM pet, event -> WHERE pet.name = event.name AND type = "litter"; +--------+------+-----------------------------+ | name | age | remark | +--------+------+-----------------------------+ | Fluffy | 2.27 | 4 kittens, 3 female, 1 male | | Buffy | 4.12 | 5 puppies, 2 female, 3 male | | Buffy | 5.10 | 3 puppies, 3 female | +--------+------+-----------------------------+ @end example There are several things to note about this query: @itemize @bullet @item The @code{FROM} clause lists two tables because the query needs to pull information from both of them. @item When combining (joining) information from multiple tables, you need to specify how records in one table can be matched to records in the other. This is easy because they both have a @code{name} column. The query uses @code{WHERE} clause to match up records in the two tables based on the @code{name} values. @item Because the @code{name} column occurs in both tables, you must be specific about which table you mean when referring to the column. This is done by prepending the table name to the column name. @end itemize You need not have two different tables to perform a join. Sometimes it is useful to join a table to itself, if you want to compare records in a table to other records in that same table. For example, to find breeding pairs among your pets, you can join the @code{pet} table with itself to pair up males and females of like species: @example mysql> SELECT p1.name, p1.sex, p2.name, p2.sex, p1.species -> FROM pet AS p1, pet AS p2 -> WHERE p1.species = p2.species AND p1.sex = "f" AND p2.sex = "m"; +--------+------+--------+------+---------+ | name | sex | name | sex | species | +--------+------+--------+------+---------+ | Fluffy | f | Claws | m | cat | | Buffy | f | Fang | m | dog | | Buffy | f | Bowser | m | dog | +--------+------+--------+------+---------+ @end example In this query, we specify aliases for the table name in order to refer to the columns and keep straight which instance of the table each column reference is associated with. @node Getting information, Examples, Database use, Tutorial @section Getting Information About Databases and Tables @cindex databases, information about @cindex tables, information about @findex DESCRIBE What if you forget the name of a database or table, or what the structure of a given table is (for example, what its columns are called)? MySQL addresses this problem through several statements that provide information about the databases and tables it supports. You have already seen @code{SHOW DATABASES}, which lists the databases managed by the server. To find out which database is currently selected, use the @code{DATABASE()} function: @example mysql> SELECT DATABASE(); +------------+ | DATABASE() | +------------+ | menagerie | +------------+ @end example If you haven't selected any database yet, the result is blank. To find out what tables the current database contains (for example, when you're not sure about the name of a table), use this command: @example mysql> SHOW TABLES; +---------------------+ | Tables in menagerie | +---------------------+ | event | | pet | +---------------------+ @end example If you want to find out about the structure of a table, the @code{DESCRIBE} command is useful; it displays information about each of a table's columns: @example mysql> DESCRIBE pet; +---------+-------------+------+-----+---------+-------+ | Field | Type | Null | Key | Default | Extra | +---------+-------------+------+-----+---------+-------+ | name | varchar(20) | YES | | NULL | | | owner | varchar(20) | YES | | NULL | | | species | varchar(20) | YES | | NULL | | | sex | char(1) | YES | | NULL | | | birth | date | YES | | NULL | | | death | date | YES | | NULL | | +---------+-------------+------+-----+---------+-------+ @end example @code{Field} indicates the column name, @code{Type} is the data type for the column, @code{Null} indicates whether or not the column can contain @code{NULL} values, @code{Key} indicates whether or not the column is indexed, and @code{Default} specifies the column's default value. If you have indexes on a table, @code{SHOW INDEX FROM tbl_name} produces information about them. @node Examples, Batch mode, Getting information, Tutorial @section Examples of Common Queries @cindex queries, examples @cindex examples, queries Here are examples of how to solve some common problems with MySQL. Some of the examples use the table @code{shop} to hold the price of each article (item number) for certain traders (dealers). Supposing that each trader has a single fixed price per article, then (@code{article}, @code{dealer}) is a primary key for the records. Start the command line tool @code{mysql} and select a database: @example mysql your-database-name @end example (In most MySQL installations, you can use the database-name 'test'). You can create the example table as: @example CREATE TABLE shop ( article INT(4) UNSIGNED ZEROFILL DEFAULT '0000' NOT NULL, dealer CHAR(20) DEFAULT '' NOT NULL, price DOUBLE(16,2) DEFAULT '0.00' NOT NULL, PRIMARY KEY(article, dealer)); INSERT INTO shop VALUES (1,'A',3.45),(1,'B',3.99),(2,'A',10.99),(3,'B',1.45),(3,'C',1.69), (3,'D',1.25),(4,'D',19.95); @end example Okay, so the example data is: @example mysql> SELECT * FROM shop; +---------+--------+-------+ | article | dealer | price | +---------+--------+-------+ | 0001 | A | 3.45 | | 0001 | B | 3.99 | | 0002 | A | 10.99 | | 0003 | B | 1.45 | | 0003 | C | 1.69 | | 0003 | D | 1.25 | | 0004 | D | 19.95 | +---------+--------+-------+ @end example @menu * example-Maximum-column:: The maximum value for a column * example-Maximum-row:: The row holding the maximum of a certain column * example-Maximum-column-group:: Maximum of column per group * example-Maximum-column-group-row:: The rows holding the group-wise maximum of a certain field * example-user-variables:: Using user variables * example-Foreign keys:: Using foreign keys * Searching on two keys:: Searching on Two Keys * Calculating days:: Calculating visits per day @end menu @node example-Maximum-column, example-Maximum-row, Examples, Examples @subsection The Maximum Value for a Column ``What's the highest item number?'' @example SELECT MAX(article) AS article FROM shop +---------+ | article | +---------+ | 4 | +---------+ @end example @node example-Maximum-row, example-Maximum-column-group, example-Maximum-column, Examples @subsection The Row Holding the Maximum of a Certain Column ``Find number, dealer, and price of the most expensive article.'' In ANSI SQL this is easily done with a sub-query: @example SELECT article, dealer, price FROM shop WHERE price=(SELECT MAX(price) FROM shop) @end example In MySQL (which does not yet have sub-selects), just do it in two steps: @enumerate @item Get the maximum price value from the table with a @code{SELECT} statement. @item Using this value compile the actual query: @example SELECT article, dealer, price FROM shop WHERE price=19.95 @end example @end enumerate Another solution is to sort all rows descending by price and only get the first row using the MySQL specific @code{LIMIT} clause: @example SELECT article, dealer, price FROM shop ORDER BY price DESC LIMIT 1 @end example @strong{NOTE}: If there are several most expensive articles (for example, each 19.95) the @code{LIMIT} solution shows only one of them! @node example-Maximum-column-group, example-Maximum-column-group-row, example-Maximum-row, Examples @subsection Maximum of Column per Group ``What's the highest price per article?'' @example SELECT article, MAX(price) AS price FROM shop GROUP BY article +---------+-------+ | article | price | +---------+-------+ | 0001 | 3.99 | | 0002 | 10.99 | | 0003 | 1.69 | | 0004 | 19.95 | +---------+-------+ @end example @node example-Maximum-column-group-row, example-user-variables, example-Maximum-column-group, Examples @subsection The Rows Holding the Group-wise Maximum of a Certain Field ``For each article, find the dealer(s) with the most expensive price.'' In ANSI SQL, I'd do it with a sub-query like this: @example SELECT article, dealer, price FROM shop s1 WHERE price=(SELECT MAX(s2.price) FROM shop s2 WHERE s1.article = s2.article); @end example In MySQL it's best do it in several steps: @enumerate @item Get the list of (article,maxprice). @item For each article get the corresponding rows that have the stored maximum price. @end enumerate This can easily be done with a temporary table: @example CREATE TEMPORARY TABLE tmp ( article INT(4) UNSIGNED ZEROFILL DEFAULT '0000' NOT NULL, price DOUBLE(16,2) DEFAULT '0.00' NOT NULL); LOCK TABLES shop read; INSERT INTO tmp SELECT article, MAX(price) FROM shop GROUP BY article; SELECT shop.article, dealer, shop.price FROM shop, tmp WHERE shop.article=tmp.article AND shop.price=tmp.price; UNLOCK TABLES; DROP TABLE tmp; @end example If you don't use a @code{TEMPORARY} table, you must also lock the 'tmp' table. ``Can it be done with a single query?'' Yes, but only by using a quite inefficient trick that I call the ``MAX-CONCAT trick'': @example SELECT article, SUBSTRING( MAX( CONCAT(LPAD(price,6,'0'),dealer) ), 7) AS dealer, 0.00+LEFT( MAX( CONCAT(LPAD(price,6,'0'),dealer) ), 6) AS price FROM shop GROUP BY article; +---------+--------+-------+ | article | dealer | price | +---------+--------+-------+ | 0001 | B | 3.99 | | 0002 | A | 10.99 | | 0003 | C | 1.69 | | 0004 | D | 19.95 | +---------+--------+-------+ @end example The last example can, of course, be made a bit more efficient by doing the splitting of the concatenated column in the client. @node example-user-variables, example-Foreign keys, example-Maximum-column-group-row, Examples @subsection Using user variables You can use MySQL user variables to remember results without having to store them in temporary variables in the client. @xref{Variables}. For example, to find the articles with the highest and lowest price you can do: @example select @@min_price:=min(price),@@max_price:=max(price) from shop; select * from shop where price=@@min_price or price=@@max_price; +---------+--------+-------+ | article | dealer | price | +---------+--------+-------+ | 0003 | D | 1.25 | | 0004 | D | 19.95 | +---------+--------+-------+ @end example @node example-Foreign keys, Searching on two keys, example-user-variables, Examples @subsection Using Foreign Keys @cindex foreign keys @cindex keys, foreign You don't need foreign keys to join 2 tables. The only thing MySQL doesn't do is @code{CHECK} to make sure that the keys you use really exist in the table(s) you're referencing and it doesn't automatically delete rows from table with a foreign key definition. If you use your keys like normal, it'll work just fine: @example CREATE TABLE persons ( id SMALLINT UNSIGNED NOT NULL AUTO_INCREMENT, name CHAR(60) NOT NULL, PRIMARY KEY (id) ); CREATE TABLE shirts ( id SMALLINT UNSIGNED NOT NULL AUTO_INCREMENT, style ENUM('t-shirt', 'polo', 'dress') NOT NULL, color ENUM('red', 'blue', 'orange', 'white', 'black') NOT NULL, owner SMALLINT UNSIGNED NOT NULL REFERENCES persons, PRIMARY KEY (id) ); INSERT INTO persons VALUES (NULL, 'Antonio Paz'); INSERT INTO shirts VALUES (NULL, 'polo', 'blue', LAST_INSERT_ID()), (NULL, 'dress', 'white', LAST_INSERT_ID()), (NULL, 't-shirt', 'blue', LAST_INSERT_ID()); INSERT INTO persons VALUES (NULL, 'Lilliana Angelovska'); INSERT INTO shirts VALUES (NULL, 'dress', 'orange', LAST_INSERT_ID()), (NULL, 'polo', 'red', LAST_INSERT_ID()), (NULL, 'dress', 'blue', LAST_INSERT_ID()), (NULL, 't-shirt', 'white', LAST_INSERT_ID()); SELECT * FROM persons; +----+---------------------+ | id | name | +----+---------------------+ | 1 | Antonio Paz | | 2 | Lilliana Angelovska | +----+---------------------+ SELECT * FROM shirts; +----+---------+--------+-------+ | id | style | color | owner | +----+---------+--------+-------+ | 1 | polo | blue | 1 | | 2 | dress | white | 1 | | 3 | t-shirt | blue | 1 | | 4 | dress | orange | 2 | | 5 | polo | red | 2 | | 6 | dress | blue | 2 | | 7 | t-shirt | white | 2 | +----+---------+--------+-------+ SELECT s.* FROM persons p, shirts s WHERE p.name LIKE 'Lilliana%' AND s.owner = p.id AND s.color <> 'white'; +----+-------+--------+-------+ | id | style | color | owner | +----+-------+--------+-------+ | 4 | dress | orange | 2 | | 5 | polo | red | 2 | | 6 | dress | blue | 2 | +----+-------+--------+-------+ @end example @node Searching on two keys, Calculating days, example-Foreign keys, Examples @subsection Searching on Two Keys @findex UNION @cindex searching, two keys @cindex keys, searching on two MySQL doesn't yet optimize when you search on two different keys combined with @code{OR} (Searching on one key with different @code{OR} parts is optimized quite good): @example SELECT field1_index, field2_index FROM test_table WHERE field1_index = '1' OR field2_index = '1' @end example The reason is that we haven't yet had time to come up with an efficient way to handle this in the general case. (The @code{AND} handling is, in comparison, now completely general and works very well). For the moment you can solve this very efficiently by using a @code{TEMPORARY} table. This type of optimization is also very good if you are using very complicated queries where the SQL server does the optimizations in the wrong order. @example CREATE TEMPORARY TABLE tmp SELECT field1_index, field2_index FROM test_table WHERE field1_index = '1'; INSERT INTO tmp SELECT field1_index, field2_index FROM test_table WHERE field2_index = '1'; SELECT * from tmp; DROP TABLE tmp; @end example The above way to solve this query is in effect an @code{UNION} of two queries. @node Calculating days, , Searching on two keys, Examples @subsection Calculating visits per day @findex BIT_OR @findex BIT_COUNT @findex << @cindex bit_functions, example The following shows an idea of how you can use the bit group functions to calculate the number of days per month a user has visited a web page. @example CREATE TABLE t1 (year YEAR(4), month INT(2) UNSIGNED ZEROFILL, day INT(2) UNSIGNED ZEROFILL); INSERT INTO t1 VALUES(2000,1,1),(2000,1,20),(2000,1,30),(2000,2,2),(2000,2,23),(2000,2,23); SELECT year,month,BIT_COUNT(BIT_OR(1<<day)) AS days FROM t1 GROUP BY year,month; Which returns: +------+-------+------+ | year | month | days | +------+-------+------+ | 2000 | 01 | 3 | | 2000 | 02 | 2 | +------+-------+------+ @end example The above calculates how many different days was used for a given year/month combination, with automatic removal of duplicate entries. @node Batch mode, Twin, Examples, Tutorial @section Using @code{mysql} in Batch Mode @cindex modes, batch @cindex batch mode @cindex running, batch mode @cindex script files @cindex files, script In the previous sections, you used @code{mysql} interactively to enter queries and view the results. You can also run @code{mysql} in batch mode. To do this, put the commands you want to run in a file, then tell @code{mysql} to read its input from the file: @example shell> mysql < batch-file @end example If you need to specify connection parameters on the command line, the command might look like this: @example shell> mysql -h host -u user -p < batch-file Enter password: ******** @end example When you use @code{mysql} this way, you are creating a script file, then executing the script. Why use a script? Here are a few reasons: @itemize @bullet @item If you run a query repeatedly (say, every day or every week), making it a script allows you to avoid retyping it each time you execute it. @item You can generate new queries from existing ones that are similar by copying and editing script files. @item Batch mode can also be useful while you're developing a query, particularly for multiple-line commands or multiple-statement sequences of commands. If you make a mistake, you don't have to retype everything. Just edit your script to correct the error, then tell @code{mysql} to execute it again. @item If you have a query that produces a lot of output, you can run the output through a pager rather than watching it scroll off the top of your screen: @example shell> mysql < batch-file | more @end example @item You can catch the output in a file for further processing: @example shell> mysql < batch-file > mysql.out @end example @item You can distribute your script to other people so they can run the commands, too. @item Some situations do not allow for interactive use, for example, when you run a query from a @code{cron} job. In this case, you must use batch mode. @end itemize The default output format is different (more concise) when you run @code{mysql} in batch mode than when you use it interactively. For example, the output of @code{SELECT DISTINCT species FROM pet} looks like this when run interactively: @example +---------+ | species | +---------+ | bird | | cat | | dog | | hamster | | snake | +---------+ @end example But like this when run in batch mode: @example species bird cat dog hamster snake @end example If you want to get the interactive output format in batch mode, use @code{mysql -t}. To echo to the output the commands that are executed, use @code{mysql -vvv}. @node Twin, Apache, Batch mode, Tutorial @section Queries from Twin Project @cindex Twin Studies, queries @cindex queries, Twin Studeis project At Analytikerna and Lentus, we have been doing the systems and field work for a big research project. This project is a collaboration between the Institute of Environmental Medicine at Karolinska Institutet Stockholm and the Section on Clinical Research in Aging and Psychology at the University of Southern California. The project involves a screening part where all twins in Sweden older than 65 years are interviewed by telephone. Twins who meet certain criteria are passed on to the next stage. In this latter stage, twins who want to participate are visited by a doctor/nurse team. Some of the examinations include physical and neuropsychological examination, laboratory testing, neuroimaging, psychological status assessment, and family history collection. In addition, data are collected on medical and environmental risk factors. More information about Twin studies can be found at: @example @url{http://www.imm.ki.se/TWIN/TWINUKW.HTM} @end example The latter part of the project is administered with a Web interface written using Perl and MySQL. Each night all data from the interviews are moved into a MySQL database. @menu * Twin pool:: Find all non-distributed twins * Twin event:: Show a table on twin pair status @end menu @node Twin pool, Twin event, Twin, Twin @subsection Find all Non-distributed Twins The following query is used to determine who goes into the second part of the project: @example select concat(p1.id, p1.tvab) + 0 as tvid, concat(p1.christian_name, " ", p1.surname) as Name, p1.postal_code as Code, p1.city as City, pg.abrev as Area, if(td.participation = "Aborted", "A", " ") as A, p1.dead as dead1, l.event as event1, td.suspect as tsuspect1, id.suspect as isuspect1, td.severe as tsevere1, id.severe as isevere1, p2.dead as dead2, l2.event as event2, h2.nurse as nurse2, h2.doctor as doctor2, td2.suspect as tsuspect2, id2.suspect as isuspect2, td2.severe as tsevere2, id2.severe as isevere2, l.finish_date from twin_project as tp /* For Twin 1 */ left join twin_data as td on tp.id = td.id and tp.tvab = td.tvab left join informant_data as id on tp.id = id.id and tp.tvab = id.tvab left join harmony as h on tp.id = h.id and tp.tvab = h.tvab left join lentus as l on tp.id = l.id and tp.tvab = l.tvab /* For Twin 2 */ left join twin_data as td2 on p2.id = td2.id and p2.tvab = td2.tvab left join informant_data as id2 on p2.id = id2.id and p2.tvab = id2.tvab left join harmony as h2 on p2.id = h2.id and p2.tvab = h2.tvab left join lentus as l2 on p2.id = l2.id and p2.tvab = l2.tvab, person_data as p1, person_data as p2, postal_groups as pg where /* p1 gets main twin and p2 gets his/her twin. */ /* ptvab is a field inverted from tvab */ p1.id = tp.id and p1.tvab = tp.tvab and p2.id = p1.id and p2.ptvab = p1.tvab and /* Just the sceening survey */ tp.survey_no = 5 and /* Skip if partner died before 65 but allow emigration (dead=9) */ (p2.dead = 0 or p2.dead = 9 or (p2.dead = 1 and (p2.death_date = 0 or (((to_days(p2.death_date) - to_days(p2.birthday)) / 365) >= 65)))) and ( /* Twin is suspect */ (td.future_contact = 'Yes' and td.suspect = 2) or /* Twin is suspect - Informant is Blessed */ (td.future_contact = 'Yes' and td.suspect = 1 and id.suspect = 1) or /* No twin - Informant is Blessed */ (ISNULL(td.suspect) and id.suspect = 1 and id.future_contact = 'Yes') or /* Twin broken off - Informant is Blessed */ (td.participation = 'Aborted' and id.suspect = 1 and id.future_contact = 'Yes') or /* Twin broken off - No inform - Have partner */ (td.participation = 'Aborted' and ISNULL(id.suspect) and p2.dead = 0)) and l.event = 'Finished' /* Get at area code */ and substring(p1.postal_code, 1, 2) = pg.code /* Not already distributed */ and (h.nurse is NULL or h.nurse=00 or h.doctor=00) /* Has not refused or been aborted */ and not (h.status = 'Refused' or h.status = 'Aborted' or h.status = 'Died' or h.status = 'Other') order by tvid; @end example Some explanations: @table @asis @item @code{concat(p1.id, p1.tvab) + 0 as tvid} We want to sort on the concatenated @code{id} and @code{tvab} in numerical order. Adding @code{0} to the result causes MySQL to treat the result as a number. @item column @code{id} This identifies a pair of twins. It is a key in all tables. @item column @code{tvab} This identifies a twin in a pair. It has a value of @code{1} or @code{2}. @item column @code{ptvab} This is an inverse of @code{tvab}. When @code{tvab} is @code{1} this is @code{2}, and vice versa. It exists to save typing and to make it easier for MySQL to optimize the query. @end table This query demonstrates, among other things, how to do lookups on a table from the same table with a join (@code{p1} and @code{p2}). In the example, this is used to check whether a twin's partner died before the age of 65. If so, the row is not returned. All of the above exist in all tables with twin-related information. We have a key on both @code{id,tvab} (all tables), and @code{id,ptvab} (@code{person_data}) to make queries faster. On our production machine (A 200MHz UltraSPARC), this query returns about 150-200 rows and takes less than one second. The current number of records in the tables used above: @multitable @columnfractions .3 .5 @item @strong{Table} @tab @strong{Rows} @item @code{person_data} @tab 71074 @item @code{lentus} @tab 5291 @item @code{twin_project} @tab 5286 @item @code{twin_data} @tab 2012 @item @code{informant_data} @tab 663 @item @code{harmony} @tab 381 @item @code{postal_groups} @tab 100 @end multitable @node Twin event, , Twin pool, Twin @subsection Show a Table on Twin Pair Status Each interview ends with a status code called @code{event}. The query shown below is used to display a table over all twin pairs combined by event. This indicates in how many pairs both twins are finished, in how many pairs one twin is finished and the other refused, and so on. @example select t1.event, t2.event, count(*) from lentus as t1, lentus as t2, twin_project as tp where /* We are looking at one pair at a time */ t1.id = tp.id and t1.tvab=tp.tvab and t1.id = t2.id /* Just the sceening survey */ and tp.survey_no = 5 /* This makes each pair only appear once */ and t1.tvab='1' and t2.tvab='2' group by t1.event, t2.event; @end example @node Apache, , Twin, Tutorial @section Using MySQL with Apache @cindex Apache The Contrib section includes programs that let you authenticate your users from a MySQL database and also let you log your log files into a MySQL table. @xref{Contrib}. You can change the Apache logging format to be easily readable by MySQL by putting the following into the Apache configuration file: @example LogFormat \ "\"%h\",%@{%Y%m%d%H%M%S@}t,%>s,\"%b\",\"%@{Content-Type@}o\", \ \"%U\",\"%@{Referer@}i\",\"%@{User-Agent@}i\"" @end example In MySQL you can do something like this: @example LOAD DATA INFILE '/local/access_log' INTO TABLE table_name FIELDS TERMINATED BY ',' OPTIONALLY ENCLOSED BY '"' ESCAPED BY '\\' @end example @node MySQL Database Administration, MySQL Optimization, Tutorial, Top @chapter MySQL Database Administration @menu * Configuring MySQL:: Configuring MySQL * Privilege system:: General Security Issues and the MySQL Access Privilege System * User Account Management:: MySQL User Account Management * Disaster Prevention:: Disaster Prevention and Recovery * Database Administration:: Database Administration Language Reference * Localization:: MySQL Localization and International Usage * Server-Side Scripts:: MySQL Server-Side Scripts and Utilities * Client-Side Scripts:: MySQL Client-Side Scripts and Utilities * Log Files:: The MySQL Log Files * Replication:: Replication in MySQL @end menu @node Configuring MySQL, Privilege system, MySQL Database Administration, MySQL Database Administration @section Configuring MySQL @menu * Command-line options:: mysqld Command-line Options * Option files:: my.cnf Option Files * Installing many servers:: Installing Many Servers on the Same Machine * Multiple servers:: Running Multiple MySQL Servers on the Same Machine @end menu @node Command-line options, Option files, Configuring MySQL, Configuring MySQL @subsection mysqld Command-line Options @findex command-line options @cindex options, command-line @cindex mysqld options @code{mysqld} accepts the following command-line options: @table @code @item --ansi Use ANSI SQL syntax instead of MySQL syntax. @xref{ANSI mode}. @item -b, --basedir=path Path to installation directory. All paths are usually resolved relative to this. @item --big-tables Allow big result sets by saving all temporary sets on file. It solves most 'table full' errors, but also slows down the queries where in-memory tables would suffice. Since Version 3.23.2, MySQL is able to solve it automatically by using memory for small temporary tables and switching to disk tables where necessary. @item --bind-address=IP IP address to bind to. @item --character-sets-dir=path Directory where character sets are. @xref{Character sets}. @item --chroot=path Chroot @code{mysqld} daemon during startup. Recommended security measure. It will somewhat limit @code{LOAD DATA INFILE} and @code{SELECT ... INTO OUTFILE} though. @item --core-file Write a core file if @code{mysqld} dies. For some systems you must also specify @code{--core-file-size} to @code{safe_mysqld}. @xref{safe_mysqld, , @code{safe_mysqld}}. @item -h, --datadir=path Path to the database root. @item --default-character-set=charset Set the default character set. @xref{Character sets}. @item --default-table-type=type Set the default table type for tables. @xref{Table types}. @item --debug[...]= If MySQL is configured with @code{--with-debug}, you can use this option to get a trace file of what @code{mysqld} is doing. @xref{Making trace files}. @item --delay-key-write-for-all-tables Don't flush key buffers between writes for any @code{MyISAM} table. @xref{Server parameters}. @item --enable-locking Enable system locking. Note that if you use this option on a system which a not fully working lockd() (as on Linux) you will easily get mysqld to deadlock. @item --enable-named-pipe Enable support for named pipes (only on NT/Win2000/XP). @item -T, --exit-info This is a bit mask of different flags one can use for debugging the mysqld server; One should not use this option if one doesn't know exactly what it does! @item --flush Flush all changes to disk after each SQL command. Normally MySQL only does a write of all changes to disk after each SQL command and lets the operating system handle the syncing to disk. @xref{Crashing}. @item -?, --help Display short help and exit. @item --init-file=file Read SQL commands from this file at startup. @item -L, --language=... Client error messages in given language. May be given as a full path. @xref{Languages}. @item -l, --log[=file] Log connections and queries to file. @xref{Query log}. @item --log-isam[=file] Log all ISAM/MyISAM changes to file (only used when debugging ISAM/MyISAM). @item --log-slow-queries[=file] Log all queries that have taken more than @code{long_query_time} seconds to execute to file. @xref{Slow query log}. @item --log-update[=file] Log updates to @code{file.#} where @code{#} is a unique number if not given. @xref{Update log}. @item --log-long-format Log some extra information to update log. If you are using @code{--log-slow-queries} then queries that are not using indexes are logged to the slow query log. @item --low-priority-updates Table-modifying operations (@code{INSERT}/@code{DELETE}/@code{UPDATE}) will have lower priority than selects. It can also be done via @code{@{INSERT | REPLACE | UPDATE | DELETE@} LOW_PRIORITY ...} to lower the priority of only one query, or by @code{SET OPTION SQL_LOW_PRIORITY_UPDATES=1} to change the priority in one thread. @xref{Table locking}. @item --memlock Lock the @code{mysqld} process in memory. This works only if your system supports the @code{mlockall()} system call (like Solaris). This may help if you have a problem where the operating system is causing @code{mysqld} to swap on disk. @item --myisam-recover [=option[,option...]]] where option is any combination of @code{DEFAULT}, @code{BACKUP}, @code{FORCE} or @code{QUICK}. You can also set this explicitely to @code{""} if you want to disable this option. If this option is used, @code{mysqld} will on open check if the table is marked as crashed or if if the table wasn't closed properly. (The last option only works if you are running with @code{--skip-locking}). If this is the case @code{mysqld} will run check on the table. If the table was corrupted, @code{mysqld} will attempt to repair it. The following options affects how the repair works. @multitable @columnfractions .3 .7 @item DEFAULT @tab The same as not giving any option to @code{--myisam-recover}. @item BACKUP @tab If the data table was changed during recover, save a backup of the @file{table_name.MYD} data file as @file{table_name-datetime.BAK}. @item FORCE @tab Run recover even if we will loose more than one row from the .MYD file. @item QUICK @tab Don't check the rows in the table if there isn't any delete blocks. @end multitable Before a table is automatically repaired, MySQL will add a note about this in the error log. If you want to be able to recover from most things without user intervention, you should use the options @code{BACKUP,FORCE}. This will force a repair of a table even if some rows would be deleted, but it will keep the old data file as a backup so that you can later examine what happened. @item --pid-file=path Path to pid file used by @code{safe_mysqld}. @item -P, --port=... Port number to listen for TCP/IP connections. @item -o, --old-protocol Use the 3.20 protocol for compatibility with some very old clients. @xref{Upgrading-from-3.20}. @item --one-thread Only use one thread (for debugging under Linux). @xref{Debugging server}. @item -O, --set-variable var=option Give a variable a value. @code{--help} lists variables. You can find a full description for all variables in the @code{SHOW VARIABLES} section in this manual. @xref{SHOW VARIABLES}. The tuning server parameters section includes information of how to optimize these. @xref{Server parameters}. @item --safe-mode Skip some optimize stages. Implies @code{--skip-delay-key-write}. @item --safe-show-database Don't show databases for which the user doesn't have any privileges. @item --safe-user-create If this is enabled, a user can't create new users with the GRANT command, if the user doesn't have @code{INSERT} privilege to the @code{mysql.user} table or any column in this table. @item --skip-concurrent-insert Turn off the ability to select and insert at the same time on @code{MyISAM} tables. (This is only to be used if you think you have found a bug in this feature). @item --skip-delay-key-write Ignore the @code{delay_key_write} option for all tables. @xref{Server parameters}. @item --skip-grant-tables This option causes the server not to use the privilege system at all. This gives everyone @emph{full access} to all databases! (You can tell a running server to start using the grant tables again by executing @code{mysqladmin flush-privileges} or @code{mysqladmin reload}.) @item --skip-host-cache Never use host name cache for faster name-ip resolution, but query DNS server on every connect instead. @xref{DNS}. @item --skip-locking Don't use system locking. To use @code{isamchk} or @code{myisamchk} you must shut down the server. @xref{Stability}. Note that in MySQL Version 3.23 you can use @code{REPAIR} and @code{CHECK} to repair/check @code{MyISAM} tables. @item --skip-name-resolve Hostnames are not resolved. All @code{Host} column values in the grant tables must be IP numbers or @code{localhost}. @xref{DNS}. @item --skip-networking Don't listen for TCP/IP connections at all. All interaction with @code{mysqld} must be made via Unix sockets. This option is highly recommended for systems where only local requests are allowed. @xref{DNS}. @item --skip-new Don't use new, possible wrong routines. Implies @code{--skip-delay-key-write}. This will also set default table type to @code{ISAM}. @xref{ISAM}. @item --skip-symlink Don't delete or rename files that a symlinked file in the data directory points to. @item --skip-safemalloc If MySQL is configured with @code{--with-debug=full}, all programs will check the memory for overruns for every memory allocation and memory freeing. As this checking is very slow, you can avoid this, when you don't need memory checking, by using this option. @item --skip-show-database Don't allow 'SHOW DATABASE' commands, unless the user has @strong{process} privilege. @item --skip-stack-trace Don't write stack traces. This option is useful when you are running @code{mysqld} under a debugger. @xref{Debugging server}. @item --skip-thread-priority Disable using thread priorities for faster response time. @item --socket=path Socket file to use for local connections instead of default @code{/tmp/mysql.sock}. @item --sql-mode=option[,option[,option...]] Option can be any combination of: @code{REAL_AS_FLOAT}, @code{PIPES_AS_CONCAT}, @code{ANSI_QUOTES}, @code{IGNORE_SPACE}, @code{SERIALIZE}, @code{ONLY_FULL_GROUP_BY}. It can also be empty (@code{""}) if you want to reset this. By specifying all of the above options is same as using --ansi. With this option one can turn on only needed SQL modes. @xref{ANSI mode}. @item transaction-isolation= @{ READ-UNCOMMITTED | READ-COMMITTED | REPEATABLE-READ | SERIALIZABLE @} Sets the default transaction isolation level. @xref{SET TRANSACTION}. @item -t, --tmpdir=path Path for temporary files. It may be useful if your default @code{/tmp} directory resides on a partition too small to hold temporary tables. @item -u, --user=user_name Run @code{mysqld} daemon as user @code{user_name}. This option is @emph{mandatory} when starting @code{mysqld} as root. @item -V, --version Output version information and exit. @item -W, --warnings Print out warnings like @code{Aborted connection...} to the @code{.err} file. @xref{Communication errors}. @end table @node Option files, Installing many servers, Command-line options, Configuring MySQL @subsection my.cnf Option Files @cindex default options @cindex option files @cindex creating, default startup options @cindex startup options, default MySQL can, since Version 3.22, read default startup options for the server and for clients from option files. MySQL reads default options from the following files on Unix: @tindex .my.cnf file @multitable @columnfractions .3 .7 @item @strong{Filename} @tab @strong{Purpose} @item @code{/etc/my.cnf} @tab Global options @item @code{DATADIR/my.cnf} @tab Server-specific options @item @code{defaults-extra-file} @tab The file specified with --defaults-extra-file=# @item @code{~/.my.cnf} @tab User-specific options @end multitable @code{DATADIR} is the MySQL data directory (typically @file{/usr/local/mysql/data} for a binary installation or @file{/usr/local/var} for a source installation). Note that this is the directory that was specified at configuration time, not the one specified with @code{--datadir} when @code{mysqld} starts up! (@code{--datadir} has no effect on where the server looks for option files, because it looks for them before it processes any command-line arguments.) MySQL reads default options from the following files on Windows: @multitable @columnfractions .3 .7 @item @strong{Filename} @tab @strong{Purpose} @item @code{windows-system-directory\my.ini} @tab Global options @item @code{C:\my.cnf} @tab Global options @item @code{C:\mysql\data\my.cnf} @tab Server-specific options @end multitable Note that on Windows, you should specify all paths with @code{/} instead of @code{\}. If you use @code{\}, you need to specify this twice, as @code{\} is the escape character in MySQL. @cindex environment variables MySQL tries to read option files in the order listed above. If multiple option files exist, an option specified in a file read later takes precedence over the same option specified in a file read earlier. Options specified on the command line take precedence over options specified in any option file. Some options can be specified using environment variables. Options specified on the command line or in option files take precedence over environment variable values. @xref{Environment variables}. The following programs support option files: @code{mysql}, @code{mysqladmin}, @code{mysqld}, @code{mysqldump}, @code{mysqlimport}, @code{mysql.server}, @code{myisamchk}, and @code{myisampack}. You can use option files to specify any long option that a program supports! Run the program with @code{--help} to get a list of available options. An option file can contain lines of the following forms: @table @code @item #comment Comment lines start with @samp{#} or @samp{;}. Empty lines are ignored. @item [group] @code{group} is the name of the program or group for which you want to set options. After a group line, any @code{option} or @code{set-variable} lines apply to the named group until the end of the option file or another group line is given. @item option This is equivalent to @code{--option} on the command line. @item option=value This is equivalent to @code{--option=value} on the command line. @item set-variable = variable=value This is equivalent to @code{--set-variable variable=value} on the command line. This syntax must be used to set a @code{mysqld} variable. @end table The @code{client} group allows you to specify options that apply to all MySQL clients (not @code{mysqld}). This is the perfect group to use to specify the password you use to connect to the server. (But make sure the option file is readable and writable only by yourself.) Note that for options and values, all leading and trailing blanks are automatically deleted. You may use the escape sequences @samp{\b}, @samp{\t}, @samp{\n}, @samp{\r}, @samp{\\}, and @samp{\s} in your value string (@samp{\s} == blank). Here is a typical global option file: @example [client] port=3306 socket=/tmp/mysql.sock [mysqld] port=3306 socket=/tmp/mysql.sock set-variable = key_buffer_size=16M set-variable = max_allowed_packet=1M [mysqldump] quick @end example Here is typical user option file: @example [client] # The following password will be sent to all standard MySQL clients password=my_password [mysql] no-auto-rehash set-variable = connect_timeout=2 [mysqlhotcopy] interactive-timeout @end example @tindex .my.cnf file If you have a source distribution, you will find sample configuration files named @file{my-xxxx.cnf} in the @file{support-files} directory. If you have a binary distribution, look in the @file{DIR/support-files} directory, where @code{DIR} is the pathname to the MySQL installation directory (typically @file{/usr/local/mysql}). Currently there are sample configuration files for small, medium, large, and very large systems. You can copy @file{my-xxxx.cnf} to your home directory (rename the copy to @file{.my.cnf}) to experiment with this. All MySQL clients that support option files support the following options: @multitable @columnfractions .40 .60 @item --no-defaults @tab Don't read any option files. @item --print-defaults @tab Print the program name and all options that it will get. @item --defaults-file=full-path-to-default-file @tab Only use the given configuration file. @item --defaults-extra-file=full-path-to-default-file @tab Read this configuration file after the global configuration file but before the user configuration file. @end multitable Note that the above options must be first on the command line to work! @code{--print-defaults} may however be used directly after the @code{--defaults-xxx-file} commands. Note for developers: Option file handling is implemented simply by processing all matching options (that is, options in the appropriate group) before any command-line arguments. This works nicely for programs that use the last instance of an option that is specified multiple times. If you have an old program that handles multiply-specified options this way but doesn't read option files, you need add only two lines to give it that capability. Check the source code of any of the standard MySQL clients to see how to do this. In shell scripts you can use the @file{my_print_defaults} command to parse the config files: @example shell> my_print_defaults client mysql --port=3306 --socket=/tmp/mysql.sock --no-auto-rehash @end example The above output contains all options for the groups 'client' and 'mysql'. @node Installing many servers, Multiple servers, Option files, Configuring MySQL @subsection Installing Many Servers on the Same Machine @cindex post-install, many servers @cindex Installing many servers @cindex Starting many servers In some cases you may want to have many different @code{mysqld} daemons (servers) running on the same machine. You may for example want to run a new version of MySQL for testing together with an old version that is in production. Another case is when you want to give different users access to different @code{mysqld} servers that they manage themselves. One way to get a new server running is by starting it with a different socket and port as follows: @tindex @code{MYSQL_UNIX_PORT} environment variable @tindex @code{MYSQL_TCP_PORT} environment variable @tindex environment variable, @code{MYSQL_UNIX_PORT} @tindex environment variable, @code{MYSQL_TCP_PORT} @example shell> MYSQL_UNIX_PORT=/tmp/mysqld-new.sock shell> MYSQL_TCP_PORT=3307 shell> export MYSQL_UNIX_PORT MYSQL_TCP_PORT shell> scripts/mysql_install_db shell> bin/safe_mysqld & @end example The environment variables appendix includes a list of other environment variables you can use to affect @code{mysqld}. @xref{Environment variables}. The above is the quick and dirty way that one commonly uses for testing. The nice thing with this is that all connections you do in the above shell will automatically be directed to the new running server! If you need to do this more permanently, you should create an option file for each server. @xref{Option files}. In your startup script that is executed at boot time (mysql.server?) you should specify for both servers: @code{safe_mysqld --default-file=path-to-option-file} At least the following options should be different per server: @table @code @item port=# @item socket=path @item pid-file=path @end table The following options should be different, if they are used: @table @code @item log=path @item log-bin=path @item log-update=path @item log-isam=path @item bdb-logdir=path @end table If you want more performance, you can also specify the following differently: @table @code @item tmpdir=path @item bdb-tmpdir=path @end table @xref{Command-line options}. If you are installing binary MySQL versions (.tar files) and start them with @code{./bin/safe_mysqld} then in most cases the only option you need to add/change is the @code{socket} and @code{port} argument to @code{safe_mysqld}. @menu * Multiple servers:: Running Multiple MySQL Servers on the Same Machine @end menu @node Multiple servers, , Installing many servers, Configuring MySQL @subsection Running Multiple MySQL Servers on the Same Machine @cindex multiple servers @cindex servers, multiple @cindex running, multiple servers There are circumstances when you might want to run multiple servers on the same machine. For example, you might want to test a new MySQL release while leaving your existing production setup undisturbed. Or you might be an Internet service provider that wants to provide independent MySQL installations for different customers. If you want to run multiple servers, the easiest way is to compile the servers with different TCP/IP ports and socket files so they are not both listening to the same TCP/IP port or socket file. @xref{mysqld_multi, , @code{mysqld_multi}}. Assume an existing server is configured for the default port number and socket file. Then configure the new server with a @code{configure} command something like this: @example shell> ./configure --with-tcp-port=port_number \ --with-unix-socket-path=file_name \ --prefix=/usr/local/mysql-3.22.9 @end example Here @code{port_number} and @code{file_name} should be different than the default port number and socket file pathname, and the @code{--prefix} value should specify an installation directory different than the one under which the existing MySQL installation is located. You can check the socket used by any currently executing MySQL server with this command: @example shell> mysqladmin -h hostname --port=port_number variables @end example Note that if you specify ``@code{localhost}'' as a hostname, @code{mysqladmin} will default to using Unix sockets instead of TCP/IP. If you have a MySQL server running on the port you used, you will get a list of some of the most important configurable variables in MySQL, including the socket name. You don't have to recompile a new MySQL server just to start with a different port and socket. You can change the port and socket to be used by specifying them at run time as options to @code{safe_mysqld}: @example shell> /path/to/safe_mysqld --socket=file_name --port=port_number @end example @code{mysqld_multi} can also take @code{safe_mysqld} (or @code{mysqld}) as an argument and pass the options from a configuration file to @code{safe_mysqld} and further to @code{mysqld}. If you run the new server on the same database directory as another server with logging enabled, you should also specify the name of the log files to @code{safe_mysqld} with @code{--log}, @code{--log-update}, or @code{--log-slow-queries}. Otherwise, both servers may be trying to write to the same log file. @strong{WARNING}: Normally you should never have two servers that update data in the same database! If your OS doesn't support fault-free system locking, this may lead to unpleasant surprises! If you want to use another database directory for the second server, you can use the @code{--datadir=path} option to @code{safe_mysqld}. @strong{NOTE} also that starting several MySQL servers (@code{mysqlds}) in different machines and letting them access one data directory over @code{NFS} is generally a @strong{BAD IDEA}! The problem is that the @code{NFS} will become the bottleneck with the speed. It is not meant for such use. And last but not least, you would still have to come up with a solution how to make sure that two or more @code{mysqlds} are not interfering with each other. At the moment there is no platform that would 100% reliable do the file locking (@code{lockd} daemon usually) in every situation. Yet there would be one more possible risk with @code{NFS}; it would make the work even more complicated for @code{lockd} daemon to handle. So make it easy for your self and forget about the idea. The working solution is to have one computer with an operating system that efficiently handles threads and have several CPUs in it. When you want to connect to a MySQL server that is running with a different port than the port that is compiled into your client, you can use one of the following methods: @itemize @bullet @item Start the client with @code{--host 'hostname' --port=port_number} to connect with TCP/IP, or @code{[--host localhost] --socket=file_name} to connect via a Unix socket. @item In your C or Perl programs, you can give the port or socket arguments when connecting to the MySQL server. @item If your are using the Perl @code{DBD::mysql} module you can read the options from the MySQL option files. @xref{Option files}. @example $dsn = "DBI:mysql:test;mysql_read_default_group=client;mysql_read_default_file=/usr/local/mysql/data/my.cnf" $dbh = DBI->connect($dsn, $user, $password); @end example @item @tindex MYSQL_UNIX_PORT environment variable @tindex MYSQL_TCP_PORT environment variable @tindex environment variable, MYSQL_UNIX_PORT @tindex environment variable, MYSQL_TCP_PORT Set the @code{MYSQL_UNIX_PORT} and @code{MYSQL_TCP_PORT} environment variables to point to the Unix socket and TCP/IP port before you start your clients. If you normally use a specific socket or port, you should place commands to set these environment variables in your @file{.login} file. @xref{Environment variables}. @item @tindex .my.cnf file Specify the default socket and TCP/IP port in the @file{.my.cnf} file in your home directory. @xref{Option files}. @end itemize @node Privilege system, User Account Management, Configuring MySQL, MySQL Database Administration @section General Security Issues and the MySQL Access Privilege System @cindex system, security @cindex access privileges @cindex privileges, access @cindex security system @cindex ACLs MySQL has an advanced but non-standard security/privilege system. This section describes how it works. @menu * General security:: General security * Security:: How to make MySQL secure against crackers * Privileges options:: Startup Options for @code{mysqld} Concerning Security * What Privileges:: What the privilege system does * Privileges:: How the privilege system works * Privileges provided:: Privileges provided by MySQL * Connecting:: Connecting to the MySQL server * Connection access:: Access control, stage 1: Connection verification * Request access:: Access control, stage 2: Request verification * Access denied:: Causes of @code{Access denied} errors @end menu @node General security, Security, Privilege system, Privilege system @subsection General Security Guidelines Anyone using MySQL on a computer connected to the Internet should read this section to avoid the most common security mistakes. In discussing security, we emphasize the necessity of fully protecting the entire server host (not simply the MySQL server) against all types of applicable attacks: eavesdropping, altering, playback, and denial of service. We do not cover all aspects of availability and fault tolerance here. MySQL uses security based on Access Control Lists (ACLs) for all connections, queries, and other operations that a user may attempt to perform. There is also some support for SSL-encrypted connections between MySQL clients and servers. Many of the concepts discussed here are not specific to MySQL at all; the same general ideas apply to almost all applications. When running MySQL, follow these guidelines whenever possible: @itemize @bullet @item DON'T EVER GIVE ANYONE (EXCEPT THE MySQL ROOT USER) ACCESS TO THE @code{user} TABLE IN THE @code{mysql} DATABASE! The encrypted password is the real password in MySQL. If you know the password listed in the @code{user} table for a given user, you can easily log in as that user if you have access to the host listed for that account. @item Learn the MySQL access privilege system. The @code{GRANT} and @code{REVOKE} commands are used for controlling access to MySQL. Do not grant any more privileges than necessary. Never grant privileges to all hosts. Checklist: @itemize @minus @item Try @code{mysql -u root}. If you are able to connect successfully to the server without being asked for a password, you have problems. Anyone can connect to your MySQL server as the MySQL @code{root} user with full privileges! Review the MySQL installation instructions, paying particular attention to the item about setting a @code{root} password. @item Use the command @code{SHOW GRANTS} and check to see who has access to what. Remove those privileges that are not necessary using the @code{REVOKE} command. @end itemize @item Do not keep any plain-text passwords in your database. When your computer becomes compromised, the intruder can take the full list of passwords and use them. Instead use @code{MD5()} or another one-way hashing function. @item Do not choose passwords from dictionaries. There are special programs to break them. Even passwords like ``xfish98'' are very bad. Much better is ``duag98'' which contains the same word ``fish'' but typed one key to the left on a standard QWERTY keyboard. Another method is to use ``Mhall'' which is taken from the first characters of each word in the sentence ``Mary had a little lamb.'' This is easy to remember and type, but difficult to guess for someone who does not know it. @item Invest in a firewall. This protects you from at least 50% of all types of exploits in any software. Put MySQL behind the firewall or in a demilitarized zone (DMZ). Checklist: @itemize @minus @item Try to scan your ports from the Internet using a tool such as @code{nmap}. MySQL uses port 3306 by default. This port should be inaccessible from untrusted hosts. Another simple way to check whether or not your MySQL port is open is to try the following command from some remote machine, where @code{server_host} is the hostname of your MySQL server: @example shell> telnet server_host 3306 @end example If you get a connection and some garbage characters, the port is open, and should be closed on your firewall or router, unless you really have a good reason to keep it open. If @code{telnet} just hangs or the connection is refused, everything is OK; the port is blocked. @end itemize @item Do not trust any data entered by your users. They can try to trick your code by entering special or escaped character sequences in Web forms, URLs, or whatever application you have built. Be sure that your application remains secure if a user enters something like ``@code{; DROP DATABASE mysql;}''. This is an extreme example, but large security leaks and data loss may occur as a result of hackers using similar techniques, if you do not prepare for them. Also remember to check numeric data. A common mistake is to protect only strings. Sometimes people think that if a database contains only publicly available data that it need not be protected. This is incorrect. At least denial-of-service type attacks can be performed on such databases. The simplest way to protect from this type of attack is to use apostrophes around the numeric constants: @code{SELECT * FROM table WHERE ID='234'} rather than @code{SELECT * FROM table WHERE ID=234}. MySQL automatically converts this string to a number and strips all non-numeric symbols from it. Checklist: @itemize @minus @item All Web applications: @itemize @bullet @item Try to enter @samp{'} and @samp{"} in all your Web forms. If you get any kind of MySQL error, investigate the problem right away. @item Try to modify any dynamic URLs by adding @code{%22} (@samp{"}), @code{%23} (@samp{#}), and @code{%27} (@samp{'}) in the URL. @item Try to modify datatypes in dynamic URLs from numeric ones to character ones containing characters from previous examples. Your application should be safe against this and similar attacks. @item Try to enter characters, spaces, and special symbols instead of numbers in numeric fields. Your application should remove them before passing them to MySQL or your application should generate an error. Passing unchecked values to MySQL is very dangerous! @item Check data sizes before passing them to MySQL. @item Consider having your application connect to the database using a different user name than the one you use for administrative purposes. Do not give your applications any more access privileges than they need. @end itemize @item Users of PHP: @itemize @bullet @item Check out the @code{addslashes()} function. As of PHP 4.0.3, a @code{mysql_escape_string()} function is available that is based on the function of the same name in the MySQL C API. @end itemize @item Users of MySQL C API: @itemize @bullet @item Check out the @code{mysql_escape_string()} API call. @end itemize @item Users of MySQL++: @itemize @bullet @item Check out the @code{escape} and @code{quote} modifiers for query streams. @end itemize @item Users of Perl DBI: @itemize @bullet @item Check out the @code{quote()} method or use placeholders. @end itemize @item Users of Java JDBC: @itemize @bullet @item Use a @code{PreparedStatement} object and placeholders. @end itemize @end itemize @item Do not transmit plain (unencrypted) data over the Internet. These data are accessible to everyone who has the time and ability to intercept it and use it for their own purposes. Instead, use an encrypted protocol such as SSL or SSH. MySQL supports internal SSL connections as of Version 3.23.9. SSH port-forwarding can be used to create an encrypted (and compressed) tunnel for the communication. @item Learn to use the @code{tcpdump} and @code{strings} utilities. For most cases, you can check whether or not MySQL data streams are unencrypted by issuing a command like the following: @example shell> tcpdump -l -i eth0 -w - src or dst port 3306 | strings @end example (This works under Linux and should work with small modifications under other systems). Warning: If you do not see data this doesn't always actually mean that it is encrypted. If you need high security, you should consult with a security expert. @end itemize @node Security, Privileges options, General security, Privilege system @subsection How to Make MySQL Secure Against Crackers @cindex crackers, security against @cindex security, against crackers When you connect to a MySQL server, you normally should use a password. The password is not transmitted in clear text over the connection, however the encryption algorithm is not very strong, and with some effort a clever attacker can crack the password if he is able to sniff the traffic between the client and the server. If the connection between the client and the server goes through an untrusted network, you should use an SSH tunnel to encrypt the communication. All other information is transferred as text that can be read by anyone who is able to watch the connection. If you are concerned about this, you can use the compressed protocol (in MySQL Version 3.22 and above) to make things much harder. To make things even more secure you should use @code{ssh}. You can find an open source @code{ssh} client at @uref{http://www.openssh.org}, and a commercial @code{ssh} client at @uref{http://www.ssh.com}. With this, you can get an encrypted TCP/IP connection between a MySQL server and a MySQL client. To make a MySQL system secure, you should strongly consider the following suggestions: @itemize @bullet @item Use passwords for all MySQL users. Remember that anyone can log in as any other person as simply as @code{mysql -u other_user db_name} if @code{other_user} has no password. It is common behavior with client/server applications that the client may specify any user name. You can change the password of all users by editing the @code{mysql_install_db} script before you run it, or only the password for the MySQL @code{root} user like this: @example shell> mysql -u root mysql mysql> UPDATE user SET Password=PASSWORD('new_password') WHERE user='root'; mysql> FLUSH PRIVILEGES; @end example @item Don't run the MySQL daemon as the Unix @code{root} user. This is very dangerous, because any user with @code{FILE} privileges will be able to create files as @code{root} (for example, @code{~root/.bashrc}). To prevent this, @code{mysqld} will refuse to run as @code{root} unless it is specified directly using a @code{--user=root} option. @code{mysqld} can be run as an ordinary unprivileged user instead. You can also create a new Unix user @code{mysql} to make everything even more secure. If you run @code{mysqld} as another Unix user, you don't need to change the @code{root} user name in the @code{user} table, because MySQL user names have nothing to do with Unix user names. To start @code{mysqld} as another Unix user, add a @code{user} line that specifies the user name to the @code{[mysqld]} group of the @file{/etc/my.cnf} option file or the @file{my.cnf} option file in the server's data directory. For example: @example [mysqld] user=mysql @end example This will cause the server to start as the designated user whether you start it manually or by using @code{safe_mysqld} or @code{mysql.server}. For more details, see @ref{Changing MySQL user, , Changing MySQL user}. @item Don't support symlinks to tables (This can be disabled with the @code{--skip-symlink} option. This is especially important if you run @code{mysqld} as root as anyone that has write access to the mysqld data directories could then delete any file in the system! @xref{Symbolic links to tables}. @item Check that the Unix user that @code{mysqld} runs as is the only user with read/write privileges in the database directories. @item Don't give the @strong{process} privilege to all users. The output of @code{mysqladmin processlist} shows the text of the currently executing queries, so any user who is allowed to execute that command might be able to see if another user issues an @code{UPDATE user SET password=PASSWORD('not_secure')} query. @code{mysqld} reserves an extra connection for users who have the @strong{process} privilege, so that a MySQL @code{root} user can log in and check things even if all normal connections are in use. @item Don't give the @strong{file} privilege to all users. Any user that has this privilege can write a file anywhere in the file system with the privileges of the @code{mysqld} daemon! To make this a bit safer, all files generated with @code{SELECT ... INTO OUTFILE} are readable to everyone, and you cannot overwrite existing files. @tindex /etc/passwd The @strong{file} privilege may also be used to read any file accessible to the Unix user that the server runs as. This could be abused, for example, by using @code{LOAD DATA} to load @file{/etc/passwd} into a table, which can then be read with @code{SELECT}. @item If you don't trust your DNS, you should use IP numbers instead of hostnames in the grant tables. In any case, you should be very careful about creating grant table entries using hostname values that contain wild cards! @item If you want to restrict the number of connections for a single user, you can do this by setting the @code{max_user_connections} variable in @code{mysqld}. @end itemize @node Privileges options, What Privileges, Security, Privilege system @subsection Startup Options for @code{mysqld} Concerning Security The following @code{mysqld} options affect security: @table @code @item --safe-show-database With this option, @code{SHOW DATABASES} returns only those databases for which the user has some kind of privilege. @item --safe-user-create If this is enabled, an user can't create new users with the @code{GRANT} command, if the user doesn't have @code{INSERT} privilege to the @code{mysql.user} table. If you want to give a user access to just create new users with those privileges that the user has right to grant, you should give the user the following privilege: @example GRANT INSERT(user) on mysql.user to 'user'@'hostname'; @end example This will ensure that the user can't change any privilege columns directly, but has to use the @code{GRANT} command to give privileges to other users. @item --skip-grant-tables This option causes the server not to use the privilege system at all. This gives everyone @emph{full access} to all databases! (You can tell a running server to start using the grant tables again by executing @code{mysqladmin flush-privileges} or @code{mysqladmin reload}.) @item --skip-name-resolve Hostnames are not resolved. All @code{Host} column values in the grant tables must be IP numbers or @code{localhost}. @item --skip-networking Don't allow TCP/IP connections over the network. All connections to @code{mysqld} must be made via Unix sockets. This option is unsuitable for systems that use MIT-pthreads, because the MIT-pthreads package doesn't support Unix sockets. @item --skip-show-database With this option, the @code{SHOW DATABASES} statement doesn't return anything. @end table @node What Privileges, Privileges, Privileges options, Privilege system @subsection What the Privilege System Does @cindex system, privilege @cindex privilege system @cindex passwords, security The primary function of the MySQL privilege system is to authenticate a user connecting from a given host, and to associate that user with privileges on a database such as @strong{select}, @strong{insert}, @strong{update} and @strong{delete}. Additional functionality includes the ability to have an anonymous user and to grant privileges for MySQL-specific functions such as @code{LOAD DATA INFILE} and administrative operations. @node Privileges, Privileges provided, What Privileges, Privilege system @subsection How the Privilege System Works @cindex privilege system, described The MySQL privilege system ensures that all users may do exactly the things that they are supposed to be allowed to do. When you connect to a MySQL server, your identity is determined by @strong{the host from which you connect} and @strong{the user name you specify}. The system grants privileges according to your identity and @strong{what you want to do}. MySQL considers both your hostname and user name in identifying you because there is little reason to assume that a given user name belongs to the same person everywhere on the Internet. For example, the user @code{bill} who connects from @code{whitehouse.gov} need not be the same person as the user @code{bill} who connects from @code{microsoft.com}. MySQL handles this by allowing you to distinguish users on different hosts that happen to have the same name: you can grant @code{bill} one set of privileges for connections from @code{whitehouse.gov}, and a different set of privileges for connections from @code{microsoft.com}. MySQL access control involves two stages: @itemize @bullet @item Stage 1: The server checks whether or not you are even allowed to connect. @item Stage 2: Assuming you can connect, the server checks each request you issue to see whether or not you have sufficient privileges to perform it. For example, if you try to select rows from a table in a database or drop a table from the database, the server makes sure you have the @strong{select} privilege for the table or the @strong{drop} privilege for the database. @end itemize The server uses the @code{user}, @code{db}, and @code{host} tables in the @code{mysql} database at both stages of access control. The fields in these grant tables are shown below: @multitable @columnfractions .2 .25 .25 .25 @item @strong{Table name} @tab @code{user} @tab @code{db} @tab @code{host} @item @strong{Scope fields} @tab @code{Host} @tab @code{Host} @tab @code{Host} @item @tab @code{User} @tab @code{Db} @tab @code{Db} @item @tab @code{Password} @tab @code{User} @tab @item @strong{Privilege fields} @tab @code{Select_priv} @tab @code{Select_priv} @tab @code{Select_priv} @item @tab @code{Insert_priv} @tab @code{Insert_priv} @tab @code{Insert_priv} @item @tab @code{Update_priv} @tab @code{Update_priv} @tab @code{Update_priv} @item @tab @code{Delete_priv} @tab @code{Delete_priv} @tab @code{Delete_priv} @item @tab @code{Index_priv} @tab @code{Index_priv} @tab @code{Index_priv} @item @tab @code{Alter_priv} @tab @code{Alter_priv} @tab @code{Alter_priv} @item @tab @code{Create_priv} @tab @code{Create_priv} @tab @code{Create_priv} @item @tab @code{Drop_priv} @tab @code{Drop_priv} @tab @code{Drop_priv} @item @tab @code{Grant_priv} @tab @code{Grant_priv} @tab @code{Grant_priv} @item @tab @code{References_priv} @tab @tab @item @tab @code{Reload_priv} @tab @tab @item @tab @code{Shutdown_priv} @tab @tab @item @tab @code{Process_priv} @tab @tab @item @tab @code{File_priv} @tab @tab @end multitable For the second stage of access control (request verification), the server may, if the request involves tables, additionally consult the @code{tables_priv} and @code{columns_priv} tables. The fields in these tables are shown below: @multitable @columnfractions .2 .25 .25 @item @strong{Table name} @tab @code{tables_priv} @tab @code{columns_priv} @item @strong{Scope fields} @tab @code{Host} @tab @code{Host} @item @tab @code{Db} @tab @code{Db} @item @tab @code{User} @tab @code{User} @item @tab @code{Table_name} @tab @code{Table_name} @item @tab @tab @code{Column_name} @item @strong{Privilege fields} @tab @code{Table_priv} @tab @code{Column_priv} @item @tab @code{Column_priv} @tab @item @strong{Other fields} @tab @code{Timestamp} @tab @code{Timestamp} @item @tab @code{Grantor} @tab @end multitable Each grant table contains scope fields and privilege fields. Scope fields determine the scope of each entry in the tables, that is, the context in which the entry applies. For example, a @code{user} table entry with @code{Host} and @code{User} values of @code{'thomas.loc.gov'} and @code{'bob'} would be used for authenticating connections made to the server by @code{bob} from the host @code{thomas.loc.gov}. Similarly, a @code{db} table entry with @code{Host}, @code{User}, and @code{Db} fields of @code{'thomas.loc.gov'}, @code{'bob'} and @code{'reports'} would be used when @code{bob} connects from the host @code{thomas.loc.gov} to access the @code{reports} database. The @code{tables_priv} and @code{columns_priv} tables contain scope fields indicating tables or table/column combinations to which each entry applies. @cindex case sensitivity, in access checking For access-checking purposes, comparisons of @code{Host} values are case insensitive. @code{User}, @code{Password}, @code{Db}, and @code{Table_name} values are case sensitive. @code{Column_name} values are case insensitive in MySQL Version 3.22.12 or later. Privilege fields indicate the privileges granted by a table entry, that is, what operations can be performed. The server combines the information in the various grant tables to form a complete description of a user's privileges. The rules used to do this are described in @ref{Request access}. Scope fields are strings, declared as shown below; the default value for each is the empty string: @multitable @columnfractions .15 .15 .7 @item @strong{Field name} @tab @strong{Type} @item @code{Host} @tab @code{CHAR(60)} @item @code{User} @tab @code{CHAR(16)} @item @code{Password} @tab @code{CHAR(16)} @item @code{Db} @tab @code{CHAR(64)} @tab (@code{CHAR(60)} for the @code{tables_priv} and @code{columns_priv} tables) @item @code{Table_name} @tab @code{CHAR(60)} @item @code{Column_name} @tab @code{CHAR(60)} @end multitable In the @code{user}, @code{db} and @code{host} tables, all privilege fields are declared as @code{ENUM('N','Y')} --- each can have a value of @code{'N'} or @code{'Y'}, and the default value is @code{'N'}. In the @code{tables_priv} and @code{columns_priv} tables, the privilege fields are declared as @code{SET} fields: @multitable @columnfractions .2 .2 .6 @item @strong{Table name} @tab @strong{Field name} @tab @strong{Possible set elements} @item @code{tables_priv} @tab @code{Table_priv} @tab @code{'Select', 'Insert', 'Update', 'Delete', 'Create', 'Drop', 'Grant', 'References', 'Index', 'Alter'} @item @code{tables_priv} @tab @code{Column_priv} @tab @code{'Select', 'Insert', 'Update', 'References'} @item @code{columns_priv} @tab @code{Column_priv} @tab @code{'Select', 'Insert', 'Update', 'References'} @end multitable Briefly, the server uses the grant tables like this: @itemize @bullet @item The @code{user} table scope fields determine whether to allow or reject incoming connections. For allowed connections, any privileges granted in the @code{user} table indicate the user's global (superuser) privileges. These privileges apply to @strong{all} databases on the server. @item The @code{db} and @code{host} tables are used together: @itemize @minus @item The @code{db} table scope fields determine which users can access which databases from which hosts. The privilege fields determine which operations are allowed. @item The @code{host} table is used as an extension of the @code{db} table when you want a given @code{db} table entry to apply to several hosts. For example, if you want a user to be able to use a database from several hosts in your network, leave the @code{Host} value empty in the user's @code{db} table entry, then populate the @code{host} table with an entry for each of those hosts. This mechanism is described more detail in @ref{Request access}. @end itemize @item The @code{tables_priv} and @code{columns_priv} tables are similar to the @code{db} table, but are more fine-grained: they apply at the table and column levels rather than at the database level. @end itemize Note that administrative privileges (@strong{reload}, @strong{shutdown}, etc.) are specified only in the @code{user} table. This is because administrative operations are operations on the server itself and are not database-specific, so there is no reason to list such privileges in the other grant tables. In fact, only the @code{user} table need be consulted to determine whether or not you can perform an administrative operation. The @strong{file} privilege is specified only in the @code{user} table, too. It is not an administrative privilege as such, but your ability to read or write files on the server host is independent of the database you are accessing. The @code{mysqld} server reads the contents of the grant tables once, when it starts up. Changes to the grant tables take effect as indicated in @ref{Privilege changes}. When you modify the contents of the grant tables, it is a good idea to make sure that your changes set up privileges the way you want. For help in diagnosing problems, see @ref{Access denied}. For advice on security issues, @pxref{Security}. A useful diagnostic tool is the @code{mysqlaccess} script, which Yves Carlier has provided for the MySQL distribution. Invoke @code{mysqlaccess} with the @code{--help} option to find out how it works. Note that @code{mysqlaccess} checks access using only the @code{user}, @code{db} and @code{host} tables. It does not check table- or column-level privileges. @node Privileges provided, Connecting, Privileges, Privilege system @subsection Privileges Provided by MySQL @cindex privilege information, location Information about user privileges is stored in the @code{user}, @code{db}, @code{host}, @code{tables_priv}, and @code{columns_priv} tables in the @code{mysql} database (that is, in the database named @code{mysql}). The MySQL server reads the contents of these tables when it starts up and under the circumstances indicated in @ref{Privilege changes}. The names used in this manual to refer to the privileges provided by MySQL are shown below, along with the table column name associated with each privilege in the grant tables and the context in which the privilege applies: @multitable @columnfractions .15 .25 .6 @item @strong{Privilege} @tab @strong{Column} @tab @strong{Context} @item @strong{select} @tab @code{Select_priv} @tab tables @item @strong{insert} @tab @code{Insert_priv} @tab tables @item @strong{update} @tab @code{Update_priv} @tab tables @item @strong{delete} @tab @code{Delete_priv} @tab tables @item @strong{index} @tab @code{Index_priv} @tab tables @item @strong{alter} @tab @code{Alter_priv} @tab tables @item @strong{create} @tab @code{Create_priv} @tab databases, tables, or indexes @item @strong{drop} @tab @code{Drop_priv} @tab databases or tables @item @strong{grant} @tab @code{Grant_priv} @tab databases or tables @item @strong{references} @tab @code{References_priv} @tab databases or tables @item @strong{reload} @tab @code{Reload_priv} @tab server administration @item @strong{shutdown} @tab @code{Shutdown_priv} @tab server administration @item @strong{process} @tab @code{Process_priv} @tab server administration @item @strong{file} @tab @code{File_priv} @tab file access on server @end multitable The @strong{select}, @strong{insert}, @strong{update}, and @strong{delete} privileges allow you to perform operations on rows in existing tables in a database. @code{SELECT} statements require the @strong{select} privilege only if they actually retrieve rows from a table. You can execute certain @code{SELECT} statements even without permission to access any of the databases on the server. For example, you could use the @code{mysql} client as a simple calculator: @example mysql> SELECT 1+1; mysql> SELECT PI()*2; @end example The @strong{index} privilege allows you to create or drop (remove) indexes. The @strong{alter} privilege allows you to use @code{ALTER TABLE}. The @strong{create} and @strong{drop} privileges allow you to create new databases and tables, or to drop (remove) existing databases and tables. Note that if you grant the @strong{drop} privilege for the @code{mysql} database to a user, that user can drop the database in which the MySQL access privileges are stored! The @strong{grant} privilege allows you to give to other users those privileges you yourself possess. The @strong{file} privilege gives you permission to read and write files on the server using the @code{LOAD DATA INFILE} and @code{SELECT ... INTO OUTFILE} statements. Any user to whom this privilege is granted can read or write any file that the MySQL server can read or write. The remaining privileges are used for administrative operations, which are performed using the @code{mysqladmin} program. The table below shows which @code{mysqladmin} commands each administrative privilege allows you to execute: @multitable @columnfractions .15 .85 @item @strong{Privilege} @tab @strong{Commands permitted to privilege holders} @item @strong{reload} @tab @code{reload}, @code{refresh}, @code{flush-privileges}, @code{flush-hosts}, @code{flush-logs}, and @code{flush-tables} @item @strong{shutdown} @tab @code{shutdown} @item @strong{process} @tab @code{processlist}, @code{kill} @end multitable The @code{reload} command tells the server to re-read the grant tables. The @code{refresh} command flushes all tables and opens and closes the log files. @code{flush-privileges} is a synonym for @code{reload}. The other @code{flush-*} commands perform functions similar to @code{refresh} but are more limited in scope, and may be preferable in some instances. For example, if you want to flush just the log files, @code{flush-logs} is a better choice than @code{refresh}. The @code{shutdown} command shuts down the server. The @code{processlist} command displays information about the threads executing within the server. The @code{kill} command kills server threads. You can always display or kill your own threads, but you need the @strong{process} privilege to display or kill threads initiated by other users. @xref{KILL}. It is a good idea in general to grant privileges only to those users who need them, but you should exercise particular caution in granting certain privileges: @itemize @bullet @item The @strong{grant} privilege allows users to give away their privileges to other users. Two users with different privileges and with the @strong{grant} privilege are able to combine privileges. @item The @strong{alter} privilege may be used to subvert the privilege system by renaming tables. @item The @strong{file} privilege can be abused to read any world-readable file on the server into a database table, the contents of which can then be accessed using @code{SELECT}. This includes the contents of all databases hosted by the server! @item The @strong{shutdown} privilege can be abused to deny service to other users entirely, by terminating the server. @item The @strong{process} privilege can be used to view the plain text of currently executing queries, including queries that set or change passwords. @item Privileges on the @code{mysql} database can be used to change passwords and other access privilege information. (Passwords are stored encrypted, so a malicious user cannot simply read them to know the plain text password). If they can access the @code{mysql.user} password column, they can use it to log into the MySQL server for the given user. (With sufficient privileges, the same user can replace a password with a different one.) @end itemize There are some things that you cannot do with the MySQL privilege system: @itemize @bullet @item You cannot explicitly specify that a given user should be denied access. That is, you cannot explicitly match a user and then refuse the connection. @item You cannot specify that a user has privileges to create or drop tables in a database but not to create or drop the database itself. @end itemize @node Connecting, Connection access, Privileges provided, Privilege system @subsection Connecting to the MySQL Server @cindex connecting, to the server @cindex default hostname @cindex hostname, default @cindex server, connecting MySQL client programs generally require that you specify connection parameters when you want to access a MySQL server: the host you want to connect to, your user name, and your password. For example, the @code{mysql} client can be started like this (optional arguments are enclosed between @samp{[} and @samp{]}): @example shell> mysql [-h host_name] [-u user_name] [-pyour_pass] @end example Alternate forms of the @code{-h}, @code{-u}, and @code{-p} options are @code{--host=host_name}, @code{--user=user_name}, and @code{--password=your_pass}. Note that there is @emph{no space} between @code{-p} or @code{--password=} and the password following it. @strong{NOTE:} Specifying a password on the command line is not secure! Any user on your system may then find out your password by typing a command like: @code{ps auxww}. @xref{Option files}. @code{mysql} uses default values for connection parameters that are missing from the command line: @itemize @bullet @item The default hostname is @code{localhost}. @item The default user name is your Unix login name. @item No password is supplied if @code{-p} is missing. @end itemize Thus, for a Unix user @code{joe}, the following commands are equivalent: @example shell> mysql -h localhost -u joe shell> mysql -h localhost shell> mysql -u joe shell> mysql @end example Other MySQL clients behave similarly. On Unix systems, you can specify different default values to be used when you make a connection, so that you need not enter them on the command line each time you invoke a client program. This can be done in a couple of ways: @itemize @bullet @item @tindex .my.cnf file You can specify connection parameters in the @code{[client]} section of the @file{.my.cnf} configuration file in your home directory. The relevant section of the file might look like this: @example [client] host=host_name user=user_name password=your_pass @end example @xref{Option files}. @item @tindex @code{MYSQL_HOST} environment variable @tindex environment variable, @code{MYSQL_HOST} @tindex @code{MYSQL_PWD} environment variable @tindex environment variable, @code{MYSQL_PWD} @tindex @code{USER} environment variable @tindex environment variable, @code{USER} You can specify connection parameters using environment variables. The host can be specified for @code{mysql} using @code{MYSQL_HOST}. The MySQL user name can be specified using @code{USER} (this is for Windows only). The password can be specified using @code{MYSQL_PWD} (but this is insecure; see the next section). @xref{Environment variables}. @end itemize @node Connection access, Request access, Connecting, Privilege system @subsection Access Control, Stage 1: Connection Verification @cindex access control @cindex control access @cindex connecting, verification @cindex testing, connection to the server When you attempt to connect to a MySQL server, the server accepts or rejects the connection based on your identity and whether or not you can verify your identity by supplying the correct password. If not, the server denies access to you completely. Otherwise, the server accepts the connection, then enters Stage 2 and waits for requests. Your identity is based on two pieces of information: @itemize @bullet @item The host from which you connect @item Your MySQL user name @end itemize Identity checking is performed using the three @code{user} table scope fields (@code{Host}, @code{User}, and @code{Password}). The server accepts the connection only if a @code{user} table entry matches your hostname and user name, and you supply the correct password. Values in the @code{user} table scope fields may be specified as follows: @itemize @bullet @item A @code{Host} value may be a hostname or an IP number, or @code{'localhost'} to indicate the local host. @item @cindex wildcards, in @code{mysql.user} table You can use the wild-card characters @samp{%} and @samp{_} in the @code{Host} field. @item A @code{Host} value of @code{'%'} matches any hostname. @item A blank @code{Host} value means that the privilege should be anded with the entry in the @code{host} table that matches the given host name. You can find more information about this in the next chapter. @cindex netmask notation, in @code{mysql.user} table @item As of MySQL Version 3.23, for @code{Host} values specified as IP numbers, you can specify a netmask indicating how many address bits to use for the network number. For example: @example GRANT ALL PRIVILEGES on db.* to david@@'192.58.197.0/255.255.255.0'; @end example This will allow everyone to connect from an IP where the following is true: @example user_ip & netmask = host_ip. @end example In the above example all IP:s in the interval 192.58.197.0 - 192.58.197.255 can connect to the MySQL server. @item @cindex anonymous user Wild-card characters are not allowed in the @code{User} field, but you can specify a blank value, which matches any name. If the @code{user} table entry that matches an incoming connection has a blank user name, the user is considered to be the anonymous user (the user with no name), rather than the name that the client actually specified. This means that a blank user name is used for all further access checking for the duration of the connection (that is, during Stage 2). @item The @code{Password} field can be blank. This does not mean that any password matches, it means the user must connect without specifying a password. @end itemize @findex PASSWORD() Non-blank @code{Password} values represent encrypted passwords. MySQL does not store passwords in plaintext form for anyone to see. Rather, the password supplied by a user who is attempting to connect is encrypted (using the @code{PASSWORD()} function). The encrypted password is then used when the client/server is checking if the password is correct (This is done without the encrypted password ever traveling over the connection.) Note that from MySQL's point of view the encrypted password is the REAL password, so you should not give anyone access to it! In particular, don't give normal users read access to the tables in the @code{mysql} database! The examples below show how various combinations of @code{Host} and @code{User} values in @code{user} table entries apply to incoming connections: @multitable @columnfractions .25 .15 .60 @item @code{Host} @strong{value} @tab @code{User} @strong{value} @tab @strong{Connections matched by entry} @item @code{'thomas.loc.gov'} @tab @code{'fred'} @tab @code{fred}, connecting from @code{thomas.loc.gov} @item @code{'thomas.loc.gov'} @tab @code{''} @tab Any user, connecting from @code{thomas.loc.gov} @item @code{'%'} @tab @code{'fred'} @tab @code{fred}, connecting from any host @item @code{'%'} @tab @code{''} @tab Any user, connecting from any host @item @code{'%.loc.gov'} @tab @code{'fred'} @tab @code{fred}, connecting from any host in the @code{loc.gov} domain @item @code{'x.y.%'} @tab @code{'fred'} @tab @code{fred}, connecting from @code{x.y.net}, @code{x.y.com},@code{x.y.edu}, etc. (this is probably not useful) @item @code{'144.155.166.177'} @tab @code{'fred'} @tab @code{fred}, connecting from the host with IP address @code{144.155.166.177} @item @code{'144.155.166.%'} @tab @code{'fred'} @tab @code{fred}, connecting from any host in the @code{144.155.166} class C subnet @item @code{'144.155.166.0/255.255.255.0'} @tab @code{'fred'} @tab Same as previous example @end multitable Because you can use IP wild-card values in the @code{Host} field (for example, @code{'144.155.166.%'} to match every host on a subnet), there is the possibility that someone might try to exploit this capability by naming a host @code{144.155.166.somewhere.com}. To foil such attempts, MySQL disallows matching on hostnames that start with digits and a dot. Thus, if you have a host named something like @code{1.2.foo.com}, its name will never match the @code{Host} column of the grant tables. Only an IP number can match an IP wild-card value. An incoming connection may be matched by more than one entry in the @code{user} table. For example, a connection from @code{thomas.loc.gov} by @code{fred} would be matched by several of the entries just shown above. How does the server choose which entry to use if more than one matches? The server resolves this question by sorting the @code{user} table after reading it at startup time, then looking through the entries in sorted order when a user attempts to connect. The first matching entry is the one that is used. @code{user} table sorting works as follows. Suppose the @code{user} table looks like this: @example +-----------+----------+- | Host | User | ... +-----------+----------+- | % | root | ... | % | jeffrey | ... | localhost | root | ... | localhost | | ... +-----------+----------+- @end example When the server reads in the table, it orders the entries with the most-specific @code{Host} values first (@code{'%'} in the @code{Host} column means ``any host'' and is least specific). Entries with the same @code{Host} value are ordered with the most-specific @code{User} values first (a blank @code{User} value means ``any user'' and is least specific). The resulting sorted @code{user} table looks like this: @example +-----------+----------+- | Host | User | ... +-----------+----------+- | localhost | root | ... | localhost | | ... | % | jeffrey | ... | % | root | ... +-----------+----------+- @end example @cindex grant tables, sorting @cindex sorting, grant tables @cindex @code{user} table, sorting When a connection is attempted, the server looks through the sorted entries and uses the first match found. For a connection from @code{localhost} by @code{jeffrey}, the entries with @code{'localhost'} in the @code{Host} column match first. Of those, the entry with the blank user name matches both the connecting hostname and user name. (The @code{'%'/'jeffrey'} entry would have matched, too, but it is not the first match in the table.) Here is another example. Suppose the @code{user} table looks like this: @example +----------------+----------+- | Host | User | ... +----------------+----------+- | % | jeffrey | ... | thomas.loc.gov | | ... +----------------+----------+- @end example The sorted table looks like this: @example +----------------+----------+- | Host | User | ... +----------------+----------+- | thomas.loc.gov | | ... | % | jeffrey | ... +----------------+----------+- @end example A connection from @code{thomas.loc.gov} by @code{jeffrey} is matched by the first entry, whereas a connection from @code{whitehouse.gov} by @code{jeffrey} is matched by the second. A common misconception is to think that for a given user name, all entries that explicitly name that user will be used first when the server attempts to find a match for the connection. This is simply not true. The previous example illustrates this, where a connection from @code{thomas.loc.gov} by @code{jeffrey} is first matched not by the entry containing @code{'jeffrey'} as the @code{User} field value, but by the entry with no user name! If you have problems connecting to the server, print out the @code{user} table and sort it by hand to see where the first match is being made. @node Request access, Access denied, Connection access, Privilege system @subsection Access Control, Stage 2: Request Verification Once you establish a connection, the server enters Stage 2. For each request that comes in on the connection, the server checks whether you have sufficient privileges to perform it, based on the type of operation you wish to perform. This is where the privilege fields in the grant tables come into play. These privileges can come from any of the @code{user}, @code{db}, @code{host}, @code{tables_priv}, or @code{columns_priv} tables. The grant tables are manipulated with @code{GRANT} and @code{REVOKE} commands. @xref{GRANT, , @code{GRANT}}. (You may find it helpful to refer to @ref{Privileges}, which lists the fields present in each of the grant tables.) The @code{user} table grants privileges that are assigned to you on a global basis and that apply no matter what the current database is. For example, if the @code{user} table grants you the @strong{delete} privilege, you can delete rows from any database on the server host! In other words, @code{user} table privileges are superuser privileges. It is wise to grant privileges in the @code{user} table only to superusers such as server or database administrators. For other users, you should leave the privileges in the @code{user} table set to @code{'N'} and grant privileges on a database-specific basis only, using the @code{db} and @code{host} tables. @cindex anonymous user @cindex wild cards, in @code{mysql.db} table @cindex wild cards, in @code{mysql.host} table The @code{db} and @code{host} tables grant database-specific privileges. Values in the scope fields may be specified as follows: @itemize @bullet @item The wild-card characters @samp{%} and @samp{_} can be used in the @code{Host} and @code{Db} fields of either table. @item A @code{'%'} @code{Host} value in the @code{db} table means ``any host.'' A blank @code{Host} value in the @code{db} table means ``consult the @code{host} table for further information.'' @item A @code{'%'} or blank @code{Host} value in the @code{host} table means ``any host.'' @item A @code{'%'} or blank @code{Db} value in either table means ``any database.'' @item A blank @code{User} value in either table matches the anonymous user. @end itemize @cindex grant tables, sorting @cindex sorting, grant tables @cindex @code{db} table, sorting @cindex @code{host} table, sorting The @code{db} and @code{host} tables are read in and sorted when the server starts up (at the same time that it reads the @code{user} table). The @code{db} table is sorted on the @code{Host}, @code{Db}, and @code{User} scope fields, and the @code{host} table is sorted on the @code{Host} and @code{Db} scope fields. As with the @code{user} table, sorting puts the most-specific values first and least-specific values last, and when the server looks for matching entries, it uses the first match that it finds. @cindex wild cards, in @code{mysql.tables_priv} table @cindex wild cards, in @code{mysql.columns_priv} table The @code{tables_priv} and @code{columns_priv} tables grant table- and column-specific privileges. Values in the scope fields may be specified as follows: @itemize @bullet @item The wild-card characters @samp{%} and @samp{_} can be used in the @code{Host} field of either table. @item A @code{'%'} or blank @code{Host} value in either table means ``any host.'' @item The @code{Db}, @code{Table_name} and @code{Column_name} fields cannot contain wild cards or be blank in either table. @end itemize The @code{tables_priv} and @code{columns_priv} tables are sorted on the @code{Host}, @code{Db}, and @code{User} fields. This is similar to @code{db} table sorting, although the sorting is simpler because only the @code{Host} field may contain wild cards. The request verification process is described below. (If you are familiar with the access-checking source code, you will notice that the description here differs slightly from the algorithm used in the code. The description is equivalent to what the code actually does; it differs only to make the explanation simpler.) For administrative requests (@strong{shutdown}, @strong{reload}, etc.), the server checks only the @code{user} table entry, because that is the only table that specifies administrative privileges. Access is granted if the entry allows the requested operation and denied otherwise. For example, if you want to execute @code{mysqladmin shutdown} but your @code{user} table entry doesn't grant the @strong{shutdown} privilege to you, access is denied without even checking the @code{db} or @code{host} tables. (They contain no @code{Shutdown_priv} column, so there is no need to do so.) For database-related requests (@strong{insert}, @strong{update}, etc.), the server first checks the user's global (superuser) privileges by looking in the @code{user} table entry. If the entry allows the requested operation, access is granted. If the global privileges in the @code{user} table are insufficient, the server determines the user's database-specific privileges by checking the @code{db} and @code{host} tables: @enumerate @item The server looks in the @code{db} table for a match on the @code{Host}, @code{Db}, and @code{User} fields. The @code{Host} and @code{User} fields are matched to the connecting user's hostname and MySQL user name. The @code{Db} field is matched to the database the user wants to access. If there is no entry for the @code{Host} and @code{User}, access is denied. @item If there is a matching @code{db} table entry and its @code{Host} field is not blank, that entry defines the user's database-specific privileges. @item If the matching @code{db} table entry's @code{Host} field is blank, it signifies that the @code{host} table enumerates which hosts should be allowed access to the database. In this case, a further lookup is done in the @code{host} table to find a match on the @code{Host} and @code{Db} fields. If no @code{host} table entry matches, access is denied. If there is a match, the user's database-specific privileges are computed as the intersection (@emph{not} the union!) of the privileges in the @code{db} and @code{host} table entries, that is, the privileges that are @code{'Y'} in both entries. (This way you can grant general privileges in the @code{db} table entry and then selectively restrict them on a host-by-host basis using the @code{host} table entries.) @end enumerate After determining the database-specific privileges granted by the @code{db} and @code{host} table entries, the server adds them to the global privileges granted by the @code{user} table. If the result allows the requested operation, access is granted. Otherwise, the server checks the user's table and column privileges in the @code{tables_priv} and @code{columns_priv} tables and adds those to the user's privileges. Access is allowed or denied based on the result. Expressed in boolean terms, the preceding description of how a user's privileges are calculated may be summarized like this: @example global privileges OR (database privileges AND host privileges) OR table privileges OR column privileges @end example It may not be apparent why, if the global @code{user} entry privileges are initially found to be insufficient for the requested operation, the server adds those privileges to the database-, table-, and column-specific privileges later. The reason is that a request might require more than one type of privilege. For example, if you execute an @code{INSERT ... SELECT} statement, you need both @strong{insert} and @strong{select} privileges. Your privileges might be such that the @code{user} table entry grants one privilege and the @code{db} table entry grants the other. In this case, you have the necessary privileges to perform the request, but the server cannot tell that from either table by itself; the privileges granted by the entries in both tables must be combined. @cindex @code{host} table @cindex tables, @code{host} The @code{host} table can be used to maintain a list of secure servers. At TcX, the @code{host} table contains a list of all machines on the local network. These are granted all privileges. You can also use the @code{host} table to indicate hosts that are @emph{not} secure. Suppose you have a machine @code{public.your.domain} that is located in a public area that you do not consider secure. You can allow access to all hosts on your network except that machine by using @code{host} table entries like this: @example +--------------------+----+- | Host | Db | ... +--------------------+----+- | public.your.domain | % | ... (all privileges set to 'N') | %.your.domain | % | ... (all privileges set to 'Y') +--------------------+----+- @end example @cindex privilege, changes @cindex changes to privileges @cindex tables, grant @cindex grant tables Naturally, you should always test your entries in the grant tables (for example, using @code{mysqlaccess}) to make sure your access privileges are actually set up the way you think they are. @node Access denied, , Request access, Privilege system @subsection Causes of @code{Access denied} Errors If you encounter @code{Access denied} errors when you try to connect to the MySQL server, the list below indicates some courses of action you can take to correct the problem: @itemize @bullet @item After installing MySQL, did you run the @code{mysql_install_db} script to set up the initial grant table contents? If not, do so. @xref{Default privileges}. Test the initial privileges by executing this command: @example shell> mysql -u root test @end example The server should let you connect without error. You should also make sure you have a file @file{user.MYD} in the MySQL database directory. Ordinarily, this is @file{PATH/var/mysql/user.MYD}, where @code{PATH} is the pathname to the MySQL installation root. @item After a fresh installation, you should connect to the server and set up your users and their access permissions: @example shell> mysql -u root mysql @end example The server should let you connect because the MySQL @code{root} user has no password initially. That is also a security risk, so setting the @code{root} password is something you should do while you're setting up your other MySQL users. If you try to connect as @code{root} and get this error: @example Access denied for user: '@@unknown' to database mysql @end example this means that you don't have an entry in the @code{user} table with a @code{User} column value of @code{'root'} and that @code{mysqld} cannot resolve the hostname for your client. In this case, you must restart the server with the @code{--skip-grant-tables} option and edit your @file{/etc/hosts} or @file{\windows\hosts} file to add an entry for your host. @item If you get an error like the following: @example shell> mysqladmin -u root -pxxxx ver Access denied for user: 'root@@localhost' (Using password: YES) @end example It means that you are using a wrong password. @xref{Passwords}. If you have forgot the root password, you can restart @code{mysqld} with @code{--skip-grant-tables} to change the password. You can find more about this option later on in this manual section. If you get the above error even if you haven't specified a password, this means that you a wrong password in some @code{my.ini} file. @xref{Option files}. You can avoid using option files with the @code{--no-defaults} option, as follows: @example shell> mysqladmin --no-defaults -u root ver @end example @item @cindex @code{mysql_fix_privilege_tables} If you updated an existing MySQL installation from a version earlier than Version 3.22.11 to Version 3.22.11 or later, did you run the @code{mysql_fix_privilege_tables} script? If not, do so. The structure of the grant tables changed with MySQL Version 3.22.11 when the @code{GRANT} statement became functional. @item If your privileges seem to have changed in the middle of a session, it may be that a superuser has changed them. Reloading the grant tables affects new client connections, but it also affects existing connections as indicated in @ref{Privilege changes}. @item If you can't get your password to work, remember that you must use the @code{PASSWORD()} function if you set the password with the @code{INSERT}, @code{UPDATE}, or @code{SET PASSWORD} statements. The @code{PASSWORD()} function is unnecessary if you specify the password using the @code{GRANT ... INDENTIFIED BY} statement or the @code{mysqladmin password} command. @xref{Passwords}. @item @code{localhost} is a synonym for your local hostname, and is also the default host to which clients try to connect if you specify no host explicitly. However, connections to @code{localhost} do not work if you are running on a system that uses MIT-pthreads (@code{localhost} connections are made using Unix sockets, which are not supported by MIT-pthreads). To avoid this problem on such systems, you should use the @code{--host} option to name the server host explicitly. This will make a TCP/IP connection to the @code{mysqld} server. In this case, you must have your real hostname in @code{user} table entries on the server host. (This is true even if you are running a client program on the same host as the server.) @item If you get an @code{Access denied} error when trying to connect to the database with @code{mysql -u user_name db_name}, you may have a problem with the @code{user} table. Check this by executing @code{mysql -u root mysql} and issuing this SQL statement: @example mysql> SELECT * FROM user; @end example The result should include an entry with the @code{Host} and @code{User} columns matching your computer's hostname and your MySQL user name. @item The @code{Access denied} error message will tell you who you are trying to log in as, the host from which you are trying to connect, and whether or not you were using a password. Normally, you should have one entry in the @code{user} table that exactly matches the hostname and user name that were given in the error message. For example if you get an error message that contains @code{Using password: NO}, this means that you tried to login without an password. @item If you get the following error when you try to connect from a different host than the one on which the MySQL server is running, then there is no row in the @code{user} table that matches that host: @example Host ... is not allowed to connect to this MySQL server @end example You can fix this by using the command-line tool @code{mysql} (on the server host!) to add a row to the @code{user}, @code{db}, or @code{host} table for the user/hostname combination from which you are trying to connect and then execute @code{mysqladmin flush-privileges}. If you are not running MySQL Version 3.22 and you don't know the IP number or hostname of the machine from which you are connecting, you should put an entry with @code{'%'} as the @code{Host} column value in the @code{user} table and restart @code{mysqld} with the @code{--log} option on the server machine. After trying to connect from the client machine, the information in the MySQL log will indicate how you really did connect. (Then replace the @code{'%'} in the @code{user} table entry with the actual hostname that shows up in the log. Otherwise, you'll have a system that is insecure.) Another reason for this error on Linux is that you are using a binary MySQL version that is compiled with a different glibc version than the one you are using. In this case you should either upgrade your OS/glibc or download the source MySQL version and compile this yourself. A source RPM is normally trivial to compile and install, so this isn't a big problem. @item If you get an error message where the hostname is not shown or where the hostname is an IP, even if you try to connect with a hostname: @example shell> mysqladmin -u root -pxxxx -h some-hostname ver Access denied for user: 'root@' (Using password: YES) @end example This means that MySQL got some error when trying to resolve the IP to a hostname. In this case you can execute @code{mysqladmin flush-hosts} to reset the internal DNS cache. @xref{DNS}. Some permanent solutions are: @itemize @minus @item Try to find out what is wrong with your DNS server and fix this. @item Specify IPs instead of hostnames in the MySQL privilege tables. @item Start @code{mysqld} with @code{--skip-name-resolve}. @item Start @code{mysqld} with @code{--skip-host-cache}. @item Connect to @code{localhost} if you are running the server and the client on the same machine. @item Put the client machine names in @code{/etc/hosts}. @end itemize @item If @code{mysql -u root test} works but @code{mysql -h your_hostname -u root test} results in @code{Access denied}, then you may not have the correct name for your host in the @code{user} table. A common problem here is that the @code{Host} value in the user table entry specifies an unqualified hostname, but your system's name resolution routines return a fully qualified domain name (or vice-versa). For example, if you have an entry with host @code{'tcx'} in the @code{user} table, but your DNS tells MySQL that your hostname is @code{'tcx.subnet.se'}, the entry will not work. Try adding an entry to the @code{user} table that contains the IP number of your host as the @code{Host} column value. (Alternatively, you could add an entry to the @code{user} table with a @code{Host} value that contains a wild card---for example, @code{'tcx.%'}. However, use of hostnames ending with @samp{%} is @emph{insecure} and is @emph{not} recommended!) @item If @code{mysql -u user_name test} works but @code{mysql -u user_name other_db_name} doesn't work, you don't have an entry for @code{other_db_name} listed in the @code{db} table. @item If @code{mysql -u user_name db_name} works when executed on the server machine, but @code{mysql -u host_name -u user_name db_name} doesn't work when executed on another client machine, you don't have the client machine listed in the @code{user} table or the @code{db} table. @item If you can't figure out why you get @code{Access denied}, remove from the @code{user} table all entries that have @code{Host} values containing wild cards (entries that contain @samp{%} or @samp{_}). A very common error is to insert a new entry with @code{Host}=@code{'%'} and @code{User}=@code{'some user'}, thinking that this will allow you to specify @code{localhost} to connect from the same machine. The reason that this doesn't work is that the default privileges include an entry with @code{Host}=@code{'localhost'} and @code{User}=@code{''}. Because that entry has a @code{Host} value @code{'localhost'} that is more specific than @code{'%'}, it is used in preference to the new entry when connecting from @code{localhost}! The correct procedure is to insert a second entry with @code{Host}=@code{'localhost'} and @code{User}=@code{'some_user'}, or to remove the entry with @code{Host}=@code{'localhost'} and @code{User}=@code{''}. @item If you get the following error, you may have a problem with the @code{db} or @code{host} table: @example Access to database denied @end example If the entry selected from the @code{db} table has an empty value in the @code{Host} column, make sure there are one or more corresponding entries in the @code{host} table specifying which hosts the @code{db} table entry applies to. If you get the error when using the SQL commands @code{SELECT ... INTO OUTFILE} or @code{LOAD DATA INFILE}, your entry in the @code{user} table probably doesn't have the @strong{file} privilege enabled. @item @cindex configuration files @cindex environment variables @tindex .my.cnf file Remember that client programs will use connection parameters specified in configuration files or environment variables. @xref{Environment variables}. If a client seems to be sending the wrong default connection parameters when you don't specify them on the command line, check your environment and the @file{.my.cnf} file in your home directory. You might also check the system-wide MySQL configuration files, though it is far less likely that client connection parameters will be specified there. @xref{Option files}. If you get @code{Access denied} when you run a client without any options, make sure you haven't specified an old password in any of your option files! @xref{Option files}. @item If you make changes to the grant tables directly (using an @code{INSERT} or @code{UPDATE} statement) and your changes seem to be ignored, remember that you must issue a @code{FLUSH PRIVILEGES} statement or execute a @code{mysqladmin flush-privileges} command to cause the server to re-read the privilege tables. Otherwise your changes have no effect until the next time the server is restarted. Remember that after you set the @code{root} password with an @code{UPDATE} command, you won't need to specify it until after you flush the privileges, because the server won't know you've changed the password yet! @item If you have access problems with a Perl, PHP, Python, or ODBC program, try to connect to the server with @code{mysql -u user_name db_name} or @code{mysql -u user_name -pyour_pass db_name}. If you are able to connect using the @code{mysql} client, there is a problem with your program and not with the access privileges. (Note that there is no space between @code{-p} and the password; you can also use the @code{--password=your_pass} syntax to specify the password. If you use the @code{-p} option alone, MySQL will prompt you for the password.) @item For testing, start the @code{mysqld} daemon with the @code{--skip-grant-tables} option. Then you can change the MySQL grant tables and use the @code{mysqlaccess} script to check whether or not your modifications have the desired effect. When you are satisfied with your changes, execute @code{mysqladmin flush-privileges} to tell the @code{mysqld} server to start using the new grant tables. @strong{Note:} Reloading the grant tables overrides the @code{--skip-grant-tables} option. This allows you to tell the server to begin using the grant tables again without bringing it down and restarting it. @item If everything else fails, start the @code{mysqld} daemon with a debugging option (for example, @code{--debug=d,general,query}). This will print host and user information about attempted connections, as well as information about each command issued. @xref{Making trace files}. @item If you have any other problems with the MySQL grant tables and feel you must post the problem to the mailing list, always provide a dump of the MySQL grant tables. You can dump the tables with the @code{mysqldump mysql} command. As always, post your problem using the @code{mysqlbug} script. @xref{Bug reports}. In some cases you may need to restart @code{mysqld} with @code{--skip-grant-tables} to run @code{mysqldump}. @end itemize @node User Account Management, Disaster Prevention, Privilege system, MySQL Database Administration @section MySQL User Account Management @menu * GRANT:: @code{GRANT} and @code{REVOKE} Syntax * User names:: MySQL User Names and Passwords * Privilege changes:: When Privilege Changes Take Effect * Default privileges:: Setting Up the Initial MySQL Privileges * Adding users:: Adding New Users to MySQL * Passwords:: Setting Up Passwords * Password security:: Keeping Your Password Secure @end menu @node GRANT, User names, User Account Management, User Account Management @subsection @code{GRANT} and @code{REVOKE} Syntax @findex GRANT @findex REVOKE @cindex privileges, granting @cindex privileges, revoking @cindex global privileges @cindex revoking, privleges @cindex granting, privleges @example GRANT priv_type [(column_list)] [, priv_type [(column_list)] ...] ON @{tbl_name | * | *.* | db_name.*@} TO user_name [IDENTIFIED BY 'password'] [, user_name [IDENTIFIED BY 'password'] ...] [WITH GRANT OPTION] REVOKE priv_type [(column_list)] [, priv_type [(column_list)] ...] ON @{tbl_name | * | *.* | db_name.*@} FROM user_name [, user_name ...] @end example @code{GRANT} is implemented in MySQL Version 3.22.11 or later. For earlier MySQL versions, the @code{GRANT} statement does nothing. The @code{GRANT} and @code{REVOKE} commands allow system administrators to create users and grant and revoke rights to MySQL users at four privilege levels: @table @strong @item Global level Global privileges apply to all databases on a given server. These privileges are stored in the @code{mysql.user} table. @item Database level Database privileges apply to all tables in a given database. These privileges are stored in the @code{mysql.db} and @code{mysql.host} tables. @item Table level Table privileges apply to all columns in a given table. These privileges are stored in the @code{mysql.tables_priv} table. @item Column level Column privileges apply to single columns in a given table. These privileges are stored in the @code{mysql.columns_priv} table. @end table If you give a grant for a users that doesn't exists, that user is created. For examples of how @code{GRANT} works, see @ref{Adding users}. For the @code{GRANT} and @code{REVOKE} statements, @code{priv_type} may be specified as any of the following: @example ALL PRIVILEGES FILE RELOAD ALTER INDEX SELECT CREATE INSERT SHUTDOWN DELETE PROCESS UPDATE DROP REFERENCES USAGE @end example @code{ALL} is a synonym for @code{ALL PRIVILEGES}. @code{REFERENCES} is not yet implemented. @code{USAGE} is currently a synonym for ``no privileges.'' It can be used when you want to create a user that has no privileges. To revoke the @strong{grant} privilege from a user, use a @code{priv_type} value of @code{GRANT OPTION}: @example REVOKE GRANT OPTION ON ... FROM ...; @end example The only @code{priv_type} values you can specify for a table are @code{SELECT}, @code{INSERT}, @code{UPDATE}, @code{DELETE}, @code{CREATE}, @code{DROP}, @code{GRANT}, @code{INDEX}, and @code{ALTER}. The only @code{priv_type} values you can specify for a column (that is, when you use a @code{column_list} clause) are @code{SELECT}, @code{INSERT}, and @code{UPDATE}. You can set global privileges by using @code{ON *.*} syntax. You can set database privileges by using @code{ON db_name.*} syntax. If you specify @code{ON *} and you have a current database, you will set the privileges for that database. (@strong{WARNING:} If you specify @code{ON *} and you @emph{don't} have a current database, you will affect the global privileges!) In order to accommodate granting rights to users from arbitrary hosts, MySQL supports specifying the @code{user_name} value in the form @code{user@@host}. If you want to specify a @code{user} string containing special characters (such as @samp{-}), or a @code{host} string containing special characters or wild-card characters (such as @samp{%}), you can quote the user or host name (for example, @code{'test-user'@@'test-hostname'}). You can specify wild cards in the hostname. For example, @code{user@@"%.loc.gov"} applies to @code{user} for any host in the @code{loc.gov} domain, and @code{user@@"144.155.166.%"} applies to @code{user} for any host in the @code{144.155.166} class C subnet. The simple form @code{user} is a synonym for @code{user@@"%"}. @strong{NOTE:} If you allow anonymous users to connect to the MySQL server (which is the default), you should also add all local users as @code{user@@localhost} because otherwise the anonymous user entry for the local host in the @code{mysql.user} table will be used when the user tries to log into the MySQL server from the local machine! Anonymous users are defined by inserting entries with @code{User=''} into the @code{mysql.user} table. You can verify if this applies to you by executing this query: @example mysql> SELECT Host,User FROM mysql.user WHERE User=''; @end example For the moment, @code{GRANT} only supports host, table, database, and column names up to 60 characters long. A user name can be up to 16 characters. The privileges for a table or column are formed from the logical OR of the privileges at each of the four privilege levels. For example, if the @code{mysql.user} table specifies that a user has a global @strong{select} privilege, this can't be denied by an entry at the database, table, or column level. The privileges for a column can be calculated as follows: @example global privileges OR (database privileges AND host privileges) OR table privileges OR column privileges @end example In most cases, you grant rights to a user at only one of the privilege levels, so life isn't normally as complicated as above. The details of the privilege-checking procedure are presented in @ref{Privilege system}. If you grant privileges for a user/hostname combination that does not exist in the @code{mysql.user} table, an entry is added and remains there until deleted with a @code{DELETE} command. In other words, @code{GRANT} may create @code{user} table entries, but @code{REVOKE} will not remove them; you must do that explicitly using @code{DELETE}. @cindex passwords, setting In MySQL Version 3.22.12 or later, if a new user is created or if you have global grant privileges, the user's password will be set to the password specified by the @code{IDENTIFIED BY} clause, if one is given. If the user already had a password, it is replaced by the new one. @strong{WARNING:} If you create a new user but do not specify an @code{IDENTIFIED BY} clause, the user has no password. This is insecure. Passwords can also be set with the @code{SET PASSWORD} command. @xref{SET OPTION, , @code{SET OPTION}}. If you grant privileges for a database, an entry in the @code{mysql.db} table is created if needed. When all privileges for the database have been removed with @code{REVOKE}, this entry is deleted. If a user doesn't have any privileges on a table, the table is not displayed when the user requests a list of tables (for example, with a @code{SHOW TABLES} statement). The @code{WITH GRANT OPTION} clause gives the user the ability to give to other users any privileges the user has at the specified privilege level. You should be careful to whom you give the @strong{grant} privilege, as two users with different privileges may be able to join privileges! You cannot grant another user a privilege you don't have yourself; the @strong{grant} privilege allows you to give away only those privileges you possess. Be aware that when you grant a user the @strong{grant} privilege at a particular privilege level, any privileges the user already possesses (or is given in the future!) at that level are also grantable by that user. Suppose you grant a user the @strong{insert} privilege on a database. If you then grant the @strong{select} privilege on the database and specify @code{WITH GRANT OPTION}, the user can give away not only the @strong{select} privilege, but also @strong{insert}. If you then grant the @strong{update} privilege to the user on the database, the user can give away the @strong{insert}, @strong{select} and @strong{update}. You should not grant @strong{alter} privileges to a normal user. If you do that, the user can try to subvert the privilege system by renaming tables! Note that if you are using table or column privileges for even one user, the server examines table and column privileges for all users and this will slow down MySQL a bit. When @code{mysqld} starts, all privileges are read into memory. Database, table, and column privileges take effect at once, and user-level privileges take effect the next time the user connects. Modifications to the grant tables that you perform using @code{GRANT} or @code{REVOKE} are noticed by the server immediately. If you modify the grant tables manually (using @code{INSERT}, @code{UPDATE}, etc.), you should execute a @code{FLUSH PRIVILEGES} statement or run @code{mysqladmin flush-privileges} to tell the server to reload the grant tables. @xref{Privilege changes}. @cindex ANSI SQL, differences from The biggest differences between the ANSI SQL and MySQL versions of @code{GRANT} are: @itemize @bullet @item In MySQL privileges are given for an username + hostname combination and not only for an username. @item ANSI SQL doesn't have global or database-level privileges, and ANSI SQL doesn't support all privilege types that MySQL supports. MySQL doesn't support the ANSI SQL @code{TRIGGER}, @code{EXECUTE} or @code{UNDER} privileges. @item ANSI SQL privileges are structured in a hierarchal manner. If you remove an user, all privileges the user has granted are revoked. In MySQL the granted privileges are not automatically revoked, but you have to revoke these yourself if needed. @item If you in MySQL have the @code{INSERT} grant on only part of the columns in a table, you can execute @code{INSERT} statements on the table; The columns for which you don't have the @code{INSERT} privilege will set to their default values. ANSI SQL requires you to have the @code{INSERT} privilege on all columns. @item When you drop a table in ANSI SQL, all privileges for the table are revoked. If you revoke a privilege in ANSI SQL, all privileges that were granted based on this privilege are also revoked. In MySQL, privileges can be dropped only with explicit @code{REVOKE} commands or by manipulating the MySQL grant tables. @end itemize @node User names, Privilege changes, GRANT, User Account Management @subsection MySQL User Names and Passwords @cindex user names, and passwords @cindex passwords, for users There are several distinctions between the way user names and passwords are used by MySQL and the way they are used by Unix or Windows: @itemize @bullet @item User names, as used by MySQL for authentication purposes, have nothing to do with Unix user names (login names) or Windows user names. Most MySQL clients by default try to log in using the current Unix user name as the MySQL user name, but that is for convenience only. Client programs allow a different name to be specified with the @code{-u} or @code{--user} options. This means that you can't make a database secure in any way unless all MySQL user names have passwords. Anyone may attempt to connect to the server using any name, and they will succeed if they specify any name that doesn't have a password. @item MySQL user names can be up to 16 characters long; Unix user names typically are limited to 8 characters. @item MySQL passwords have nothing to do with Unix passwords. There is no necessary connection between the password you use to log in to a Unix machine and the password you use to access a database on that machine. @item MySQL encrypts passwords using a different algorithm than the one used during the Unix login process. See the descriptions of the @code{PASSWORD()} and @code{ENCRYPT()} functions in @ref{Miscellaneous functions}. Note that even if the password is stored 'scrambled', and knowing your 'scrambled' password is enough to be able to connect to the MySQL server! @end itemize MySQL users and their privileges are normally created with the @code{GRANT} command. @xref{GRANT}. When you login to a MySQL server with a command line client you should specify the password with @code{--password=your-password}. @xref{Connecting}. @example mysql --user=monty --password=guess database_name @end example If you want the client to prompt for a password, you should use @code{--password} without any argument @example mysql --user=monty --password database_name @end example or the short form: @example mysql -u monty -p database_name @end example Note that in the last example the password is @strong{NOT} 'database_name'. If you want to use the @code{-p} option to supply a password you should do so like this: @example mysql -u monty -pguess database_name @end example On some systems, the library call that MySQL uses to prompt for a password will automatically cut the password to 8 characters. Internally MySQL doesn't have any limit for the length of the password. @node Privilege changes, Default privileges, User names, User Account Management @subsection When Privilege Changes Take Effect When @code{mysqld} starts, all grant table contents are read into memory and become effective at that point. Modifications to the grant tables that you perform using @code{GRANT}, @code{REVOKE}, or @code{SET PASSWORD} are noticed by the server immediately. If you modify the grant tables manually (using @code{INSERT}, @code{UPDATE}, etc.), you should execute a @code{FLUSH PRIVILEGES} statement or run @code{mysqladmin flush-privileges} or @code{mysqladmin reload} to tell the server to reload the grant tables. Otherwise your changes will have @emph{no effect} until you restart the server. If you change the grant tables manually but forget to reload the privileges, you will be wondering why your changes don't seem to make any difference! When the server notices that the grant tables have been changed, existing client connections are affected as follows: @itemize @bullet @item Table and column privilege changes take effect with the client's next request. @item Database privilege changes take effect at the next @code{USE db_name} command. @end itemize Global privilege changes and password changes take effect the next time the client connects. @node Default privileges, Adding users, Privilege changes, User Account Management @subsection Setting Up the Initial MySQL Privileges @cindex privileges, default @cindex default, privileges @cindex root password @cindex superuser @cindex users, root @cindex anonymous user @cindex password, root user After installing MySQL, you set up the initial access privileges by running @code{scripts/mysql_install_db}. @xref{Quick install}. The @code{mysql_install_db} script starts up the @code{mysqld} server, then initializes the grant tables to contain the following set of privileges: @itemize @bullet @item The MySQL @code{root} user is created as a superuser who can do anything. Connections must be made from the local host. @strong{NOTE:} The initial @code{root} password is empty, so anyone can connect as @code{root} @emph{without a password} and be granted all privileges. @item @cindex anonymous user An anonymous user is created that can do anything with databases that have a name of @code{'test'} or starting with @code{'test_'}. Connections must be made from the local host. This means any local user can connect without a password and be treated as the anonymous user. @item Other privileges are denied. For example, normal users can't use @code{mysqladmin shutdown} or @code{mysqladmin processlist}. @end itemize @strong{NOTE:} The default privileges are different for Windows. @xref{Windows running}. Because your installation is initially wide open, one of the first things you should do is specify a password for the MySQL @code{root} user. You can do this as follows (note that you specify the password using the @code{PASSWORD()} function): @example shell> mysql -u root mysql mysql> UPDATE user SET Password=PASSWORD('new_password') WHERE user='root'; mysql> FLUSH PRIVILEGES; @end example You can, in MySQL Version 3.22 and above, use the @code{SET PASSWORD} statement: @example shell> mysql -u root mysql mysql> SET PASSWORD FOR root=PASSWORD('new_password'); @end example Another way to set the password is by using the @code{mysqladmin} command: @example shell> mysqladmin -u root password new_password @end example Only users with write/update access to the @code{mysql} database can change the password for others users. All normal users (not anonymous ones) can only change their own password with either of the above commands or with @code{SET PASSWORD=PASSWORD('new password')}. Note that if you update the password in the @code{user} table directly using the first method, you must tell the server to re-read the grant tables (with @code{FLUSH PRIVILEGES}), because the change will go unnoticed otherwise. Once the @code{root} password has been set, thereafter you must supply that password when you connect to the server as @code{root}. You may wish to leave the @code{root} password blank so that you don't need to specify it while you perform additional setup or testing. However, be sure to set it before using your installation for any real production work. See the @code{scripts/mysql_install_db} script to see how it sets up the default privileges. You can use this as a basis to see how to add other users. If you want the initial privileges to be different than those just described above, you can modify @code{mysql_install_db} before you run it. @cindex grant tables, re-creating @cindex re-creating, grant tables To re-create the grant tables completely, remove all the @file{.frm}, @file{.MYI}, and @file{.MYD} files in the directory containing the @code{mysql} database. (This is the directory named @file{mysql} under the database directory, which is listed when you run @code{mysqld --help}.) Then run the @code{mysql_install_db} script, possibly after editing it first to have the privileges you want. @strong{NOTE:} For MySQL versions older than Version 3.22.10, you should NOT delete the @file{.frm} files. If you accidentally do this, you should copy them back from your MySQL distribution before running @code{mysql_install_db}. @node Adding users, Passwords, Default privileges, User Account Management @subsection Adding New Users to MySQL @findex GRANT statement @findex statements, GRANT @cindex privileges, adding @cindex adding, new user privileges @cindex user privileges, adding You can add users two different ways: by using @code{GRANT} statements or by manipulating the MySQL grant tables directly. The preferred method is to use @code{GRANT} statements, because they are more concise and less error-prone. @xref{GRANT}. There is also a lot of contributed programs like @code{phpmyadmin} that can be used to create and administrate users. @xref{Contrib}. The examples below show how to use the @code{mysql} client to set up new users. These examples assume that privileges are set up according to the defaults described in the previous section. This means that to make changes, you must be on the same machine where @code{mysqld} is running, you must connect as the MySQL @code{root} user, and the @code{root} user must have the @strong{insert} privilege for the @code{mysql} database and the @strong{reload} administrative privilege. Also, if you have changed the @code{root} user password, you must specify it for the @code{mysql} commands below. You can add new users by issuing @code{GRANT} statements: @example shell> mysql --user=root mysql mysql> GRANT ALL PRIVILEGES ON *.* TO monty@@localhost IDENTIFIED BY 'some_pass' WITH GRANT OPTION; mysql> GRANT ALL PRIVILEGES ON *.* TO monty@@"%" IDENTIFIED BY 'some_pass' WITH GRANT OPTION; mysql> GRANT RELOAD,PROCESS ON *.* TO admin@@localhost; mysql> GRANT USAGE ON *.* TO dummy@@localhost; @end example These @code{GRANT} statements set up three new users: @table @code @item monty A full superuser who can connect to the server from anywhere, but who must use a password @code{'some_pass'} to do so. Note that we must issue @code{GRANT} statements for both @code{monty@@localhost} and @code{monty@@"%"}. If we don't add the entry with @code{localhost}, the anonymous user entry for @code{localhost} that is created by @code{mysql_install_db} will take precedence when we connect from the local host, because it has a more specific @code{Host} field value and thus comes earlier in the @code{user} table sort order. @item admin A user who can connect from @code{localhost} without a password and who is granted the @strong{reload} and @strong{process} administrative privileges. This allows the user to execute the @code{mysqladmin reload}, @code{mysqladmin refresh}, and @code{mysqladmin flush-*} commands, as well as @code{mysqladmin processlist} . No database-related privileges are granted. (They can be granted later by issuing additional @code{GRANT} statements.) @item dummy A user who can connect without a password, but only from the local host. The global privileges are all set to @code{'N'} --- the @code{USAGE} privilege type allows you to create a user with no privileges. It is assumed that you will grant database-specific privileges later. @end table @findex INSERT statement, grant privileges @findex statements, INSERT You can also add the same user access information directly by issuing @code{INSERT} statements and then telling the server to reload the grant tables: @example shell> mysql --user=root mysql mysql> INSERT INTO user VALUES('localhost','monty',PASSWORD('some_pass'), 'Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y'); mysql> INSERT INTO user VALUES('%','monty',PASSWORD('some_pass'), 'Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y'); mysql> INSERT INTO user SET Host='localhost',User='admin', Reload_priv='Y', Process_priv='Y'; mysql> INSERT INTO user (Host,User,Password) VALUES('localhost','dummy',''); mysql> FLUSH PRIVILEGES; @end example Depending on your MySQL version, you may have to use a different number of @code{'Y'} values above (versions prior to Version 3.22.11 had fewer privilege columns). For the @code{admin} user, the more readable extended @code{INSERT} syntax that is available starting with Version 3.22.11 is used. Note that to set up a superuser, you need only create a @code{user} table entry with the privilege fields set to @code{'Y'}. No @code{db} or @code{host} table entries are necessary. The privilege columns in the @code{user} table were not set explicitly in the last @code{INSERT} statement (for the @code{dummy} user), so those columns are assigned the default value of @code{'N'}. This is the same thing that @code{GRANT USAGE} does. The following example adds a user @code{custom} who can connect from hosts @code{localhost}, @code{server.domain}, and @code{whitehouse.gov}. He wants to access the @code{bankaccount} database only from @code{localhost}, the @code{expenses} database only from @code{whitehouse.gov}, and the @code{customer} database from all three hosts. He wants to use the password @code{stupid} from all three hosts. To set up this user's privileges using @code{GRANT} statements, run these commands: @example shell> mysql --user=root mysql mysql> GRANT SELECT,INSERT,UPDATE,DELETE,CREATE,DROP ON bankaccount.* TO custom@@localhost IDENTIFIED BY 'stupid'; mysql> GRANT SELECT,INSERT,UPDATE,DELETE,CREATE,DROP ON expenses.* TO custom@@whitehouse.gov IDENTIFIED BY 'stupid'; mysql> GRANT SELECT,INSERT,UPDATE,DELETE,CREATE,DROP ON customer.* TO custom@@'%' IDENTIFIED BY 'stupid'; @end example The reason that we do to grant statements for the user 'custom' is that we want the give the user access to MySQL both from the local machine with Unix sockets and from the remote machine 'whitehouse.gov' over TCP/IP. To set up the user's privileges by modifying the grant tables directly, run these commands (note the @code{FLUSH PRIVILEGES} at the end): @example shell> mysql --user=root mysql mysql> INSERT INTO user (Host,User,Password) VALUES('localhost','custom',PASSWORD('stupid')); mysql> INSERT INTO user (Host,User,Password) VALUES('server.domain','custom',PASSWORD('stupid')); mysql> INSERT INTO user (Host,User,Password) VALUES('whitehouse.gov','custom',PASSWORD('stupid')); mysql> INSERT INTO db (Host,Db,User,Select_priv,Insert_priv,Update_priv,Delete_priv, Create_priv,Drop_priv) VALUES ('localhost','bankaccount','custom','Y','Y','Y','Y','Y','Y'); mysql> INSERT INTO db (Host,Db,User,Select_priv,Insert_priv,Update_priv,Delete_priv, Create_priv,Drop_priv) VALUES ('whitehouse.gov','expenses','custom','Y','Y','Y','Y','Y','Y'); mysql> INSERT INTO db (Host,Db,User,Select_priv,Insert_priv,Update_priv,Delete_priv, Create_priv,Drop_priv) VALUES('%','customer','custom','Y','Y','Y','Y','Y','Y'); mysql> FLUSH PRIVILEGES; @end example The first three @code{INSERT} statements add @code{user} table entries that allow user @code{custom} to connect from the various hosts with the given password, but grant no permissions to him (all privileges are set to the default value of @code{'N'}). The next three @code{INSERT} statements add @code{db} table entries that grant privileges to @code{custom} for the @code{bankaccount}, @code{expenses}, and @code{customer} databases, but only when accessed from the proper hosts. As usual, when the grant tables are modified directly, the server must be told to reload them (with @code{FLUSH PRIVILEGES}) so that the privilege changes take effect. If you want to give a specific user access from any machine in a given domain, you can issue a @code{GRANT} statement like the following: @example mysql> GRANT ... ON *.* TO myusername@@"%.mydomainname.com" IDENTIFIED BY 'mypassword'; @end example To do the same thing by modifying the grant tables directly, do this: @example mysql> INSERT INTO user VALUES ('%.mydomainname.com', 'myusername', PASSWORD('mypassword'),...); mysql> FLUSH PRIVILEGES; @end example You can also use @code{xmysqladmin}, @code{mysql_webadmin}, and even @code{xmysql} to insert, change, and update values in the grant tables. You can find these utilities in the @uref{http://www.mysql.com/Downloads/Contrib/,Contrib directory of the MySQL Website}. @node Passwords, Password security, Adding users, User Account Management @subsection Setting Up Passwords @findex PASSWORD() @findex SET PASSWORD statement @cindex passwords, setting @cindex setting, passwords In most cases you should use @code{GRANT} to set up your users/passwords, so the following only applies for advanced users. @xref{GRANT, , @code{GRANT}}. The examples in the preceding sections illustrate an important principle: when you store a non-empty password using @code{INSERT} or @code{UPDATE} statements, you must use the @code{PASSWORD()} function to encrypt it. This is because the @code{user} table stores passwords in encrypted form, not as plaintext. If you forget that fact, you are likely to attempt to set passwords like this: @example shell> mysql -u root mysql mysql> INSERT INTO user (Host,User,Password) VALUES('%','jeffrey','biscuit'); mysql> FLUSH PRIVILEGES; @end example The result is that the plaintext value @code{'biscuit'} is stored as the password in the @code{user} table. When the user @code{jeffrey} attempts to connect to the server using this password, the @code{mysql} client encrypts it with @code{PASSWORD()}, generates an authentification vector based on @strong{encrypted} password and a random number, obtained from server, and sends the result to the server. The server uses the @code{password} value in the @code{user} table (that is @strong{not encrypted} value @code{'biscuit'}) to perform the same calculations, and compares results. The comparison fails and the server rejects the connection: @example shell> mysql -u jeffrey -pbiscuit test Access denied @end example Passwords must be encrypted when they are inserted in the @code{user} table, so the @code{INSERT} statement should have been specified like this instead: @example mysql> INSERT INTO user (Host,User,Password) VALUES('%','jeffrey',PASSWORD('biscuit')); @end example You must also use the @code{PASSWORD()} function when you use @code{SET PASSWORD} statements: @example mysql> SET PASSWORD FOR jeffrey@@"%" = PASSWORD('biscuit'); @end example If you set passwords using the @code{GRANT ... IDENTIFIED BY} statement or the @code{mysqladmin password} command, the @code{PASSWORD()} function is unnecessary. They both take care of encrypting the password for you, so you would specify a password of @code{'biscuit'} like this: @example mysql> GRANT USAGE ON *.* TO jeffrey@@"%" IDENTIFIED BY 'biscuit'; @end example or @example shell> mysqladmin -u jeffrey password biscuit @end example @strong{NOTE:} @code{PASSWORD()} does not perform password encryption in the same way that Unix passwords are encrypted. You should not assume that if your Unix password and your MySQL password are the same, that @code{PASSWORD()} will result in the same encrypted value as is stored in the Unix password file. @xref{User names}. @node Password security, , Passwords, User Account Management @subsection Keeping Your Password Secure It is inadvisable to specify your password in a way that exposes it to discovery by other users. The methods you can use to specify your password when you run client programs are listed below, along with an assessment of the risks of each method: @itemize @bullet @item Never give a normal user access to the @code{mysql.user} table. Knowing the encrypted password for a user makes it possible to login as this user. The passwords are only scrambled so that one shouldn't be able to see the real password you used (if you happen to use a similar password with your other applications). @item Use a @code{-pyour_pass} or @code{--password=your_pass} option on the command line. This is convenient but insecure, because your password becomes visible to system status programs (such as @code{ps}) that may be invoked by other users to display command lines. (MySQL clients typically overwrite the command-line argument with zeroes during their initialization sequence, but there is still a brief interval during which the value is visible.) @item Use a @code{-p} or @code{--password} option (with no @code{your_pass} value specified). In this case, the client program solicits the password from the terminal: @findex -p option @findex -password option @example shell> mysql -u user_name -p Enter password: ******** @end example The @samp{*} characters represent your password. It is more secure to enter your password this way than to specify it on the command line because it is not visible to other users. However, this method of entering a password is suitable only for programs that you run interactively. If you want to invoke a client from a script that runs non-interactively, there is no opportunity to enter the password from the terminal. On some systems, you may even find that the first line of your script is read and interpreted (incorrectly) as your password! @item @tindex .my.cnf file Store your password in a configuration file. For example, you can list your password in the @code{[client]} section of the @file{.my.cnf} file in your home directory: @example [client] password=your_pass @end example If you store your password in @file{.my.cnf}, the file should not be group or world readable or writable. Make sure the file's access mode is @code{400} or @code{600}. @xref{Option files}. @item You can store your password in the @code{MYSQL_PWD} environment variable, but this method must be considered extremely insecure and should not be used. Some versions of @code{ps} include an option to display the environment of running processes; your password will be in plain sight for all to see if you set @code{MYSQL_PWD}. Even on systems without such a version of @code{ps}, it is unwise to assume there is no other method to observe process environments. @xref{Environment variables}. @end itemize All in all, the safest methods are to have the client program prompt for the password or to specify the password in a properly protected @file{.my.cnf} file. @node Disaster Prevention, Database Administration, User Account Management, MySQL Database Administration @section Disaster Prevention and Recovery @menu * Backup:: Database Backups * BACKUP TABLE:: @code{BACKUP TABLE} Syntax * RESTORE TABLE:: @code{RESTORE TABLE} Syntax * CHECK TABLE:: @code{CHECK TABLE} Syntax * REPAIR TABLE:: @code{REPAIR TABLE} Syntax * Table maintenance:: Using @code{myisamchk} for Table Maintenance and Crash Recovery * Maintenance regimen:: Setting Up a Table Maintenance Regimen * Table-info:: Getting Information About a Table @end menu @node Backup, BACKUP TABLE, Disaster Prevention, Disaster Prevention @subsection Database Backups @cindex databases, backups @cindex backups Because MySQL tables are stored as files, it is easy to do a backup. To get a consistent backup, do a @code{LOCK TABLES} on the relevant tables followed by @code{FLUSH TABLES} for the tables. @xref{LOCK TABLES, , @code{LOCK TABLES}}. @xref{FLUSH, , @code{FLUSH}}. You only need a read lock; this allows other threads to continue to query the tables while you are making a copy of the files in the database directory. The @code{FLUSH TABLE} is needed to ensure that the all active index pages is written to disk before you start the backup. If you want to make a SQL level backup of a table, you can use @code{SELECT INTO OUTFILE} or @code{BACKUP TABLE}. @xref{SELECT}. @xref{BACKUP TABLE}. Another way to back up a database is to use the @code{mysqldump} program or the @code{mysqlhotcopy script}. @xref{mysqldump, , @code{mysqldump}}. @xref{mysqlhotcopy, , @code{mysqlhotcopy}}. @enumerate @item Do a full backup of your databases: @example shell> mysqldump --tab=/path/to/some/dir --opt --full or shell> mysqlhotcopy database /path/to/some/dir @end example You can also simply copy all table files (@file{*.frm}, @file{*.MYD}, and @file{*.MYI} files) as long as the server isn't updating anything. The script @code{mysqlhotcopy} does use this method. @item @cindex log files, names Stop @code{mysqld} if it's running, then start it with the @code{--log-update[=file_name]} option. @xref{Update log}. The update log file(s) provide you with the information you need to replicate changes to the database that are made subsequent to the point at which you executed @code{mysqldump}. @end enumerate If you have to restore something, try to recover your tables using @code{REPAIR TABLE} or @code{myisamchk -r} first. That should work in 99.9% of all cases. If @code{myisamchk} fails, try the following procedure: (This will only work if you have started MySQL with @code{--log-update}. @xref{Update log}.): @enumerate @item Restore the original @code{mysqldump} backup. @item Execute the following command to re-run the updates in the binary log: @example shell> mysqlbinlog hostname-bin.[0-9]* | mysql @end example If you are using the update log you can use: @example shell> ls -1 -t -r hostname.[0-9]* | xargs cat | mysql @end example @end enumerate @code{ls} is used to get all the update log files in the right order. You can also do selective backups with @code{SELECT * INTO OUTFILE 'file_name' FROM tbl_name} and restore with @code{LOAD DATA INFILE 'file_name' REPLACE ...} To avoid duplicate records, you need a @code{PRIMARY KEY} or a @code{UNIQUE} key in the table. The @code{REPLACE} keyword causes old records to be replaced with new ones when a new record duplicates an old record on a unique key value. If you get performance problems in making backups on your system, you can solve this by setting up replication and do the backups on the slave instead of on the master. @xref{Replication Intro}. If you are using a Veritas file system, you can do: @enumerate @item Execute in a client (perl ?) @code{FLUSH TABLES WITH READ LOCK} @item Fork a shell or execute in another client @code{mount vxfs snapshot}. @item Execute in the first client @code{UNLOCK TABLES} @item Copy files from snapshot @item Unmount snapshot @end enumerate @node BACKUP TABLE, RESTORE TABLE, Backup, Disaster Prevention @subsection @code{BACKUP TABLE} Syntax @findex BACKUP TABLE @cindex backups, database @example BACKUP TABLE tbl_name[,tbl_name...] TO '/path/to/backup/directory' @end example Make a copy of all the table files to the backup directory that are the minimum needed to restore it. Currenlty only works for @code{MyISAM} tables. For @code{MyISAM} table, copies @code{.frm} (definition) and @code{.MYD} (data) files. The index file can be rebuilt from those two. Before using this command, please see @xref{Backup}. During the backup, read lock will be held for each table, one at time, as they are being backed up. If you want to backup several tables as a snapshot, you must first issue @code{LOCK TABLES} obtaining a read lock for each table in the group. The command returns a table with the following columns: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Value} @item Table @tab Table name @item Op @tab Always ``backup'' @item Msg_type @tab One of @code{status}, @code{error}, @code{info} or @code{warning}. @item Msg_text @tab The message. @end multitable Note that @code{BACKUP TABLE} is only available in MySQL version 3.23.25 and later. @node RESTORE TABLE, CHECK TABLE, BACKUP TABLE, Disaster Prevention @subsection @code{RESTORE TABLE} Syntax @findex RESTORE TABLE @example RESTORE TABLE tbl_name[,tbl_name...] FROM '/path/to/backup/directory' @end example Restores the table(s) from the backup that was made with @code{BACKUP TABLE}. Existing tables will not be overwritten - if you try to restore over an existing table, you will get an error. Restore will take longer than BACKUP due to the need to rebuilt the index. The more keys you have, the longer it is going to take. Just as @code{BACKUP TABLE}, currently only works of @code{MyISAM} tables. The command returns a table with the following columns: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Value} @item Table @tab Table name @item Op @tab Always ``restore'' @item Msg_type @tab One of @code{status}, @code{error}, @code{info} or @code{warning}. @item Msg_text @tab The message. @end multitable @node CHECK TABLE, REPAIR TABLE, RESTORE TABLE, Disaster Prevention @subsection @code{CHECK TABLE} Syntax @findex CHECK TABLE @example CHECK TABLE tbl_name[,tbl_name...] [option [option...]] option = QUICK | FAST | MEDIUM | EXTENDED | CHANGED @end example @code{CHECK TABLE} only works on @code{MyISAM} tables. On @code{MyISAM} tables it's the same thing as running @code{myisamchk -m table_name} on the table. If you don't specify any option @code{MEDIUM} is used. Checks the table(s) for errors. For @code{MyISAM} tables the key statistics is updated. The command returns a table with the following columns: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Value} @item Table @tab Table name. @item Op @tab Always ``check''. @item Msg_type @tab One of @code{status}, @code{error}, @code{info}, or @code{warning}. @item Msg_text @tab The message. @end multitable Note that you can get many rows of information for each checked table. The last row will be of @code{Msg_type status} and should normally be @code{OK}. If you don't get @code{OK}, or @code{Not checked} you should normally run a repair of the table. @xref{Table maintenance}. @code{Not checked} means that the table the given @code{TYPE} told MySQL that there wasn't any need to check the table. The different check types stand for the following: @multitable @columnfractions .20 .80 @item @strong{Type} @tab @strong{Meaning} @item @code{QUICK} @tab Don't scan the rows to check for wrong links. @item @code{FAST} @tab Only check tables which haven't been closed properly. @item @code{CHANGED} @tab Only check tables which have been changed since last check or haven't been closed properly. @item @code{MEDIUM} @tab Scan rows to verify that deleted links are ok. This also calculates a key checksum for the rows and verifies this with a calcualted checksum for the keys. @item @code{EXTENDED} @tab Do a full key lookup for all keys for each row. This ensures that the table is 100 % consistent, but will take a long time! @end multitable For dynamic sized @code{MyISAM} tables a started check will always do a @code{MEDIUM} check. For static size rows we skip the row scan for @code{QUICK} and @code{FAST} as the rows are very seldom corrupted. You can combine check options as in: @example CHECK TABLE test_table FAST QUICK; @end example Which only would do a quick check on the table if it wasn't closed properly. @strong{NOTE:} that in some case @code{CHECK TABLE} will change the table! This happens if the table is marked as 'corrupted' or 'not closed properly' but @code{CHECK TABLE} didn't find any problems in the table. In this case @code{CHECK TABLE} will mark the table as ok. If a table is corrupted, then it's most likely that the problem is in the indexes and not in the data part. All of the above check types checks the indexes throughly and should thus find most errors. If you just want to check a table that you assume is ok, you should use no check options or the @code{QUICK} option. The later should be used when you are in a hurry and can take the very small risk that @code{QUICK} didn't find an error in the data file (In most cases MySQL should find, under normal usage, any error in the data file. If this happens then the table will be marked as 'corrupted', in which case the table can't be used until it's repaired). @code{FAST} and @code{CHANGED} are mostly intended to be used from a script (for example to be executed from cron) if you want to check your table from time to time. In most cases you @code{FAST} is to be prefered over @code{CHANGED}. (The only case when it isn't is when you suspect a bug you have found a bug in the @code{MyISAM} code.). @code{EXTENDED} is only to be used after you have run a normal check but still get strange errors from a table when MySQL tries to update a row or find a row by key (this is VERY unlikely to happen if a normal check has succeeded!). Some things reported by check table, can't be corrected automatically: @itemize @bullet @item @code{Found row where the auto_increment column has the value 0}. This means that you have in the table a row where the @code{auto_increment} index column contains the value 0. (It's possible to create a row where the auto_increment column is 0 by explicitely setting the column to 0 with an @code{UPDATE} statement) This isn't an error in itself, but could cause trouble if you decide to dump the table and restore it or do an @code{ALTER TABLE} on the table. In this case the auto_increment column will change value, according to the rules of auto_increment columns, which could cause problems like a duplicate key error. To get rid of the warning, just execute an @code{UPDATE} statement to set the column to some other value than 0. @end itemize @node REPAIR TABLE, Table maintenance, CHECK TABLE, Disaster Prevention @subsection @code{REPAIR TABLE} Syntax @findex REPAIR TABLE @example REPAIR TABLE tbl_name[,tbl_name...] [QUICK] [EXTENDED] @end example @code{REPAIR TABLE} only works on @code{MyISAM} tables and is the same as running @code{myisamchk -r table_name} on the table. Normally you should never have to run this command, but if disaster strikes you are very likely to get back all your data from a MyISAM table with @code{REPAIR TABLE}. If your tables get corrupted a lot you should try to find the reason for this! @xref{Crashing}. @xref{MyISAM table problems}. @code{REPAIR TABLE} repairs a possible corrupted table. The command returns a table with the following columns: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Value} @item Table @tab Table name @item Op @tab Always ``repair'' @item Msg_type @tab One of @code{status}, @code{error}, @code{info} or @code{warning}. @item Msg_text @tab The message. @end multitable Note that you can get many rows of information for each repaired table. The last one row will be of @code{Msg_type status} and should normally be @code{OK}. If you don't get @code{OK}, you should try repairing the table with @code{myisamchk -o}, as @code{REPAIR TABLE} does not yet implement all the options of @code{myisamchk}. In the near future, we will make it more flexible. If @code{QUICK} is given then MySQL will try to do a @code{REPAIR} of only the index tree. If you use @code{EXTENDED} then MySQL will create the index row by row instead of creating one index at a time with sorting; This may be better than sorting on fixed-length keys if you have long @code{char()} keys that compress very good. @node Table maintenance, Maintenance regimen, REPAIR TABLE, Disaster Prevention @subsection Using @code{myisamchk} for Table Maintenance and Crash Recovery Starting with MySQL Version 3.23.13, you can check MyISAM tables with the @code{CHECK TABLE} command. @xref{CHECK TABLE}. You can repair tables with the @code{REPAIR TABLE} command. @xref{REPAIR TABLE}. To check/repair MyISAM tables (@code{.MYI} and @code{.MYD}) you should use the @code{myisamchk} utility. To check/repair ISAM tables (@code{.ISM} and @code{.ISD}) you should use the @code{isamchk} utility. @xref{Table types}. In the following text we will talk about @code{myisamchk}, but everything also applies to the old @code{isamchk}. You can use the @code{myisamchk} utility to get information about your database tables, check and repair them, or optimize them. The following sections describe how to invoke @code{myisamchk} (including a description of its options), how to set up a table maintenance schedule, and how to use @code{myisamchk} to perform its various functions. You can, in most cases, also use the command @code{OPTIMIZE TABLES} to optimize and repair tables, but this is not as fast or reliable (in case of real fatal errors) as @code{myisamchk}. On the other hand, @code{OPTIMIZE TABLE} is easier to use and you don't have to worry about flushing tables. @xref{OPTIMIZE TABLE, , @code{OPTIMIZE TABLE}}. Even that the repair in @code{myisamchk} is quite secure, it's always a good idea to make a backup BEFORE doing a repair (or anything that could make a lot of changes to a table) @menu * myisamchk syntax:: @code{myisamchk} Invocation Syntax * myisamchk general options:: General Options for @code{myisamchk} * myisamchk check options:: Check Options for @code{myisamchk} * myisamchk repair options:: Repair Options for myisamchk * myisamchk other options:: Other Options for @code{myisamchk} * myisamchk memory:: @code{myisamchk} Memory Usage * Crash recovery:: Using @code{myisamchk} for Crash Recovery * Check:: How to Check Tables for Errors * Repair:: How to Repair Tables * Optimization:: Table Optimization @end menu @node myisamchk syntax, myisamchk general options, Table maintenance, Table maintenance @subsubsection @code{myisamchk} Invocation Syntax @code{myisamchk} is invoked like this: @example shell> myisamchk [options] tbl_name @end example The @code{options} specify what you want @code{myisamchk} to do. They are described below. (You can also get a list of options by invoking @code{myisamchk --help}.) With no options, @code{myisamchk} simply checks your table. To get more information or to tell @code{myisamchk} to take corrective action, specify options as described below and in the following sections. @code{tbl_name} is the database table you want to check/repair. If you run @code{myisamchk} somewhere other than in the database directory, you must specify the path to the file, because @code{myisamchk} has no idea where your database is located. Actually, @code{myisamchk} doesn't care whether or not the files you are working on are located in a database directory; you can copy the files that correspond to a database table into another location and perform recovery operations on them there. You can name several tables on the @code{myisamchk} command line if you wish. You can also specify a name as an index file name (with the @file{.MYI} suffix), which allows you to specify all tables in a directory by using the pattern @file{*.MYI}. For example, if you are in a database directory, you can check all the tables in the directory like this: @example shell> myisamchk *.MYI @end example If you are not in the database directory, you can check all the tables there by specifying the path to the directory: @example shell> myisamchk /path/to/database_dir/*.MYI @end example You can even check all tables in all databases by specifying a wild card with the path to the MySQL data directory: @example shell> myisamchk /path/to/datadir/*/*.MYI @end example The recommended way to quickly check all tables is: @example myisamchk --silent --fast /path/to/datadir/*/*.MYI isamchk --silent /path/to/datadir/*/*.ISM @end example If you want to check all tables and repair all tables that are corrupted, you can use the following line: @example myisamchk --silent --force --fast --update-state -O key_buffer=64M -O sort_buffer=64M -O read_buffer=1M -O write_buffer=1M /path/to/datadir/*/*.MYI isamchk --silent --force -O key_buffer=64M -O sort_buffer=64M -O read_buffer=1M -O write_buffer=1M /path/to/datadir/*/*.ISM @end example The above assumes that you have more than 64 M free. Note that if you get an error like: @example myisamchk: warning: 1 clients is using or hasn't closed the table properly @end example This means that you are trying to check a table that has been updated by the another program (like the @code{mysqld} server) that hasn't yet closed the file or that has died without closing the file properly. If you @code{mysqld} is running, you must force a sync/close of all tables with @code{FLUSH TABLES} and ensure that no one is using the tables while you are running @code{myisamchk}. In MySQL Version 3.23 the easiest way to avoid this problem is to use @code{CHECK TABLE} instead of @code{myisamchk} to check tables. @menu * myisamchk general options:: General Options for @code{myisamchk} * myisamchk check options:: Check Options for @code{myisamchk} * myisamchk repair options:: Repair Options for myisamchk * myisamchk other options:: Other Options for @code{myisamchk} @end menu @node myisamchk general options, myisamchk check options, myisamchk syntax, Table maintenance @subsubsection General Options for @code{myisamchk} @cindex options, @code{myisamchk} @cindex @code{myisamchk}, options @code{myisamchk} supports the following options. @table @code @item -# or --debug=debug_options Output debug log. The @code{debug_options} string often is @code{'d:t:o,filename'}. @item -? or --help Display a help message and exit. @item -O var=option, --set-variable var=option Set the value of a variable. The possible variables and their default values for myisamchk can be examined with @code{myisamchk --help}: @multitable @columnfractions .3 .7 @item key_buffer_size @tab 523264 @item read_buffer_size @tab 262136 @item write_buffer_size @tab 262136 @item sort_buffer_size @tab 2097144 @item sort_key_blocks @tab 16 @item decode_bits @tab 9 @end multitable @code{sort_buffer_size} is used when the keys are repaired by sorting keys, which is the normal case when you use @code{--recover}. @code{key_buffer_size} is used when you are checking the table with @code{--extended-check} or when the keys are repaired by inserting key row by row in to the table (like when doing normal inserts). Repairing through the key buffer is used in the following cases: @itemize @bullet @item If you use @code{--safe-recover}. @item If you are using a @code{FULLTEXT} index. @item If the temporary files needed to sort the keys would be more than twice as big as when creating the key file directly. This is often the case when you have big @code{CHAR}, @code{VARCHAR} or @code{TEXT} keys as the sort needs to store the whole keys during sorting. If you have lots of temporary space and you can force @code{myisamchk} to repair by sorting you can use the @code{--sort-recover} option. @end itemize Reparing through the key buffer takes much less disk space than using sorting, but is also much slower. If you want a faster repair, set the above variables to about 1/4 of your available memory. You can set both variables to big values, as only one of the above buffers will be used at a time. @item -s or --silent Silent mode. Write output only when errors occur. You can use @code{-s} twice (@code{-ss}) to make @code{myisamchk} very silent. @item -v or --verbose Verbose mode. Print more information. This can be used with @code{-d} and @code{-e}. Use @code{-v} multiple times (@code{-vv}, @code{-vvv}) for more verbosity! @item -V or --version Print the @code{myisamchk} version and exit. @item -w or, --wait Instead of giving an error if the table is locked, wait until the table is unlocked before continuing. Note that if you are running @code{mysqld} on the table with @code{--skip-locking}, the table can only be locked by another @code{myisamchk} command. @end table @node myisamchk check options, myisamchk repair options, myisamchk general options, Table maintenance @subsubsection Check Options for @code{myisamchk} @cindex check options, myisamchk @cindex tables, checking @table @code @item -c or --check Check table for errors. This is the default operation if you are not giving @code{myisamchk} any options that override this. @item -e or --extend-check Check the table VERY thoroughly (which is quite slow if you have many indexes). This option should only be used in extreme cases. Normally, @code{myisamchk} or @code{myisamchk --medium-check} should, in most cases, be able to find out if there are any errors in the table. If you are using @code{--extended-check} and have much memory, you should increase the value of @code{key_buffer_size} a lot! @item -F or --fast Check only tables that haven't been closed properly. @item -C or --check-only-changed Check only tables that have changed since the last check. @item -f or --force Restart @code{myisamchk} with @code{-r} (repair) on the table, if @code{myisamchk} finds any errors in the table. @item -i or --information Print informational statistics about the table that is checked. @item -m or --medium-check Faster than extended-check, but only finds 99.99% of all errors. Should, however, be good enough for most cases. @item -U or --update-state Store in the @file{.MYI} file when the table was checked and if the table crashed. This should be used to get full benefit of the @code{--check-only-changed} option, but you shouldn't use this option if the @code{mysqld} server is using the table and you are running @code{mysqld} with @code{--skip-locking}. @item -T or --read-only Don't mark table as checked. This is useful if you use @code{myisamchk} to check a table that is in use by some other application that doesn't use locking (like @code{mysqld --skip-locking}). @end table @node myisamchk repair options, myisamchk other options, myisamchk check options, Table maintenance @subsubsection Repair Options for myisamchk @cindex repair options, myisamchk @cindex files, repairing The following options are used if you start @code{myisamchk} with @code{-r} or @code{-o}: @table @code @item -D # or --data-file-length=# Max length of data file (when re-creating data file when it's 'full'). @item -e or --extend-check Try to recover every possible row from the data file. Normally this will also find a lot of garbage rows. Don't use this option if you are not totally desperate. @item -f or --force Overwrite old temporary files (@code{table_name.TMD}) instead of aborting. @item -k # or keys-used=# If you are using ISAM, tells the ISAM table handler to update only the first @code{#} indexes. If you are using @code{MyISAM}, tells which keys to use, where each binary bit stands for one key (first key is bit 0). This can be used to get faster inserts! Deactivated indexes can be reactivated by using @code{myisamchk -r}. keys. @item -l or --no-symlinks Do not follow symbolic links. Normally @code{myisamchk} repairs the table a symlink points at. This option doesn't exist in MySQL 4.0, as MySQL 4.0 will not remove symlinks during repair. @item -r or --recover Can fix almost anything except unique keys that aren't unique (which is an extremely unlikely error with ISAM/MyISAM tables). If you want to recover a table, this is the option to try first. Only if myisamchk reports that the table can't be recovered by @code{-r}, you should then try @code{-o}. (Note that in the unlikely case that @code{-r} fails, the data file is still intact.) If you have lots of memory, you should increase the size of @code{sort_buffer_size}! @item -o or --safe-recover Uses an old recovery method (reads through all rows in order and updates all index trees based on the found rows); this is a magnitude slower than @code{-r}, but can handle a couple of very unlikely cases that @code{-r} cannot handle. This recovery method also uses much less disk space than @code{-r}. Normally one should always first repair with @code{-r}, and only if this fails use @code{-o}. If you have lots of memory, you should increase the size of @code{key_buffer_size}! @item -n or --sort-recover Force @code{myisamchk} to use sorting to resolve the keys even if the temporary files should be very big. This will not have any effect if you have fulltext keys in the table. @item --character-sets-dir=... Directory where character sets are stored. @item --set-character-set=name Change the character set used by the index @item .t or --tmpdir=path Path for storing temporary files. If this is not set, @code{myisamchk} will use the environment variable @code{TMPDIR} for this. @item -q or --quick Faster repair by not modifying the data file. One can give a second @code{-q} to force @code{myisamchk} to modify the original datafile in case of duplicate keys @item -u or --unpack Unpack file packed with myisampack. @end table @node myisamchk other options, myisamchk memory, myisamchk repair options, Table maintenance @subsubsection Other Options for @code{myisamchk} Other actions that @code{myisamchk} can do, besides repair and check tables: @table @code @item -a or --analyze Analyze the distribution of keys. This improves join performance by enabling the join optimizer to better choose in which order it should join the tables and which keys it should use: @code{myisamchk --describe --verbose table_name'} or using @code{SHOW KEYS} in MySQL. @item -d or --description Prints some information about table. @item -A or --set-auto-increment[=value] Force auto_increment to start at this or higher value. If no value is given, then sets the next auto_increment value to the highest used value for the auto key + 1. @item -S or --sort-index Sort the index tree blocks in high-low order. This will optimize seeks and will make table scanning by key faster. @item -R or --sort-records=# Sorts records according to an index. This makes your data much more localized and may speed up ranged @code{SELECT} and @code{ORDER BY} operations on this index. (It may be VERY slow to do a sort the first time!) To find out a table's index numbers, use @code{SHOW INDEX}, which shows a table's indexes in the same order that @code{myisamchk} sees them. Indexes are numbered beginning with 1. @end table @node myisamchk memory, Crash recovery, myisamchk other options, Table maintenance @subsubsection @code{myisamchk} Memory Usage @cindex memory usage, myisamchk Memory allocation is important when you run @code{myisamchk}. @code{myisamchk} uses no more memory than you specify with the @code{-O} options. If you are going to use @code{myisamchk} on very large files, you should first decide how much memory you want it to use. The default is to use only about 3M to fix things. By using larger values, you can get @code{myisamchk} to operate faster. For example, if you have more than 32M RAM, you could use options such as these (in addition to any other options you might specify): @example shell> myisamchk -O sort=16M -O key=16M -O read=1M -O write=1M ... @end example Using @code{-O sort=16M} should probably be enough for most cases. Be aware that @code{myisamchk} uses temporary files in @code{TMPDIR}. If @code{TMPDIR} points to a memory file system, you may easily get out of memory errors. If this happens, set @code{TMPDIR} to point at some directory with more space and restart @code{myisamchk}. When repairing, @code{myisamchk} will also need a lot of disk space: @itemize @bullet @item Double the size of the record file (the original one and a copy). This space is not needed if one does a repair with @code{--quick}, as in this case only the index file will be re-created. This space is needed on the same disk as the original record file! @item Space for the new index file that replaces the old one. The old index file is truncated at start, so one usually ignore this space. This space is needed on the same disk as the original index file! @item When using @code{--recover} or @code{--sort-recover} (but not when using @code{--safe-recover}, you will need space for a sort buffer for: @code{(largest_key + row_pointer_length)*number_of_rows * 2}. You can check the length of the keys and the row_pointer_length with @code{myisamchk -dv table}. This space is allocated on the temporary disk (specified by @code{TMPDIR} or @code{--tmpdir=#}). @end itemize If you have a problem with disk space during repair, you can try to use @code{--safe-recover} instead of @code{--recover}. @node Crash recovery, Check, myisamchk memory, Table maintenance @subsubsection Using @code{myisamchk} for Crash Recovery @cindex crash, recovery @cindex recovery, from crash If you run @code{mysqld} with @code{--skip-locking} (which is the default on some systems, like Linux), you can't reliably use @code{myisamchk} to check a table when @code{mysqld} is using the same table. If you can be sure that no one is accessing the tables through @code{mysqld} while you run @code{myisamchk}, you only have to do @code{mysqladmin flush-tables} before you start checking the tables. If you can't guarantee the above, then you must take down @code{mysqld} while you check the tables. If you run @code{myisamchk} while @code{mysqld} is updating the tables, you may get a warning that a table is corrupt even if it isn't. If you are not using @code{--skip-locking}, you can use @code{myisamchk} to check tables at any time. While you do this, all clients that try to update the table will wait until @code{myisamchk} is ready before continuing. If you use @code{myisamchk} to repair or optimize tables, you @strong{MUST} always ensure that the @code{mysqld} server is not using the table (this also applies if you are using @code{--skip-locking}). If you don't take down @code{mysqld} you should at least do a @code{mysqladmin flush-tables} before you run @code{myisamchk}. This chapter describes how to check for and deal with data corruption in MySQL databases. If your tables get corrupted a lot you should try to find the reason for this! @xref{Crashing}. The @code{MyISAM} table section contains reason for why a table could be corrupted. @xref{MyISAM table problems}. When performing crash recovery, it is important to understand that each table @code{tbl_name} in a database corresponds to three files in the database directory: @multitable @columnfractions .2 .8 @item @strong{File} @tab @strong{Purpose} @item @file{tbl_name.frm} @tab Table definition (form) file @item @file{tbl_name.MYD} @tab Data file @item @file{tbl_name.MYI} @tab Index file @end multitable Each of these three file types is subject to corruption in various ways, but problems occur most often in data files and index files. @code{myisamchk} works by creating a copy of the @file{.MYD} (data) file row by row. It ends the repair stage by removing the old @file{.MYD} file and renaming the new file to the original file name. If you use @code{--quick}, @code{myisamchk} does not create a temporary @file{.MYD} file, but instead assumes that the @file{.MYD} file is correct and only generates a new index file without touching the @file{.MYD} file. This is safe, because @code{myisamchk} automatically detects if the @file{.MYD} file is corrupt and aborts the repair in this case. You can also give two @code{--quick} options to @code{myisamchk}. In this case, @code{myisamchk} does not abort on some errors (like duplicate key) but instead tries to resolve them by modifying the @file{.MYD} file. Normally the use of two @code{--quick} options is useful only if you have too little free disk space to perform a normal repair. In this case you should at least make a backup before running @code{myisamchk}. @node Check, Repair, Crash recovery, Table maintenance @subsubsection How to Check Tables for Errors @cindex checking, tables for errors @cindex tables, error checking @cindex errors, checking tables for To check a MyISAM table, use the following commands: @table @code @item myisamchk tbl_name This finds 99.99% of all errors. What it can't find is corruption that involves @strong{ONLY} the data file (which is very unusual). If you want to check a table, you should normally run @code{myisamchk} without options or with either the @code{-s} or @code{--silent} option. @item myisamchk -m tbl_name This finds 99.999% of all errors. It checks first all index entries for errors and then it reads through all rows. It calculates a checksum for all keys in the rows and verifies that they checksum matches the checksum for the keys in the index tree. @item myisamchk -e tbl_name This does a complete and thorough check of all data (@code{-e} means ``extended check''). It does a check-read of every key for each row to verify that they indeed point to the correct row. This may take a LONG time on a big table with many keys. @code{myisamchk} will normally stop after the first error it finds. If you want to obtain more information, you can add the @code{--verbose} (@code{-v}) option. This causes @code{myisamchk} to keep going, up through a maximum of 20 errors. In normal usage, a simple @code{myisamchk} (with no arguments other than the table name) is sufficient. @item myisamchk -e -i tbl_name Like the previous command, but the @code{-i} option tells @code{myisamchk} to print some informational statistics, too. @end table @node Repair, Optimization, Check, Table maintenance @subsubsection How to Repair Tables @cindex tables, repairing @cindex repairing, tables In the following section we only talk about using @code{myisamchk} on @code{MyISAM} tables (extensions @code{.MYI} and @code{.MYD}). If you are using @code{ISAM} tables (extensions @code{.ISM} and @code{.ISD}), you should use @code{isamchk} instead. Starting with MySQL Version 3.23.14, you can repair MyISAM tables with the @code{REPAIR TABLE} command. @xref{REPAIR TABLE}. The symptoms of a corrupted table include queries that abort unexpectedly and observable errors such as these: @itemize @bullet @item @file{tbl_name.frm} is locked against change @item Can't find file @file{tbl_name.MYI} (Errcode: ###) @item Unexpected end of file @item Record file is crashed @item Got error ### from table handler To get more information about the error you can run @code{perror ###}. Here is the most common errors that indicates a problem with the table: @example shell> perror 126 127 132 134 135 136 141 144 145 126 = Index file is crashed / Wrong file format 127 = Record-file is crashed 132 = Old database file 134 = Record was already deleted (or record file crashed) 135 = No more room in record file 136 = No more room in index file 141 = Duplicate unique key or constraint on write or update 144 = Table is crashed and last repair failed 145 = Table was marked as crashed and should be repaired @end example Note that error 135, no more room in record file, is not an error that can be fixed by a simple repair. In this case you have to do: @example ALTER TABLE table MAX_ROWS=xxx AVG_ROW_LENGTH=yyy; @end example @end itemize In the other cases, you must repair your tables. @code{myisamchk} can usually detect and fix most things that go wrong. The repair process involves up to four stages, described below. Before you begin, you should @code{cd} to the database directory and check the permissions of the table files. Make sure they are readable by the Unix user that @code{mysqld} runs as (and to you, because you need to access the files you are checking). If it turns out you need to modify files, they must also be writable by you. If you are using MySQL Version 3.23.16 and above, you can (and should) use the @code{CHECK} and @code{REPAIR} commands to check and repair @code{MyISAM} tables. @xref{CHECK TABLE}. @xref{REPAIR TABLE}. The manual section about table maintenance includes the options to @code{isamchk}/@code{myisamchk}. @xref{Table maintenance}. The following section is for the cases where the above command fails or if you want to use the extended features that @code{isamchk}/@code{myisamchk} provides. If you are going to repair a table from the command line, you must first take down the @code{mysqld} server. Note that when you do @code{mysqladmin shutdown} on a remote server, the @code{mysqld} server will still be alive for a while after @code{mysqladmin} returns, until all queries are stopped and all keys have been flushed to disk. @noindent @strong{Stage 1: Checking your tables} Run @code{myisamchk *.MYI} or @code{myisamchk -e *.MYI} if you have more time. Use the @code{-s} (silent) option to suppress unnecessary information. If the @code{mysqld} server is done you should use the --update option to tell @code{myisamchk} to mark the table as 'checked'. You have to repair only those tables for which @code{myisamchk} announces an error. For such tables, proceed to Stage 2. If you get weird errors when checking (such as @code{out of memory} errors), or if @code{myisamchk} crashes, go to Stage 3. @noindent @strong{Stage 2: Easy safe repair} NOTE: If you want repairing to go much faster, you should add: @code{-O sort_buffer=# -O key_buffer=#} (where # is about 1/4 of the available memory) to all @code{isamchk/myisamchk} commands. First, try @code{myisamchk -r -q tbl_name} (@code{-r -q} means ``quick recovery mode''). This will attempt to repair the index file without touching the data file. If the data file contains everything that it should and the delete links point at the correct locations within the data file, this should work, and the table is fixed. Start repairing the next table. Otherwise, use the following procedure: @enumerate @item Make a backup of the data file before continuing. @item Use @code{myisamchk -r tbl_name} (@code{-r} means ``recovery mode''). This will remove incorrect records and deleted records from the data file and reconstruct the index file. @item If the preceding step fails, use @code{myisamchk --safe-recover tbl_name}. Safe recovery mode uses an old recovery method that handles a few cases that regular recovery mode doesn't (but is slower). @end enumerate If you get weird errors when repairing (such as @code{out of memory} errors), or if @code{myisamchk} crashes, go to Stage 3. @noindent @strong{Stage 3: Difficult repair} You should only reach this stage if the first 16K block in the index file is destroyed or contains incorrect information, or if the index file is missing. In this case, it's necessary to create a new index file. Do so as follows: @enumerate @item Move the data file to some safe place. @item Use the table description file to create new (empty) data and index files: @example shell> mysql db_name mysql> SET AUTOCOMMIT=1; mysql> TRUNCATE TABLE table_name; mysql> quit @end example If your SQL version doesn't have @code{TRUNCATE TABLE}, use @code{DELETE FROM table_name} instead. @item Copy the old data file back onto the newly created data file. (Don't just move the old file back onto the new file; you want to retain a copy in case something goes wrong.) @end enumerate Go back to Stage 2. @code{myisamchk -r -q} should work now. (This shouldn't be an endless loop.) @noindent @strong{Stage 4: Very difficult repair} You should reach this stage only if the description file has also crashed. That should never happen, because the description file isn't changed after the table is created: @enumerate @item Restore the description file from a backup and go back to Stage 3. You can also restore the index file and go back to Stage 2. In the latter case, you should start with @code{myisamchk -r}. @item If you don't have a backup but know exactly how the table was created, create a copy of the table in another database. Remove the new data file, then move the description and index files from the other database to your crashed database. This gives you new description and index files, but leaves the data file alone. Go back to Stage 2 and attempt to reconstruct the index file. @end enumerate @node Optimization, , Repair, Table maintenance @subsubsection Table Optimization @cindex tables, optimizing @cindex optimizing, tables To coalesce fragmented records and eliminate wasted space resulting from deleting or updating records, run @code{myisamchk} in recovery mode: @example shell> myisamchk -r tbl_name @end example You can optimize a table in the same way using the SQL @code{OPTIMIZE TABLE} statement. @code{OPTIMIZE TABLE} does a repair of the table, a key analyzes and also sorts the index tree to give faster key lookups. There is also no possibility of unwanted interaction between a utility and the server, because the server does all the work when you use @code{OPTIMIZE TABLE}. @xref{OPTIMIZE TABLE}. @code{myisamchk} also has a number of other options you can use to improve the performance of a table: @table @code @item -S, --sort-index @item -R index_num, --sort-records=index_num @item -a, --analyze @end table For a full description of the option. @xref{myisamchk syntax}. @node Maintenance regimen, Table-info, Table maintenance, Disaster Prevention @subsection Setting Up a Table Maintenance Regimen @cindex maintaining, tables @cindex tables, maintenance regimen Starting with MySQL Version 3.23.13, you can check MyISAM tables with the @code{CHECK TABLE} command. @xref{CHECK TABLE}. You can repair tables with the @code{REPAIR TABLE} command. @xref{REPAIR TABLE}. It is a good idea to perform table checks on a regular basis rather than waiting for problems to occur. For maintenance purposes, you can use @code{myisamchk -s} to check tables. The @code{-s} option (short for @code{--silent}) causes @code{myisamchk} to run in silent mode, printing messages only when errors occur. @tindex .pid (process ID) file It's also a good idea to check tables when the server starts up. For example, whenever the machine has done a reboot in the middle of an update, you usually need to check all the tables that could have been affected. (This is an ``expected crashed table''.) You could add a test to @code{safe_mysqld} that runs @code{myisamchk} to check all tables that have been modified during the last 24 hours if there is an old @file{.pid} (process ID) file left after a reboot. (The @file{.pid} file is created by @code{mysqld} when it starts up and removed when it terminates normally. The presence of a @file{.pid} file at system startup time indicates that @code{mysqld} terminated abnormally.) An even better test would be to check any table whose last-modified time is more recent than that of the @file{.pid} file. You should also check your tables regularly during normal system operation. At MySQL AB, we run a @code{cron} job to check all our important tables once a week, using a line like this in a @file{crontab} file: @example 35 0 * * 0 /path/to/myisamchk --fast --silent /path/to/datadir/*/*.MYI @end example This prints out information about crashed tables so we can examine and repair them when needed. As we haven't had any unexpectedly crashed tables (tables that become corrupted for reasons other than hardware trouble) for a couple of years now (this is really true), once a week is more than enough for us. We recommend that to start with, you execute @code{myisamchk -s} each night on all tables that have been updated during the last 24 hours, until you come to trust MySQL as much as we do. @cindex tables, defragment Normally you don't need to maintain MySQL tables that much. If you are changing tables with dynamic size rows (tables with @code{VARCHAR}, @code{BLOB} or @code{TEXT} columns) or have tables with many deleted rows you may want to from time to time (once a month?) defragment/reclaim space from the tables. You can do this by using @code{OPTIMIZE TABLE} on the tables in question or if you can take the @code{mysqld} server down for a while do: @example isamchk -r --silent --sort-index -O sort_buffer_size=16M */*.ISM myisamchk -r --silent --sort-index -O sort_buffer_size=16M */*.MYI @end example @node Table-info, , Maintenance regimen, Disaster Prevention @subsection Getting Information About a Table @cindex tables, information To get a description of a table or statistics about it, use the commands shown below. We explain some of the information in more detail later: @table @code @item myisamchk -d tbl_name Runs @code{myisamchk} in ``describe mode'' to produce a description of your table. If you start the MySQL server using the @code{--skip-locking} option, @code{myisamchk} may report an error for a table that is updated while it runs. However, because @code{myisamchk} doesn't change the table in describe mode, there isn't any risk of destroying data. @item myisamchk -d -v tbl_name To produce more information about what @code{myisamchk} is doing, add @code{-v} to tell it to run in verbose mode. @item myisamchk -eis tbl_name Shows only the most important information from a table. It is slow because it must read the whole table. @item myisamchk -eiv tbl_name This is like @code{-eis}, but tells you what is being done. @end table @cindex examples, @code{myisamchk} output @cindex @code{myisamchk}, example output Example of @code{myisamchk -d} output: @example MyISAM file: company.MYI Record format: Fixed length Data records: 1403698 Deleted blocks: 0 Recordlength: 226 table description: Key Start Len Index Type 1 2 8 unique double 2 15 10 multip. text packed stripped 3 219 8 multip. double 4 63 10 multip. text packed stripped 5 167 2 multip. unsigned short 6 177 4 multip. unsigned long 7 155 4 multip. text 8 138 4 multip. unsigned long 9 177 4 multip. unsigned long 193 1 text @end example Example of @code{myisamchk -d -v} output: @example MyISAM file: company Record format: Fixed length File-version: 1 Creation time: 1999-10-30 12:12:51 Recover time: 1999-10-31 19:13:01 Status: checked Data records: 1403698 Deleted blocks: 0 Datafile parts: 1403698 Deleted data: 0 Datafilepointer (bytes): 3 Keyfile pointer (bytes): 3 Max datafile length: 3791650815 Max keyfile length: 4294967294 Recordlength: 226 table description: Key Start Len Index Type Rec/key Root Blocksize 1 2 8 unique double 1 15845376 1024 2 15 10 multip. text packed stripped 2 25062400 1024 3 219 8 multip. double 73 40907776 1024 4 63 10 multip. text packed stripped 5 48097280 1024 5 167 2 multip. unsigned short 4840 55200768 1024 6 177 4 multip. unsigned long 1346 65145856 1024 7 155 4 multip. text 4995 75090944 1024 8 138 4 multip. unsigned long 87 85036032 1024 9 177 4 multip. unsigned long 178 96481280 1024 193 1 text @end example Example of @code{myisamchk -eis} output: @example Checking MyISAM file: company Key: 1: Keyblocks used: 97% Packed: 0% Max levels: 4 Key: 2: Keyblocks used: 98% Packed: 50% Max levels: 4 Key: 3: Keyblocks used: 97% Packed: 0% Max levels: 4 Key: 4: Keyblocks used: 99% Packed: 60% Max levels: 3 Key: 5: Keyblocks used: 99% Packed: 0% Max levels: 3 Key: 6: Keyblocks used: 99% Packed: 0% Max levels: 3 Key: 7: Keyblocks used: 99% Packed: 0% Max levels: 3 Key: 8: Keyblocks used: 99% Packed: 0% Max levels: 3 Key: 9: Keyblocks used: 98% Packed: 0% Max levels: 4 Total: Keyblocks used: 98% Packed: 17% Records: 1403698 M.recordlength: 226 Packed: 0% Recordspace used: 100% Empty space: 0% Blocks/Record: 1.00 Record blocks: 1403698 Delete blocks: 0 Recorddata: 317235748 Deleted data: 0 Lost space: 0 Linkdata: 0 User time 1626.51, System time 232.36 Maximum resident set size 0, Integral resident set size 0 Non physical pagefaults 0, Physical pagefaults 627, Swaps 0 Blocks in 0 out 0, Messages in 0 out 0, Signals 0 Voluntary context switches 639, Involuntary context switches 28966 @end example Example of @code{myisamchk -eiv} output: @example Checking MyISAM file: company Data records: 1403698 Deleted blocks: 0 - check file-size - check delete-chain block_size 1024: index 1: index 2: index 3: index 4: index 5: index 6: index 7: index 8: index 9: No recordlinks - check index reference - check data record references index: 1 Key: 1: Keyblocks used: 97% Packed: 0% Max levels: 4 - check data record references index: 2 Key: 2: Keyblocks used: 98% Packed: 50% Max levels: 4 - check data record references index: 3 Key: 3: Keyblocks used: 97% Packed: 0% Max levels: 4 - check data record references index: 4 Key: 4: Keyblocks used: 99% Packed: 60% Max levels: 3 - check data record references index: 5 Key: 5: Keyblocks used: 99% Packed: 0% Max levels: 3 - check data record references index: 6 Key: 6: Keyblocks used: 99% Packed: 0% Max levels: 3 - check data record references index: 7 Key: 7: Keyblocks used: 99% Packed: 0% Max levels: 3 - check data record references index: 8 Key: 8: Keyblocks used: 99% Packed: 0% Max levels: 3 - check data record references index: 9 Key: 9: Keyblocks used: 98% Packed: 0% Max levels: 4 Total: Keyblocks used: 9% Packed: 17% - check records and index references [LOTS OF ROW NUMBERS DELETED] Records: 1403698 M.recordlength: 226 Packed: 0% Recordspace used: 100% Empty space: 0% Blocks/Record: 1.00 Record blocks: 1403698 Delete blocks: 0 Recorddata: 317235748 Deleted data: 0 Lost space: 0 Linkdata: 0 User time 1639.63, System time 251.61 Maximum resident set size 0, Integral resident set size 0 Non physical pagefaults 0, Physical pagefaults 10580, Swaps 0 Blocks in 4 out 0, Messages in 0 out 0, Signals 0 Voluntary context switches 10604, Involuntary context switches 122798 @end example Here are the sizes of the data and index files for the table used in the preceding examples: @example -rw-rw-r-- 1 monty tcx 317235748 Jan 12 17:30 company.MYD -rw-rw-r-- 1 davida tcx 96482304 Jan 12 18:35 company.MYM @end example Explanations for the types of information @code{myisamchk} produces are given below. The ``keyfile'' is the index file. ``Record'' and ``row'' are synonymous: @table @code @item ISAM file Name of the ISAM (index) file. @item Isam-version Version of ISAM format. Currently always 2. @item Creation time When the data file was created. @item Recover time When the index/data file was last reconstructed. @item Data records How many records are in the table. @item Deleted blocks How many deleted blocks still have reserved space. You can optimize your table to minimize this space. @xref{Optimization}. @item Datafile: Parts For dynamic record format, this indicates how many data blocks there are. For an optimized table without fragmented records, this is the same as @code{Data records}. @item Deleted data How many bytes of non-reclaimed deleted data there are. You can optimize your table to minimize this space. @xref{Optimization}. @item Datafile pointer The size of the data file pointer, in bytes. It is usually 2, 3, 4, or 5 bytes. Most tables manage with 2 bytes, but this cannot be controlled from MySQL yet. For fixed tables, this is a record address. For dynamic tables, this is a byte address. @item Keyfile pointer The size of the index file pointer, in bytes. It is usually 1, 2, or 3 bytes. Most tables manage with 2 bytes, but this is calculated automatically by MySQL. It is always a block address. @item Max datafile length How long the table's data file (@code{.MYD} file) can become, in bytes. @item Max keyfile length How long the table's key file (@code{.MYI} file) can become, in bytes. @item Recordlength How much space each record takes, in bytes. @item Record format The format used to store table rows. The examples shown above use @code{Fixed length}. Other possible values are @code{Compressed} and @code{Packed}. @item table description A list of all keys in the table. For each key, some low-level information is presented: @table @code @item Key This key's number. @item Start Where in the record this index part starts. @item Len How long this index part is. For packed numbers, this should always be the full length of the column. For strings, it may be shorter than the full length of the indexed column, because you can index a prefix of a string column. @item Index @code{unique} or @code{multip.} (multiple). Indicates whether or not one value can exist multiple times in this index. @item Type What data-type this index part has. This is an ISAM data-type with the options @code{packed}, @code{stripped} or @code{empty}. @item Root Address of the root index block. @item Blocksize The size of each index block. By default this is 1024, but the value may be changed at compile time. @item Rec/key This is a statistical value used by the optimizer. It tells how many records there are per value for this key. A unique key always has a value of 1. This may be updated after a table is loaded (or greatly changed) with @code{myisamchk -a}. If this is not updated at all, a default value of 30 is given. @end table @item In the first example above, the 9th key is a multi-part key with two parts. @item Keyblocks used What percentage of the keyblocks are used. Because the table used in the examples had just been reorganized with @code{myisamchk}, the values are very high (very near the theoretical maximum). @item Packed MySQL tries to pack keys with a common suffix. This can only be used for @code{CHAR}/@code{VARCHAR}/@code{DECIMAL} keys. For long strings like names, this can significantly reduce the space used. In the third example above, the 4th key is 10 characters long and a 60% reduction in space is achieved. @item Max levels How deep the B-tree for this key is. Large tables with long keys get high values. @item Records How many rows are in the table. @item M.recordlength The average record length. For tables with fixed-length records, this is the exact record length. @item Packed MySQL strips spaces from the end of strings. The @code{Packed} value indicates the percentage of savings achieved by doing this. @item Recordspace used What percentage of the data file is used. @item Empty space What percentage of the data file is unused. @item Blocks/Record Average number of blocks per record (that is, how many links a fragmented record is composed of). This is always 1 for fixed-format tables. This value should stay as close to 1.0 as possible. If it gets too big, you can reorganize the table with @code{myisamchk}. @xref{Optimization}. @item Recordblocks How many blocks (links) are used. For fixed format, this is the same as the number of records. @item Deleteblocks How many blocks (links) are deleted. @item Recorddata How many bytes in the data file are used. @item Deleted data How many bytes in the data file are deleted (unused). @item Lost space If a record is updated to a shorter length, some space is lost. This is the sum of all such losses, in bytes. @item Linkdata When the dynamic table format is used, record fragments are linked with pointers (4 to 7 bytes each). @code{Linkdata} is the sum of the amount of storage used by all such pointers. @end table If a table has been compressed with @code{myisampack}, @code{myisamchk -d} prints additional information about each table column. See @ref{myisampack, , @code{myisampack}}, for an example of this information and a description of what it means. @node Database Administration, Localization, Disaster Prevention, MySQL Database Administration @section Database Administration Language Reference @menu * OPTIMIZE TABLE:: @code{OPTIMIZE TABLE} Syntax * ANALYZE TABLE:: @code{ANALYZE TABLE} Syntax * FLUSH:: @code{FLUSH} Syntax * KILL:: @code{KILL} Syntax * SHOW:: @code{SHOW} Syntax @end menu @node OPTIMIZE TABLE, ANALYZE TABLE, Database Administration, Database Administration @subsection @code{OPTIMIZE TABLE} Syntax @findex OPTIMIZE TABLE @cindex tables, defragmenting @cindex tables, fragmentation @example OPTIMIZE TABLE tbl_name[,tbl_name]... @end example @code{OPTIMIZE TABLE} should be used if you have deleted a large part of a table or if you have made many changes to a table with variable-length rows (tables that have @code{VARCHAR}, @code{BLOB}, or @code{TEXT} columns). Deleted records are maintained in a linked list and subsequent @code{INSERT} operations reuse old record positions. You can use @code{OPTIMIZE TABLE} to reclaim the unused space and to defragment the data file. For the moment @code{OPTIMIZE TABLE} only works on @strong{MyISAM} and @code{BDB} tables. For @code{BDB} tables, @code{OPTIMIZE TABLE} is currently mapped to @code{ANALYZE TABLE}. @xref{ANALYZE TABLE}. You can get optimize table to work on other table types by starting @code{mysqld} with @code{--skip-new} or @code{--safe-mode}, but in this case @code{OPTIMIZE TABLE} is just mapped to @code{ALTER TABLE}. @code{OPTIMIZE TABLE} works the following way: @itemize @bullet @item If the table has deleted or split rows, repair the table. @item If the index pages are not sorted, sort them. @item If the statistics are not up to date (and the repair couldn't be done by sorting the index), update them. @end itemize @code{OPTIMIZE TABLE} for @code{MyISAM} tables is equvialent of running @code{myisamchk --quick --check-changed-tables --sort-index --analyze} on the table. Note that the table is locked during the time @code{OPTIMIZE TABLE} is running! @node ANALYZE TABLE, FLUSH, OPTIMIZE TABLE, Database Administration @subsection @code{ANALYZE TABLE} Syntax @findex ANALYZE TABLE @example ANALYZE TABLE tbl_name[,tbl_name...] @end example Analyze and store the key distribution for the table. During the analyze the table is locked with a read lock. This works on @code{MyISAM} and @code{BDB} tables. This is equivalent to running @code{myisamchk -a} on the table. MySQL uses the stored key distribution to decide in which order tables should be joined when one does a join on something else than a constant. The command returns a table with the following columns: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Value} @item Table @tab Table name @item Op @tab Always ``analyze'' @item Msg_type @tab One of @code{status}, @code{error}, @code{info} or @code{warning}. @item Msg_text @tab The message. @end multitable You can check the stored key distribution with the @code{SHOW INDEX} command. @xref{SHOW DATABASE INFO}. If the table hasn't changed since the last @code{ANALYZE TABLE} command, the table will not be analyzed again. @node FLUSH, KILL, ANALYZE TABLE, Database Administration @subsection @code{FLUSH} Syntax @findex FLUSH @cindex @code{mysqladmin} @cindex clearing, caches @cindex caches, clearing @example FLUSH flush_option [,flush_option] @end example You should use the @code{FLUSH} command if you want to clear some of the internal caches MySQL uses. To execute @code{FLUSH}, you must have the @strong{RELOAD} privilege. @code{flush_option} can be any of the following: @multitable @columnfractions .15 .85 @item @code{HOSTS} @tab Empties the host cache tables. You should flush the host tables if some of your hosts change IP number or if you get the error message @code{Host ... is blocked}. When more than @code{max_connect_errors} errors occur in a row for a given host while connection to the MySQL server, MySQL assumes something is wrong and blocks the host from further connection requests. Flushing the host tables allows the host to attempt to connect again. @xref{Blocked host}.) You can start @code{mysqld} with @code{-O max_connection_errors=999999999} to avoid this error message. @item @code{LOGS} @tab Closes and reopens all log files. If you have specified the update log file or a binary log file without an extension, the extension number of the log file will be incremented by one relative to the previous file. If you have used an extension in the file name, MySQL will close and reopen the update log file. @xref{Update log}. This is the same thing as sending the @code{SIGHUP} signal to the @code{mysqld} server. @item @code{PRIVILEGES} @tab Reloads the privileges from the grant tables in the @code{mysql} database. @item @code{TABLES} @tab Closes all open tables and force all tables in use to be closed. @item @code{[TABLE | TABLES] table_name [,table_name...]} @tab Flushes only the given tables. @item @code{TABLES WITH READ LOCK} @tab Closes all open tables and locks all tables for all databases with a read until one executes @code{UNLOCK TABLES}. This is very convenient way to get backups if you have a file system, like Veritas,that can take snapshots in time. @item @code{STATUS} @tab Resets most status variables to zero. This is something one should only use when debugging a query. @end multitable You can also access each of the commands shown above with the @code{mysqladmin} utility, using the @code{flush-hosts}, @code{flush-logs}, @code{reload}, or @code{flush-tables} commands. Take also a look at the @code{RESET} command used with replication. @xref{Replication SQL}. @node KILL, SHOW, FLUSH, Database Administration @subsection @code{KILL} Syntax @findex KILL @cindex @code{mysqladmin} @example KILL thread_id @end example Each connection to @code{mysqld} runs in a separate thread. You can see which threads are running with the @code{SHOW PROCESSLIST} command and kill a thread with the @code{KILL thread_id} command. If you have the @strong{process} privilege, you can see and kill all threads. Otherwise, you can see and kill only your own threads. You can also use the @code{mysqladmin processlist} and @code{mysqladmin kill} commands to examine and kill threads. When you do a @code{KILL}, a thread specific @code{kill flag} is set for the thread. In most cases it may take some time for the thread to die as the kill flag is only checked at specific intervals. @itemize @bullet @item In @code{SELECT}, @code{ORDER BY} and @code{GROUP BY} loops, the flag is checked after reading a block of rows. If the kill flag is set the statement is aborted @item When doing an @code{ALTER TABLE} the kill flag is checked before each block of rows are read from the original table. If the kill flag was set the command is aborted and the temporary table is deleted. @item When doing an @code{UPDATE TABLE} and @code{DELETE TABLE}, the kill flag is checked after each block read and after each updated or delete row. If the kill flag is set the statement is aborted. Note that if you are not using transactions, the changes will not be rolled back! @item @code{GET_LOCK()} will abort with @code{NULL}. @item An @code{INSERT DELAYED} thread will quickly flush all rows it has in memory and die. @item If the thread is in the table lock handler (state: @code{Locked}), the table lock will be quickly aborted. @item If the thread is waiting for free disk space in a @code{write} call, the write is aborted with an disk full error message. @end itemize @menu * SHOW:: @code{SHOW} Syntax @end menu @node SHOW, , KILL, Database Administration @subsection @code{SHOW} Syntax @c FIX more index hits needed @findex SHOW DATABASE INFO @findex SHOW DATABASES @findex SHOW TABLES @findex SHOW COLUMNS @findex SHOW FIELDS @findex SHOW INDEX @findex SHOW KEYS @findex SHOW STATUS @findex SHOW VARIABLES @findex SHOW PROCESSLIST @findex SHOW TABLE STATUS @findex SHOW GRANTS @findex SHOW CREATE TABLE @findex SHOW MASTER STATUS @findex SHOW MASTER LOGS @findex SHOW SLAVE STATUS @example SHOW DATABASES [LIKE wild] or SHOW [OPEN] TABLES [FROM db_name] [LIKE wild] or SHOW [FULL] COLUMNS FROM tbl_name [FROM db_name] [LIKE wild] or SHOW INDEX FROM tbl_name [FROM db_name] or SHOW TABLE STATUS [FROM db_name] [LIKE wild] or SHOW STATUS [LIKE wild] or SHOW VARIABLES [LIKE wild] or SHOW LOGS or SHOW [FULL] PROCESSLIST or SHOW GRANTS FOR user or SHOW CREATE TABLE table_name or SHOW MASTER STATUS or SHOW MASTER LOGS or SHOW SLAVE STATUS @end example @code{SHOW} provides information about databases, tables, columns, or status information about the server. If the @code{LIKE wild} part is used, the @code{wild} string can be a string that uses the SQL @samp{%} and @samp{_} wild-card characters. @menu * SHOW DATABASE INFO:: Retrieving information about Database, Tables, Columns, and Indexes * SHOW TABLE STATUS:: @code{SHOW TABLE STATUS} * SHOW STATUS:: @code{SHOW STATUS} * SHOW VARIABLES:: @code{SHOW VARIABLES} * SHOW LOGS:: @code{SHOW LOGS} * SHOW PROCESSLIST:: @code{SHOW PROCESSLIST} * SHOW GRANTS:: @code{SHOW GRANTS} * SHOW CREATE TABLE:: @code{SHOW CREATE TABLE} @end menu @node SHOW DATABASE INFO, SHOW TABLE STATUS, SHOW, SHOW @subsubsection Retrieving information about Database, Tables, Columns, and Indexes @cindex displaying, information, @code{SHOW} You can use @code{db_name.tbl_name} as an alternative to the @code{tbl_name FROM db_name} syntax. These two statements are equivalent: @example mysql> SHOW INDEX FROM mytable FROM mydb; mysql> SHOW INDEX FROM mydb.mytable; @end example @code{SHOW DATABASES} lists the databases on the MySQL server host. You can also get this list using the @code{mysqlshow} command. @code{SHOW TABLES} lists the tables in a given database. You can also get this list using the @code{mysqlshow db_name} command. @strong{NOTE:} If a user doesn't have any privileges for a table, the table will not show up in the output from @code{SHOW TABLES} or @code{mysqlshow db_name}. @code{SHOW OPEN TABLES} lists the tables that are currently open in the table cache. @xref{Table cache}. The @code{Comment} field tells how many times the table is @code{cached} and @code{in_use}. @code{SHOW COLUMNS} lists the columns in a given table. If you specify the @code{FULL} option, you will also get the privileges you have for each column. If the column types are different than you expect them to be based on a @code{CREATE TABLE} statement, note that MySQL sometimes changes column types. @xref{Silent column changes}. The @code{DESCRIBE} statement provides information similar to @code{SHOW COLUMNS}. @xref{DESCRIBE, , @code{DESCRIBE}}. @code{SHOW FIELDS} is a synonym for @code{SHOW COLUMNS}, and @code{SHOW KEYS} is a synonym for @code{SHOW INDEX}. You can also list a table's columns or indexes with @code{mysqlshow db_name tbl_name} or @code{mysqlshow -k db_name tbl_name}. @code{SHOW INDEX} returns the index information in a format that closely resembles the @code{SQLStatistics} call in ODBC. The following columns are returned: @multitable @columnfractions .35 .65 @item @strong{Column} @tab @strong{Meaning} @item @code{Table} @tab Name of the table. @item @code{Non_unique} @tab 0 if the index can't contain duplicates. @item @code{Key_name} @tab Name of the index. @item @code{Seq_in_index} @tab Column sequence number in index, starting with 1. @item @code{Column_name} @tab Column name. @item @code{Collation} @tab How the column is sorted in the index. In MySQL, this can have values @samp{A} (Ascending) or @code{NULL} (Not sorted). @item @code{Cardinality} @tab Number of unique values in the index. This is updated by running @code{isamchk -a}. @item @code{Sub_part} @tab Number of indexed characters if the column is only partly indexed. @code{NULL} if the entire key is indexed. @item @code{Comment} @tab Various remarks. For now, it tells whether index is FULLTEXT or not. @end multitable Note that as the @code{Cardinality} is counted based on statistics stored as integers, it's not necessarily accurate for small tables. @node SHOW TABLE STATUS, SHOW STATUS, SHOW DATABASE INFO, SHOW @subsubsection @code{SHOW TABLE STATUS} @cindex displaying, table status @cindex tables, displaying status @cindex status, tables @example SHOW TABLE STATUS [FROM db_name] [LIKE wild] @end example @code{SHOW TABLE STATUS} (new in Version 3.23) works likes @code{SHOW STATUS}, but provides a lot of information about each table. You can also get this list using the @code{mysqlshow --status db_name} command. The following columns are returned: @multitable @columnfractions .30 .70 @item @strong{Column} @tab @strong{Meaning} @item @code{Name} @tab Name of the table. @item @code{Type} @tab Type of table. @xref{Table types}. @item @code{Row_format} @tab The row storage format (Fixed, Dynamic, or Compressed). @item @code{Rows} @tab Number of rows. @item @code{Avg_row_length} @tab Average row length. @item @code{Data_length} @tab Length of the data file. @item @code{Max_data_length} @tab Max length of the data file. @item @code{Index_length} @tab Length of the index file. @item @code{Data_free} @tab Number of allocated but not used bytes. @item @code{Auto_increment} @tab Next autoincrement value. @item @code{Create_time} @tab When the table was created. @item @code{Update_time} @tab When the data file was last updated. @item @code{Check_time} @tab When the table was last checked. @item @code{Create_options} @tab Extra options used with @code{CREATE TABLE}. @item @code{Comment} @tab The comment used when creating the table (or some information why MySQL couldn't access the table information). @end multitable @code{InnoDB} tables will report the free space in the tablespace in the table comment. @node SHOW STATUS, SHOW VARIABLES, SHOW TABLE STATUS, SHOW @subsubsection @code{SHOW STATUS} @cindex @code{mysqladmin} @code{SHOW STATUS} provides server status information (like @code{mysqladmin extended-status}). The output resembles that shown below, though the format and numbers probably differ: @example +--------------------------+------------+ | Variable_name | Value | +--------------------------+------------+ | Aborted_clients | 0 | | Aborted_connects | 0 | | Bytes_received | 155372598 | | Bytes_sent | 1176560426 | | Connections | 30023 | | Created_tmp_disk_tables | 0 | | Created_tmp_tables | 8340 | | Created_tmp_files | 60 | | Delayed_insert_threads | 0 | | Delayed_writes | 0 | | Delayed_errors | 0 | | Flush_commands | 1 | | Handler_delete | 462604 | | Handler_read_first | 105881 | | Handler_read_key | 27820558 | | Handler_read_next | 390681754 | | Handler_read_prev | 6022500 | | Handler_read_rnd | 30546748 | | Handler_read_rnd_next | 246216530 | | Handler_update | 16945404 | | Handler_write | 60356676 | | Key_blocks_used | 14955 | | Key_read_requests | 96854827 | | Key_reads | 162040 | | Key_write_requests | 7589728 | | Key_writes | 3813196 | | Max_used_connections | 0 | | Not_flushed_key_blocks | 0 | | Not_flushed_delayed_rows | 0 | | Open_tables | 1 | | Open_files | 2 | | Open_streams | 0 | | Opened_tables | 44600 | | Questions | 2026873 | | Select_full_join | 0 | | Select_full_range_join | 0 | | Select_range | 99646 | | Select_range_check | 0 | | Select_scan | 30802 | | Slave_running | OFF | | Slave_open_temp_tables | 0 | | Slow_launch_threads | 0 | | Slow_queries | 0 | | Sort_merge_passes | 30 | | Sort_range | 500 | | Sort_rows | 30296250 | | Sort_scan | 4650 | | Table_locks_immediate | 1920382 | | Table_locks_waited | 0 | | Threads_cached | 0 | | Threads_created | 30022 | | Threads_connected | 1 | | Threads_running | 1 | | Uptime | 80380 | +--------------------------+------------+ @end example @cindex variables, status The status variables listed above have the following meaning: @multitable @columnfractions .35 .65 @item @strong{Variable} @tab @strong{Meaning} @item @code{Aborted_clients} @tab Number of connections aborted because the client died without closing the connection properly. @xref{Communication errors}. @item @code{Aborted_connects} @tab Number of tries to connect to the MySQL server that failed. @xref{Communication errors}. @item @code{Bytes_received} @tab Number of bytes received from all clients. @item @code{Bytes_sent} @tab Number of bytes sent to all clients. @item @code{Com_xxxx} @tab Number of times the xxx commands has been executed. @item @code{Connections} @tab Number of connection attempts to the MySQL server. @item @code{Created_tmp_disk_tables} @tab Number of implicit temporary tables on disk created while executing statements. @item @code{Created_tmp_tables} @tab Number of implicit temporary tables in memory created while executing statements. @item @code{Created_tmp_files} @tab How many temporary files @code{mysqld} have created. @item @code{Delayed_insert_threads} @tab Number of delayed insert handler threads in use. @item @code{Delayed_writes} @tab Number of rows written with @code{INSERT DELAYED}. @item @code{Delayed_errors} @tab Number of rows written with @code{INSERT DELAYED} for which some error occurred (probably @code{duplicate key}). @item @code{Flush_commands} @tab Number of executed @code{FLUSH} commands. @item @code{Handler_delete} @tab Number of times a row was deleted from a table. @item @code{Handler_read_first} @tab Number of times the first entry was read from an index. If this is high, it suggests that the server is doing a lot of full index scans, for example, @code{SELECT col1 FROM foo}, assuming that col1 is indexed. @item @code{Handler_read_key} @tab Number of requests to read a row based on a key. If this is high, it is a good indication that your queries and tables are properly indexed. @item @code{Handler_read_next} @tab Number of requests to read next row in key order. This will be incremented if you are querying an index column with a range constraint. This also will be incremented if you are doing an index scan. @item @code{Handler_read_rnd} @tab Number of requests to read a row based on a fixed position. This will be high if you are doing a lot of queries that require sorting of the result. @item @code{Handler_read_rnd_next} @tab Number of requests to read the next row in the datafile. This will be high if you are doing a lot of table scans. Generally this suggests that your tables are not properly indexed or that your queries are not written to take advantage of the indexes you have. @item @code{Handler_update} @tab Number of requests to update a row in a table. @item @code{Handler_write} @tab Number of requests to insert a row in a table. @item @code{Key_blocks_used} @tab The number of used blocks in the key cache. @item @code{Key_read_requests} @tab The number of requests to read a key block from the cache. @item @code{Key_reads} @tab The number of physical reads of a key block from disk. @item @code{Key_write_requests} @tab The number of requests to write a key block to the cache. @item @code{Key_writes} @tab The number of physical writes of a key block to disk. @item @code{Max_used_connections} @tab The maximum number of connections in use simultaneously. @item @code{Not_flushed_key_blocks} @tab Keys blocks in the key cache that has changed but hasn't yet been flushed to disk. @item @code{Not_flushed_delayed_rows} @tab Number of rows waiting to be written in @code{INSERT DELAY} queues. @item @code{Open_tables} @tab Number of tables that are open. @item @code{Open_files} @tab Number of files that are open. @item @code{Open_streams} @tab Number of streams that are open (used mainly for logging). @item @code{Opened_tables} @tab Number of tables that have been opened. @item @code{Select_full_join} @tab Number of joins without keys (If this is 0, you should carefully check the index of your tables). @item @code{Select_full_range_join} @tab Number of joins where we used a range search on reference table. @item @code{Select_range} @tab Number of joins where we used ranges on the first table. (It's normally not critical even if this is big.) @item @code{Select_scan} @tab Number of joins where we did a full scann of the first table. @item @code{Select_range_check} @tab Number of joins without keys where we check for key usage after each row (If this is 0, you should carefully check the index of your tables). @item @code{Questions} @tab Number of queries sent to the server. @item @code{Slave_open_temp_tables} @tab Number of temporary tables currently open by the slave thread @item @code{Slow_launch_threads} @tab Number of threads that have taken more than @code{slow_launch_time} to create. @item @code{Slow_queries} @tab Number of queries that have taken more than @code{long_query_time}. @xref{Slow query log}. @item @code{Sort_merge_passes} @tab Number of merges passes the sort algoritm have had to do. If this value is large you should consider increasing @code{sort_buffer}. @item @code{Sort_range} @tab Number of sorts that where done with ranges. @item @code{Sort_rows} @tab Number of sorted rows. @item @code{Sort_scan} @tab Number of sorts that where done by scanning the table. @item @code{Table_locks_immediate} @tab Number of times a table lock was acquired immediately. Available after 3.23.33. @item @code{Table_locks_waited} @tab Number of times a table lock could not be acquired immediately and a wait was needed. If this is high, and you have performance problems, you should first optimize your queries, and then either split your table(s) or use replication. Available after 3.23.33. @item @code{Threads_cached} @tab Number of threads in the thread cache. @item @code{Threads_connected} @tab Number of currently open connections. @item @code{Threads_created} @tab Number of threads created to handle connections. @item @code{Threads_running} @tab Number of threads that are not sleeping. @item @code{Uptime} @tab How many seconds the server has been up. @end multitable Some comments about the above: @itemize @bullet @item If @code{Opened_tables} is big, then your @code{table_cache} variable is probably too small. @item If @code{key_reads} is big, then your @code{key_buffer_size} variable is probably too small. The cache hit rate can be calculated with @code{key_reads}/@code{key_read_requests}. @item If @code{Handler_read_rnd} is big, then you probably have a lot of queries that require MySQL to scan whole tables or you have joins that don't use keys properly. @item If @code{Threads_created} is big, you may want to increase the @code{thread_cache_size} variable. @end itemize @node SHOW VARIABLES, SHOW LOGS, SHOW STATUS, SHOW @subsubsection @code{SHOW VARIABLES} @example SHOW VARIABLES [LIKE wild] @end example @code{SHOW VARIABLES} shows the values of some MySQL system variables. You can also get this information using the @code{mysqladmin variables} command. If the default values are unsuitable, you can set most of these variables using command-line options when @code{mysqld} starts up. @xref{Command-line options}. The output resembles that shown below, though the format and numbers may differ somewhat: @example +-------------------------+---------------------------+ | Variable_name | Value | +-------------------------+---------------------------+ | ansi_mode | OFF | | back_log | 50 | | basedir | /my/monty/ | | bdb_cache_size | 16777216 | | bdb_log_buffer_size | 32768 | | bdb_home | /my/monty/data/ | | bdb_max_lock | 10000 | | bdb_logdir | | | bdb_shared_data | OFF | | bdb_tmpdir | /tmp/ | | binlog_cache_size | 32768 | | concurrent_insert | ON | | connect_timeout | 5 | | datadir | /my/monty/data/ | | delay_key_write | ON | | delayed_insert_limit | 100 | | delayed_insert_timeout | 300 | | delayed_queue_size | 1000 | | flush | OFF | | flush_time | 0 | | have_bdb | YES | | have_innodb | YES | | have_raid | YES | | have_ssl | NO | | init_file | | | interactive_timeout | 28800 | | join_buffer_size | 131072 | | key_buffer_size | 16776192 | | language | /my/monty/share/english/ | | large_files_support | ON | | log | OFF | | log_update | OFF | | log_bin | OFF | | log_slave_updates | OFF | | long_query_time | 10 | | low_priority_updates | OFF | | lower_case_table_names | 0 | | max_allowed_packet | 1048576 | | max_binlog_cache_size | 4294967295 | | max_connections | 100 | | max_connect_errors | 10 | | max_delayed_threads | 20 | | max_heap_table_size | 16777216 | | max_join_size | 4294967295 | | max_sort_length | 1024 | | max_tmp_tables | 32 | | max_write_lock_count | 4294967295 | | myisam_recover_options | DEFAULT | | myisam_sort_buffer_size | 8388608 | | net_buffer_length | 16384 | | net_read_timeout | 30 | | net_retry_count | 10 | | net_write_timeout | 60 | | open_files_limit | 0 | | pid_file | /my/monty/data/donna.pid | | port | 3306 | | protocol_version | 10 | | record_buffer | 131072 | | query_buffer_size | 0 | | safe_show_database | OFF | | server_id | 0 | | skip_locking | ON | | skip_networking | OFF | | skip_show_database | OFF | | slow_launch_time | 2 | | socket | /tmp/mysql.sock | | sort_buffer | 2097116 | | table_cache | 64 | | table_type | MYISAM | | thread_cache_size | 4 | | thread_stack | 65536 | | tmp_table_size | 1048576 | | tmpdir | /tmp/ | | version | 3.23.29a-gamma-debug | | wait_timeout | 28800 | +-------------------------+---------------------------+ @end example Each option is described below. Values for buffer sizes, lengths, and stack sizes are given in bytes. You can specify values with a suffix of @samp{K} or @samp{M} to indicate kilobytes or megabytes. For example, @code{16M} indicates 16 megabytes. The case of suffix letters does not matter; @code{16M} and @code{16m} are equivalent: @cindex variables, values @table @code @item @code{ansi_mode}. Is @code{ON} if @code{mysqld} was started with @code{--ansi}. @xref{ANSI mode}. @item @code{back_log} The number of outstanding connection requests MySQL can have. This comes into play when the main MySQL thread gets @strong{VERY} many connection requests in a very short time. It then takes some time (although very little) for the main thread to check the connection and start a new thread. The @code{back_log} value indicates how many requests can be stacked during this short time before MySQL momentarily stops answering new requests. You need to increase this only if you expect a large number of connections in a short period of time. In other words, this value is the size of the listen queue for incoming TCP/IP connections. Your operating system has its own limit on the size of this queue. The manual page for the Unix @code{listen(2)} system call should have more details. Check your OS documentation for the maximum value for this variable. Attempting to set @code{back_log} higher than your operating system limit will be ineffective. @item @code{basedir} The value of the @code{--basedir} option. @item @code{bdb_cache_size} The buffer that is allocated to cache index and rows for @code{BDB} tables. If you don't use @code{BDB} tables, you should start @code{mysqld} with @code{--skip-bdb} to not waste memory for this cache. @item @code{bdb_log_buffer_size} The buffer that is allocated to cache index and rows for @code{BDB} tables. If you don't use @code{BDB} tables, you should set this to 0 or start @code{mysqld} with @code{--skip-bdb} to not waste memory for this cache. @item @code{bdb_home} The value of the @code{--bdb-home} option. @item @code{bdb_max_lock} The maximum number of locks (1000 by default) you can have active on a BDB table. You should increase this if you get errors of type @code{bdb: Lock table is out of available locks} or @code{Got error 12 from ...} when you have do long transactions or when @code{mysqld} has to examine a lot of rows to calculate the query. @item @code{bdb_logdir} The value of the @code{--bdb-logdir} option. @item @code{bdb_shared_data} Is @code{ON} if you are using @code{--bdb-shared-data}. @item @code{bdb_tmpdir} The value of the @code{--bdb-tmpdir} option. @item @code{binlog_cache_size}. The size of the cache to hold the SQL statements for the binary log during a transaction. If you often use big, multi-statement transactions you can increase this to get more performance. @xref{COMMIT}. @item @code{character_set} The default character set. @item @code{character_sets} The supported character sets. @item @code{concurrent_inserts} If @code{ON} (the default), MySQL will allow you to use @code{INSERT} on @code{MyISAM} tables at the same time as you run @code{SELECT} queries on them. You can turn this option off by starting @code{mysqld} with @code{--safe} or @code{--skip-new}. @cindex timeout @item @code{connect_timeout} The number of seconds the @code{mysqld} server is waiting for a connect packet before responding with @code{Bad handshake}. @item @code{datadir} The value of the @code{--datadir} option. @item @code{delay_key_write} If enabled (is on by default), MySQL will honor the @code{delay_key_write} option @code{CREATE TABLE}. This means that the key buffer for tables with this option will not get flushed on every index update, but only when a table is closed. This will speed up writes on keys a lot, but you should add automatic checking of all tables with @code{myisamchk --fast --force} if you use this. Note that if you start @code{mysqld} with the @code{--delay-key-write-for-all-tables} option this means that all tables will be treated as if they were created with the @code{delay_key_write} option. You can clear this flag by starting @code{mysqld} with @code{--skip-new} or @code{--safe-mode}. @item @code{delayed_insert_limit} After inserting @code{delayed_insert_limit} rows, the @code{INSERT DELAYED} handler will check if there are any @code{SELECT} statements pending. If so, it allows these to execute before continuing. @item @code{delayed_insert_timeout} How long a @code{INSERT DELAYED} thread should wait for @code{INSERT} statements before terminating. @item @code{delayed_queue_size} What size queue (in rows) should be allocated for handling @code{INSERT DELAYED}. If the queue becomes full, any client that does @code{INSERT DELAYED} will wait until there is room in the queue again. @item @code{flush} This is @code{ON} if you have started MySQL with the @code{--flush} option. @item @code{flush_time} If this is set to a non-zero value, then every @code{flush_time} seconds all tables will be closed (to free up resources and sync things to disk). We only recommend this option on Win95, Win98, or on systems where you have very little resources. @item @code{have_bdb} @code{YES} if @code{mysqld} supports Berkeley DB tables. @code{DISABLED} if @code{--skip-bdb} is used. @item @code{have_innodb} @code{YES} if @code{mysqld} supports InnoDB tables. @code{DISABLED} if @code{--skip-innodb} is used. @item @code{have_raid} @code{YES} if @code{mysqld} supports the @code{RAID} option. @item @code{have_ssl} @code{YES} if @code{mysqld} supports SSL (encryption) on the client/server protocol. @item @code{init_file} The name of the file specified with the @code{--init-file} option when you start the server. This is a file of SQL statements you want the server to execute when it starts. @item @code{interactive_timeout} The number of seconds the server waits for activity on an interactive connection before closing it. An interactive client is defined as a client that uses the @code{CLIENT_INTERACTIVE} option to @code{mysql_real_connect()}. See also @code{wait_timeout}. @item @code{join_buffer_size} The size of the buffer that is used for full joins (joins that do not use indexes). The buffer is allocated one time for each full join between two tables. Increase this value to get a faster full join when adding indexes is not possible. (Normally the best way to get fast joins is to add indexes.) @c Make texi2html support index @anchor{Index cache size}. Then change @c some xrefs to point here @cindex indexes, block size @item @code{key_buffer_size} Index blocks are buffered and are shared by all threads. @code{key_buffer_size} is the size of the buffer used for index blocks. Increase this to get better index handling (for all reads and multiple writes) to as much as you can afford; 64M on a 256M machine that mainly runs MySQL is quite common. If you, however, make this too big (more than 50% of your total memory?) your system may start to page and become REALLY slow. Remember that because MySQL does not cache data read, that you will have to leave some room for the OS filesystem cache. You can check the performance of the key buffer by doing @code{show status} and examine the variables @code{Key_read_requests}, @code{Key_reads}, @code{Key_write_requests}, and @code{Key_writes}. The @code{Key_reads/Key_read_request} ratio should normally be < 0.01. The @code{Key_write/Key_write_requests} is usually near 1 if you are using mostly updates/deletes but may be much smaller if you tend to do updates that affect many at the same time or if you are using @code{delay_key_write}. @xref{SHOW}. To get even more speed when writing many rows at the same time, use @code{LOCK TABLES}. @xref{LOCK TABLES, , @code{LOCK TABLES}}. @item @code{language} The language used for error messages. @item @code{large_file_support} If @code{mysqld} was compiled with options for big file support. @item @code{locked_in_memory} If @code{mysqld} was locked in memory with @code{--memlock} @item @code{log} If logging of all queries is enabled. @item @code{log_update} If the update log is enabled. @item @code{log_bin} If the binary log is enabled. @item @code{log_slave_updates} If the updates from the slave should be logged. @item @code{long_query_time} If a query takes longer than this (in seconds), the @code{Slow_queries} counter will be incremented. If you are using @code{--log-slow-queries}, the query will be logged to the slow query logfile. @xref{Slow query log}. @item @code{lower_case_table_names} If set to 1 table names are stored in lowercase on disk and table names will be case-insensitive. @xref{Name case sensitivity}. @item @code{max_allowed_packet} The maximum size of one packet. The message buffer is initialized to @code{net_buffer_length} bytes, but can grow up to @code{max_allowed_packet} bytes when needed. This value by default is small, to catch big (possibly wrong) packets. You must increase this value if you are using big @code{BLOB} columns. It should be as big as the biggest @code{BLOB} you want to use. The current protocol limits @code{max_allowed_packet} to 16M. @item @code{max_binlog_cache_size} If a multi-statement transaction requires more than this amount of memory, one will get the error "Multi-statement transaction required more than 'max_binlog_cache_size' bytes of storage". @item @code{max_binlog_size} Available after 3.23.33. If a write to the binary (replication) log exceeds the given value, rotate the logs. You cannot set it to less than 1024 bytes, or more than 1 GB. Default is 1 GB. @item @code{max_connections} The number of simultaneous clients allowed. Increasing this value increases the number of file descriptors that @code{mysqld} requires. See below for comments on file descriptor limits. @xref{Too many connections}. @item @code{max_connect_errors} If there is more than this number of interrupted connections from a host this host will be blocked from further connections. You can unblock a host with the command @code{FLUSH HOSTS}. @item @code{max_delayed_threads} Don't start more than this number of threads to handle @code{INSERT DELAYED} statements. If you try to insert data into a new table after all @code{INSERT DELAYED} threads are in use, the row will be inserted as if the @code{DELAYED} attribute wasn't specified. @item @code{max_heap_table_size} Don't allow creation of heap tables bigger than this. @item @code{max_join_size} Joins that are probably going to read more than @code{max_join_size} records return an error. Set this value if your users tend to perform joins that lack a @code{WHERE} clause, that take a long time, and that return millions of rows. @item @code{max_sort_length} The number of bytes to use when sorting @code{BLOB} or @code{TEXT} values (only the first @code{max_sort_length} bytes of each value are used; the rest are ignored). @item @code{max_user_connections} The maximum number of active connections for a single user (0 = no limit). @item @code{max_tmp_tables} (This option doesn't yet do anything.) Maximum number of temporary tables a client can keep open at the same time. @item @code{max_write_lock_count} After this many write locks, allow some read locks to run in between. @item @code{myisam_recover_options} The value of the @code{--myisam-recover} option. @item @code{myisam_sort_buffer_size} The buffer that is allocated when sorting the index when doing a @code{REPAIR} or when creating indexes with @code{CREATE INDEX} or @code{ALTER TABLE}. @item @code{myisam_max_extra_sort_file_size}. If the creating of the temporary file for fast index creation would be this much bigger than using the key cache, then prefer the key cache method. This is mainly used to force long character keys in large tables to use the slower key cache method to create the index. @strong{NOTE} that this parameter is given in megabytes! @item @code{myisam_max_sort_file_size} The maximum size of the temporary file MySQL is allowed to use while recreating the index (during @code{REPAIR}, @code{ALTER TABLE} or @code{LOAD DATA INFILE}. If the file size would be bigger than this, the index will be created through the key cache (which is slower). @strong{NOTE} that this parameter is given in megabytes! @item @code{net_buffer_length} The communication buffer is reset to this size between queries. This should not normally be changed, but if you have very little memory, you can set it to the expected size of a query. (That is, the expected length of SQL statements sent by clients. If statements exceed this length, the buffer is automatically enlarged, up to @code{max_allowed_packet} bytes.) @item @code{net_read_timeout} Number of seconds to wait for more data from a connection before aborting the read. Note that when we don't expect data from a connection, the timeout is defined by @code{write_timeout}. See also @code{slave_read_timeout}. @item @code{net_retry_count} If a read on a communication port is interrupted, retry this many times before giving up. This value should be quite high on @code{FreeBSD} as internal interrupts are sent to all threads. @item @code{net_write_timeout} Number of seconds to wait for a block to be written to a connection before aborting the write. @item @code{open_files_limit} If this is not 0, then @code{mysqld} will use this value to reserve file descriptors to use with @code{setrlimit()}. If this value is 0 then @code{mysqld} will reserve @code{max_connections*5} or @code{max_connections + table_cache*2} (whichever is larger) number of files. You should try increasing this if @code{mysqld} gives you the error 'Too many open files'. @item @code{pid_file} The value of the @code{--pid-file} option. @item @code{port} The value of the @code{--port} option. @item @code{protocol_version} The protocol version used by the MySQL server. @item @code{record_buffer} Each thread that does a sequential scan allocates a buffer of this size for each table it scans. If you do many sequential scans, you may want to increase this value. @item @code{record_rnd_buffer} When reading rows in sorted order after a sort, the rows are read through this buffer to avoid a disk seeks. If not set, then it's set to the value of @code{record_buffer}. @item @code{query_buffer_size} The initial allocation of the query buffer. If most of your queries are long (like when inserting blobs), you should increase this! @item @code{safe_show_databases} Don't show databases for which the user doesn't have any database or table privileges. This can improve security if you're concerned about people being able to see what databases other users have. See also @code{skip_show_databases}. @item @code{server_id} The value of the @code{--server-id} option. @item @code{skip_locking} Is OFF if @code{mysqld} uses external locking. @item @code{skip_networking} Is ON if we only allow local (socket) connections. @item @code{skip_show_databases} This prevents people from doing @code{SHOW DATABASES} if they don't have the @code{PROCESS_PRIV} privilege. This can improve security if you're concerned about people being able to see what databases other users have. See also @code{safe_show_databases}. @item @code{slave_read_timeout} Number of seconds to wait for more data from a master/slave connection before aborting the read. @item @code{slow_launch_time} If creating the thread takes longer than this value (in seconds), the @code{Slow_launch_threads} counter will be incremented. @item @code{socket} The Unix socket used by the server. @item @code{sort_buffer} Each thread that needs to do a sort allocates a buffer of this size. Increase this value for faster @code{ORDER BY} or @code{GROUP BY} operations. @xref{Temporary files}. @item @code{table_cache} The number of open tables for all threads. Increasing this value increases the number of file descriptors that @code{mysqld} requires. MySQL needs two file descriptors for each unique open table. See below for comments on file descriptor limits. You can check if you need to increase the table cache by checking the @code{Opened_tables} variable. @xref{SHOW}. If this variable is big and you don't do @code{FLUSH TABLES} a lot (which just forces all tables to be closed and reopenend), then you should increase the value of this variable. Make sure that your operating system can handle the number of open file descriptors implied by the @code{table_cache} setting. If @code{table_cache} is set too high, MySQL may run out of file descriptors and refuse connections, fail to perform queries, and be very unreliable. For information about how the table cache works, see @ref{Table cache}. @item @code{table_type} The default table type @item @code{thread_cache_size} How many threads we should keep in a cache for reuse. When a client disconnects, the client's threads are put in the cache if there aren't more than @code{thread_cache_size} threads from before. All new threads are first taken from the cache, and only when the cache is empty is a new thread created. This variable can be increased to improve performance if you have a lot of new connections. (Normally this doesn't give a notable performance improvement if you have a good thread implementation.) By examing the difference between the @code{Connections} and @code{Threads_created} you can see how efficient the current thread cache is for you. @item @code{thread_concurrency} On Solaris, @code{mysqld} will call @code{thr_setconcurrency()} with this value. @code{thr_setconcurrency()} permits the application to give the threads system a hint for the desired number of threads that should be run at the same time. @item @code{thread_stack} The stack size for each thread. Many of the limits detected by the @code{crash-me} test are dependent on this value. The default is large enough for normal operation. @xref{MySQL Benchmarks}. @item @code{timezone} The timezone for the server. @item @code{tmp_table_size} If an in-memory temporary table exceeds this size, MySQL will automatically convert it to an on-disk @code{MyISAM} table. Increase the value of @code{tmp_table_size} if you do many advanced @code{GROUP BY} queries and you have lots of memory. @item @code{tmpdir} The directory used for temporary files and temporary tables. @item @code{version} The version number for the server. @item @code{wait_timeout} The number of seconds the server waits for activity on a connection before closing it. See also @code{interactive_timeout}. @end table The manual section that describes tuning MySQL contains some information of how to tune the above variables. @xref{Server parameters}. @node SHOW LOGS, SHOW PROCESSLIST, SHOW VARIABLES, SHOW @subsubsection @code{SHOW LOGS} @code{SHOW LOGS} shows you status information about existing log files. It currently only displays information about Berkeley DB log files. @itemize @bullet @item @code{File} shows the full path to the log file @item @code{Type} shows the type of the log file (@code{BDB} for Berkeley DB log files) @item @code{Status} shows the status of the log file (@code{FREE} if the file can be removed, or @code{IN USE} if the file is needed by the transaction subsystem) @end itemize @node SHOW PROCESSLIST, SHOW GRANTS, SHOW LOGS, SHOW @subsubsection @code{SHOW PROCESSLIST} @findex threads @findex PROCESSLIST @cindex threads, display @cindex processes, display @code{SHOW PROCESSLIST} shows you which threads are running. You can also get this information using the @code{mysqladmin processlist} command. If you have the @strong{process} privilege, you can see all threads. Otherwise, you can see only your own threads. @xref{KILL, , @code{KILL}}. If you don't use the @code{FULL} option, then only the first 100 characters of each query will be shown. This command is very useful if you get the 'too many connections' error message and want to find out what's going on. MySQL reserves one extra connection for a client with the @code{Process_priv} privilege to ensure that you should always be able to login and check the system (assuming you are not giving this privilege to all your users). @node SHOW GRANTS, SHOW CREATE TABLE, SHOW PROCESSLIST, SHOW @subsubsection @code{SHOW GRANTS} @cindex privileges, display @code{SHOW GRANTS FOR user} lists the grant commands that must be issued to duplicate the grants for a user. @example mysql> SHOW GRANTS FOR root@@localhost; +---------------------------------------------------------------------+ | Grants for root@@localhost | +---------------------------------------------------------------------+ | GRANT ALL PRIVILEGES ON *.* TO 'root'@@'localhost' WITH GRANT OPTION | +---------------------------------------------------------------------+ @end example @node SHOW CREATE TABLE, , SHOW GRANTS, SHOW @subsubsection @code{SHOW CREATE TABLE} Shows a @code{CREATE TABLE} statement that will create the given table: @example mysql> show create table t\G *************************** 1. row *************************** Table: t Create Table: CREATE TABLE t ( id int(11) default NULL auto_increment, s char(60) default NULL, PRIMARY KEY (id) ) TYPE=MyISAM @end example @code{SHOW CREATE TABLE} will quote table and column names according to @code{SQL_QUOTE_SHOW_CREATE} option. @ref{SET OPTION, , @code{SET OPTION SQL_QUOTE_SHOW_CREATE}}. @node Localization, Server-Side Scripts, Database Administration, MySQL Database Administration @section MySQL Localization and International Usage @menu * Character sets:: The Character Set Used for Data and Sorting * Languages:: Non-English Error Messages * Adding character set:: Adding a New Character Set * Character arrays:: The character definition arrays * String collating:: String Collating Support * Multi-byte characters:: Multi-byte Character Support @end menu @node Character sets, Languages, Localization, Localization @subsection The Character Set Used for Data and Sorting @cindex character sets @cindex data, character sets @cindex sorting, character sets By default, MySQL uses the ISO-8859-1 (Latin1) character set with sorting according to Swedish/Finnish. This is the character set suitable in the USA and western Europe. All standard MySQL binaries are compiled with @code{--with-extra-charsets=complex}. This will add code to all standard programs to be able to handle @code{latin1} and all multi-byte character sets within the binary. Other character sets will be loaded from a character-set definition file when needed. The character set determines what characters are allowed in names and how things are sorted by the @code{ORDER BY} and @code{GROUP BY} clauses of the @code{SELECT} statement. You can change the character set with the @code{--default-character-set} option when you start the server. The character sets available depend on the @code{--with-charset=charset} and @code{--with-extra-charset= list-of-charset | complex | all} options to @code{configure}, and the character set configuration files listed in @file{SHAREDIR/charsets/Index}. @xref{configure options}. If you change the character set when running MySQL (which may also change the sort order), you must run myisamchk -r -q on all tables. Otherwise your indexes may not be ordered correctly. When a client connects to a MySQL server, the server sends the default character set in use to the client. The client will switch to use this character set for this connection. One should use @code{mysql_real_escape_string()} when escaping strings for a SQL query. @code{mysql_real_escape_string()} is identical to the old @code{mysql_escape_string()} function, except that it takes the MYSQL connection handle as the first parameter. If the client is compiled with different paths than where the server is installed and the user who configured MySQL didn't included all character sets in the MySQL binary, one must specify for the client where it can find the additional character sets it will need if the server runs with a different character set than the client. One can specify this by putting in a MySQL option file: @example [client] character-sets-dir=/usr/local/mysql/share/mysql/charsets @end example where the path points to where the dynamic MySQL character sets are stored. One can force the client to use specific character set by specifying: @example [client] default-character-set=character-set-name @end example but normally this is never needed. @node Languages, Adding character set, Character sets, Localization @subsection Non-English Error Messages @cindex error messages, languages @cindex messages, languages @cindex files, error messages @cindex language support @code{mysqld} can issue error messages in the following languages: Czech, Danish, Dutch, English (the default), Estonian, French, German, Greek, Hungarian, Italian, Japanese, Korean, Norwegian, Norwegian-ny, Polish, Portuguese, Romanian, Russian, Slovak, Spanish, and Swedish. To start @code{mysqld} with a particular language, use either the @code{--language=lang} or @code{-L lang} options. For example: @example shell> mysqld --language=swedish @end example or: @example shell> mysqld --language=/usr/local/share/swedish @end example Note that all language names are specified in lowercase. The language files are located (by default) in @file{@var{mysql_base_dir}/share/@var{LANGUAGE}/}. To update the error message file, you should edit the @file{errmsg.txt} file and execute the following command to generate the @file{errmsg.sys} file: @example shell> comp_err errmsg.txt errmsg.sys @end example If you upgrade to a newer version of MySQL, remember to repeat your changes with the new @file{errmsg.txt} file. @node Adding character set, Character arrays, Languages, Localization @subsection Adding a New Character Set @cindex character sets, adding @cindex adding, character sets To add another character set to MySQL, use the following procedure. Decide if the set is simple or complex. If the character set does not need to use special string collating routines for sorting and does not need multi-byte character support, it is simple. If it needs either of those features, it is complex. For example, @code{latin1} and @code{danish} are simple charactersets while @code{big5} or @code{czech} are complex character sets. In the following section, we have assumed that you name your character set @code{MYSET}. For a simple character set do the following: @enumerate @item Add MYSET to the end of the @file{sql/share/charsets/Index} file Assign an unique number to it. @item Create the file @file{sql/share/charsets/MYSET.conf}. (You can use @file{sql/share/charsets/latin1.conf} as a base for this). The syntax for the file very simple: @itemize @bullet @item Comments start with a '#' character and proceed to the end of the line. @item Words are separated by arbitrary amounts of whitespace. @item When defining the character set, every word must be a number in hexadecimal format @item The @code{ctype} array takes up the first 257 words. The @code{to_lower}, @code{to_upper} and @code{sort_order} arrays take up 256 words each after that. @end itemize @xref{Character arrays}. @item Add the character set name to the @code{CHARSETS_AVAILABLE} and @code{COMPILED_CHARSETS} lists in @code{configure.in}. @item Reconfigure, recompile, and test. @end enumerate For a complex character set do the following: @enumerate @item Create the file @file{strings/ctype-MYSET.c} in the MySQL source distribution. @item Add MYSET to the end of the @file{sql/share/charsets/Index} file. Assign an unique number to it. @item Look at one of the existing @file{ctype-*.c} files to see what needs to be defined, for example @file{strings/ctype-big5.c}. Note that the arrays in your file must have names like @code{ctype_MYSET}, @code{to_lower_MYSET}, and so on. This corresponds to the arrays in the simple character set. @xref{Character arrays}. For a complex character set @item Near the top of the file, place a special comment like this: @example /* * This comment is parsed by configure to create ctype.c, * so don't change it unless you know what you are doing. * * .configure. number_MYSET=MYNUMBER * .configure. strxfrm_multiply_MYSET=N * .configure. mbmaxlen_MYSET=N */ @end example The @code{configure} program uses this comment to include the character set into the MySQL library automatically. The strxfrm_multiply and mbmaxlen lines will be explained in the following sections. Only include them if you the string collating functions or the multi-byte character set functions, respectively. @item You should then create some of the following functions: @itemize @bullet @item @code{my_strncoll_MYSET()} @item @code{my_strcoll_MYSET()} @item @code{my_strxfrm_MYSET()} @item @code{my_like_range_MYSET()} @end itemize @xref{String collating}. @item Add the character set name to the @code{CHARSETS_AVAILABLE} and @code{COMPILED_CHARSETS} lists in @code{configure.in}. @item Reconfigure, recompile, and test. @end enumerate The file @file{sql/share/charsets/README} includes some more instructions. If you want to have the character set included in the MySQL distribution, mail a patch to @email{internals@@lists.mysql.com}. @node Character arrays, String collating, Adding character set, Localization @subsection The character definition arrays @code{to_lower[]} and @code{to_upper[]} are simple arrays that hold the lowercase and uppercase characters corresponding to each member of the character set. For example: @example to_lower['A'] should contain 'a' to_upper['a'] should contain 'A' @end example @code{sort_order[]} is a map indicating how characters should be ordered for comparison and sorting purposes. For many character sets, this is the same as @code{to_upper[]} (which means sorting will be case insensitive). MySQL will sort characters based on the value of @code{sort_order[character]}. For more complicated sorting rules, see the discussion of string collating below. @xref{String collating}. @code{ctype[]} is an array of bit values, with one element for one character. (Note that @code{to_lower[]}, @code{to_upper[]}, and @code{sort_order[]} are indexed by character value, but @code{ctype[]} is indexed by character value + 1. This is an old legacy to be able to handle EOF.) You can find the following bitmask definitions in @file{m_ctype.h}: @example #define _U 01 /* Uppercase */ #define _L 02 /* Lowercase */ #define _N 04 /* Numeral (digit) */ #define _S 010 /* Spacing character */ #define _P 020 /* Punctuation */ #define _C 040 /* Control character */ #define _B 0100 /* Blank */ #define _X 0200 /* heXadecimal digit */ @end example The @code{ctype[]} entry for each character should be the union of the applicable bitmask values that describe the character. For example, @code{'A'} is an uppercase character (@code{_U}) as well as a hexadecimal digit (@code{_X}), so @code{ctype['A'+1]} should contain the value: @example _U + _X = 01 + 0200 = 0201 @end example @node String collating, Multi-byte characters, Character arrays, Localization @subsection String Collating Support @cindex collating, strings @cindex string collating If the sorting rules for your language are too complex to be handled with the simple @code{sort_order[]} table, you need to use the string collating functions. Right now the best documentation on this is the character sets that are already implemented. Look at the big5, czech, gbk, sjis, and tis160 character sets for examples. You must specify the @code{strxfrm_multiply_MYSET=N} value in the special comment at the top of the file. @code{N} should be set to the maximum ratio the strings may grow during @code{my_strxfrm_MYSET} (it must be a positive integer). @node Multi-byte characters, , String collating, Localization @subsection Multi-byte Character Support @cindex characters, multi-byte @cindex multi-byte characters If your want to add support for a new character set that includes multi-byte characters, you need to use the multi-byte character functions. Right now the best documentation on this is the character sets that are already implemented. Look at the euc_kr, gb2312, gbk, sjis and ujis character sets for examples. These are implemented in the @code{ctype-'charset'.c} files in the @file{strings} directory. You must specify the @code{mbmaxlen_MYSET=N} value in the special comment at the top of the source file. @code{N} should be set to the size in bytes of the largest character in the set. @node Server-Side Scripts, Client-Side Scripts, Localization, MySQL Database Administration @section MySQL Server-Side Scripts and Utilities @menu * Server-Side Overview:: Overview of the Server-Side Scripts and Utilities * safe_mysqld:: safe_mysqld, the wrapper around mysqld * mysqld_multi:: mysqld_multi, program for managing multiple MySQL servers * myisampack:: myisampack, The MySQL Compressed Read-only Table Generator * mysqld-max:: mysqld-max, An extended mysqld server @end menu @node Server-Side Overview, safe_mysqld, Server-Side Scripts, Server-Side Scripts @subsection Overview of the Server-Side Scripts and Utilities @cindex environment variables @cindex programs, list of All MySQL clients that communicate with the server using the @code{mysqlclient} library use the following environment variables: @tindex @code{MYSQL_UNIX_PORT} environment variable @tindex environment variable, @code{MYSQL_UNIX_PORT} @tindex @code{MYSQL_TCP_PORT} environment variable @tindex environment variable, @code{MYSQL_TCP_PORT} @tindex @code{MYSQL_PWD} environment variable @tindex environment variable, @code{MYSQL_PWD} @tindex @code{MYSQL_DEBUG} environment variable @tindex environment variable, @code{MYSQL_DEBUG} @multitable @columnfractions .25 .75 @item @strong{Name} @tab @strong{Description} @item @code{MYSQL_UNIX_PORT} @tab The default socket; used for connections to @code{localhost} @item @code{MYSQL_TCP_PORT} @tab The default TCP/IP port @item @code{MYSQL_PWD} @tab The default password @item @code{MYSQL_DEBUG} @tab Debug-trace options when debugging @item @code{TMPDIR} @tab The directory where temporary tables/files are created @end multitable Use of @code{MYSQL_PWD} is insecure. @xref{Connecting}. @tindex @code{MYSQL_HISTFILE} environment variable @tindex environment variable, @code{MYSQL_HISTFILE} @tindex @code{HOME} environment variable @tindex environment variable, @code{HOME} @cindex history file @cindex command line history @tindex .mysql_history file The @file{mysql} client uses the file named in the @code{MYSQL_HISTFILE} environment variable to save the command-line history. The default value for the history file is @file{$HOME/.mysql_history}, where @code{$HOME} is the value of the @code{HOME} environment variable. @xref{Environment variables}. All MySQL programs take many different options. However, every MySQL program provides a @code{--help} option that you can use to get a full description of the program's different options. For example, try @code{mysql --help}. You can override default options for all standard client programs with an option file. @ref{Option files}. The list below briefly describes the MySQL programs: @table @code @cindex @code{myisamchk} @item myisamchk Utility to describe, check, optimize, and repair MySQL tables. Because @code{myisamchk} has many functions, it is described in its own chapter. @xref{MySQL Database Administration}. @cindex @code{make_binary_distribution} @item make_binary_distribution Makes a binary release of a compiled MySQL. This could be sent by FTP to @file{/pub/mysql/Incoming} on @code{support.mysql.com} for the convenience of other MySQL users. @cindex @code{msql2mysql} @item msql2mysql A shell script that converts @code{mSQL} programs to MySQL. It doesn't handle all cases, but it gives a good start when converting. @cindex @code{mysqlaccess} @item mysqlaccess A script that checks the access privileges for a host, user, and database combination. @cindex @code{mysqladmin} @item mysqladmin Utility for performing administrative operations, such as creating or dropping databases, reloading the grant tables, flushing tables to disk, and reopening log files. @code{mysqladmin} can also be used to retrieve version, process, and status information from the server. @xref{mysqladmin, , @code{mysqladmin}}. @cindex @code{mysqlbug} @item mysqlbug The MySQL bug report script. This script should always be used when filing a bug report to the MySQL list. @cindex @code{mysqld} @item mysqld The SQL daemon. This should always be running. @cindex @code{mysqldump} @item mysqldump Dumps a MySQL database into a file as SQL statements or as tab-separated text files. Enhanced freeware originally by Igor Romanenko. @xref{mysqldump, , @code{mysqldump}}. @cindex @code{mysqlimport} @item mysqlimport Imports text files into their respective tables using @code{LOAD DATA INFILE}. @xref{mysqlimport, , @code{mysqlimport}}. @cindex @code{mysqlshow} @item mysqlshow Displays information about databases, tables, columns, and indexes. @cindex @code{mysql_install_db} @item mysql_install_db Creates the MySQL grant tables with default privileges. This is usually executed only once, when first installing MySQL on a system. @cindex @code{replace} @item replace A utility program that is used by @code{msql2mysql}, but that has more general applicability as well. @code{replace} changes strings in place in files or on the standard input. Uses a finite state machine to match longer strings first. Can be used to swap strings. For example, this command swaps @code{a} and @code{b} in the given files: @example shell> replace a b b a -- file1 file2 ... @end example @end table @node safe_mysqld, mysqld_multi, Server-Side Overview, Server-Side Scripts @subsection safe_mysqld, the wrapper around mysqld @cindex tools, safe_mysqld @cindex scripts @cindex @code{safe_mysqld} @code{safe_mysqld} is the recommended way to start a @code{mysqld} daemon on Unix. @code{safe_mysqld} adds some safety features such as restarting the server when an error occurs and logging run-time information to a log file. If you don't use @code{--mysqld=#} or @code{--mysqld-version=#} @code{safe_mysqld} will use an executable named @code{mysqld-max} if it exists. If not, @code{safe_mysqld} will start @code{mysqld}. This makes it very easy to test to use @code{mysqld-max} instead of @code{mysqld}; Just copy @code{mysqld-max} to where you have @code{mysqld} and it will be used. Normally one should never edit the @code{safe_mysqld} script, but instead put the options to @code{safe_mysqld} in the @code{[safe_mysqld]} section in the @code{my.cnf} file. @code{safe_mysqld} will read all options from the @code{[mysqld]}, @code{[server]} and @code{[safe_mysqld]} sections from the option files. @xref{Option files}. Note that all options on the command line to @code{safe_mysqld} are passed to @code{mysqld}. If you wants to use any options in @code{safe_mysqld} that @code{mysqld} doesn't support, you must specify these in the option file. Most of the options to @code{safe_mysqld} are the same as the options to @code{mysqld}. @xref{Command-line options}. @code{safe_mysqld} supports the following options: @table @code @item --basedir=path @item --core-file-size=# Size of the core file @code{mysqld} should be able to create. Passed to @code{ulimit -c}. @item --datadir=path @item --defaults-extra-file=path @item --defaults-file=path @item --err-log=path @item --ledir=path Path to @code{mysqld} @item --log=path @item --mysqld=mysqld-version Name of the @code{mysqld} version in the @code{ledir} directory you want to start. @item --mysqld-version=version Similar to @code{--mysqld=} but here you only give the suffix for @code{mysqld}. For example if you use @code{--mysqld-version=max}, @code{safe_mysqld} will start the @code{ledir/mysqld-max} version. If the argument to @code{--mysqld-version} is empty, @code{ledir/mysqld} will be used. @item --no-defaults @item --open-files-limit=# Number of files @code{mysqld} should be able to open. Passed to @code{ulimit -n}. Note that you need to start @code{safe_mysqld} as root for this to work properly! @item --pid-file=path @item --port=# @item --socket=path @item --timezone=# Set the timezone (the @code{TZ}) variable to the value of this parameter. @item --user=# @end table The @code{safe_mysqld} script is written so that it normally is able to start a server that was installed from either a source or a binary version of MySQL, even if these install the server in slightly different locations. @code{safe_mysqld} expects one of these conditions to be true: @itemize @bullet @item The server and databases can be found relative to the directory from which @code{safe_mysqld} is invoked. @code{safe_mysqld} looks under its working directory for @file{bin} and @file{data} directories (for binary distributions) or for @file{libexec} and @file{var} directories (for source distributions). This condition should be met if you execute @code{safe_mysqld} from your MySQL installation directory (for example, @file{/usr/local/mysql} for a binary distribution). @item If the server and databases cannot be found relative to the working directory, @code{safe_mysqld} attempts to locate them by absolute pathnames. Typical locations are @file{/usr/local/libexec} and @file{/usr/local/var}. The actual locations are determined when the distribution was built from which @code{safe_mysqld} comes. They should be correct if MySQL was installed in a standard location. @end itemize Because @code{safe_mysqld} will try to find the server and databases relative to its own working directory, you can install a binary distribution of MySQL anywhere, as long as you start @code{safe_mysqld} from the MySQL installation directory: @example shell> cd mysql_installation_directory shell> bin/safe_mysqld & @end example If @code{safe_mysqld} fails, even when invoked from the MySQL installation directory, you can modify it to use the path to @code{mysqld} and the pathname options that are correct for your system. Note that if you upgrade MySQL in the future, your modified version of @code{safe_mysqld} will be overwritten, so you should make a copy of your edited version that you can reinstall. @node mysqld_multi, myisampack, safe_mysqld, Server-Side Scripts @subsection mysqld_multi, program for managing multiple MySQL servers @cindex tools, mysqld_multi @cindex scripts @cindex multi mysqld @cindex @code{mysqld_multi} @code{mysqld_multi} is meant for managing several @code{mysqld} processes running in different UNIX sockets and TCP/IP ports. The program will search for group(s) named [mysqld#] from my.cnf (or the given --config-file=...), where # can be any positive number starting from 1. These groups should be the same as the usual @code{[mysqld]} group (e.g. options to mysqld, see MySQL manual for detailed information about this group), but with those port, socket etc. options that are wanted for each separate @code{mysqld} processes. The number in the group name has another function; it can be used for starting, stopping, or reporting some specific @code{mysqld} servers with this program. See the usage and options below for more information. @example Usage: mysqld_multi [OPTIONS] @{start|stop|report@} [GNR,GNR,GNR...] or mysqld_multi [OPTIONS] @{start|stop|report@} [GNR-GNR,GNR,GNR-GNR,...] @end example The GNR above means the group number. You can start, stop or report any GNR, or several of them at the same time. (See --example) The GNRs list can be comma separated, or a dash combined, of which the latter means that all the GNRs between GNR1-GNR2 will be affected. Without GNR argument all the found groups will be either started, stopped, or reported. Note that you must not have any white spaces in the GNR list. Anything after a white space is ignored. @code{mysqld_multi} supports the following options: @table @code @cindex config-file option @item --config-file=... Alternative config file. NOTE: This will not affect this program's own options (group @code{[mysqld_multi]}), but only groups [mysqld#]. Without this option everything will be searched from the ordinary my.cnf file. @cindex example option @item --example Give an example of a config file. @cindex help option @item --help Print this help and exit. @cindex log option @item --log=... Log file. Full path to and the name for the log file. NOTE: If the file exists, everything will be appended. @cindex mysqladmin option @item --mysqladmin=... @code{mysqladmin} binary to be used for a server shutdown. @cindex mysqld option @item --mysqld=... @code{mysqld} binary to be used. Note that you can give @code{safe_mysqld} to this option also. The options are passed to @code{mysqld}. Just make sure you have @code{mysqld} in your environment variable @code{PATH} or fix @code{safe_mysqld}. @cindex no-log option @item --no-log Print to stdout instead of the log file. By default the log file is turned on. @cindex password option @item --password=... Password for user for @code{mysqladmin}. @cindex tcp-ip option @item --tcp-ip Connect to the MySQL server(s) via the TCP/IP port instead of the UNIX socket. This affects stopping and reporting. If a socket file is missing, the server may still be running, but can be accessed only via the TCP/IP port. By default connecting is done via the UNIX socket. @cindex user option @item --user=... MySQL user for @code{mysqladmin}. @cindex version option @item --version Print the version number and exit. @end table Some notes about @code{mysqld_multi}: @itemize @bullet @item Make sure that the MySQL user, who is stopping the @code{mysqld} services (e.g using the @code{mysqladmin}) have the same password and username for all the data directories accessed (to the 'mysql' database) And make sure that the user has the 'Shutdown_priv' privilege! If you have many data- directories and many different 'mysql' databases with different passwords for the MySQL 'root' user, you may want to create a common 'multi_admin' user for each using the same password (see below). Example how to do it: @example shell> mysql -u root -S /tmp/mysql.sock -proot_password -e "GRANT SHUTDOWN ON *.* TO multi_admin@@localhost IDENTIFIED BY 'multipass'" @xref{Privileges}. @end example You will have to do the above for each @code{mysqld} running in each data directory, that you have (just change the socket, -S=...) @item @code{pid-file} is very important, if you are using @code{safe_mysqld} to start @code{mysqld} (e.g. --mysqld=safe_mysqld) Every @code{mysqld} should have its own @code{pid-file}. The advantage using @code{safe_mysqld} instead of @code{mysqld} directly here is, that @code{safe_mysqld} 'guards' every @code{mysqld} process and will restart it, if a @code{mysqld} process fails due to signal kill -9, or similar. (Like segmentation fault, which MySQL should never do, of course ;) Please note that @code{safe_mysqld} script may require that you start it from a certain place. This means that you may have to CD to a certain directory, before you start the @code{mysqld_multi}. If you have problems starting, please see the @code{safe_mysqld} script. Check especially the lines: @example -------------------------------------------------------------------------- MY_PWD=`pwd` Check if we are starting this relative (for the binary release) if test -d /data/mysql -a -f ./share/mysql/english/errmsg.sys -a -x ./bin/mysqld -------------------------------------------------------------------------- @xref{safe_mysqld, , @code{safe_mysqld}}. @end example The above test should be successful, or you may encounter problems. @item Beware of the dangers starting multiple @code{mysqlds} in the same data directory. Use separate data directories, unless you @strong{KNOW} what you are doing! @item The socket file and the TCP/IP port must be different for every @code{mysqld}. @item The first and fifth @code{mysqld} group were intentionally left out from the example. You may have 'gaps' in the config file. This gives you more flexibility. The order in which the @code{mysqlds} are started or stopped depends on the order in which they appear in the config file. @item When you want to refer to a certain group using GNR with this program, just use the number in the end of the group name ( [mysqld# <== ). @item You may want to use option '--user' for @code{mysqld}, but in order to do this you need to be root when you start the @code{mysqld_multi} script. Having the option in the config file doesn't matter; you will just get a warning, if you are not the superuser and the @code{mysqlds} are started under @strong{YOUR} UNIX account. @strong{IMPORTANT}: Make sure that the @code{pid-file} and the data directory are read+write(+execute for the latter one) accessible for @strong{THAT} UNIX user, who the specific @code{mysqld} process is started as. @strong{DON'T} use the UNIX root account for this, unless you @strong{KNOW} what you are doing! @item @strong{MOST IMPORTANT}: Make sure that you understand the meanings of the options that are passed to the @code{mysqlds} and why @strong{WOULD YOU WANT} to have separate @code{mysqld} processes. Starting multiple @code{mysqlds} in one data directory @strong{WILL NOT} give you extra performance in a threaded system! @end itemize @xref{Multiple servers}. This is an example of the config file on behalf of @code{mysqld_multi}. @example # This file should probably be in your home dir (~/.my.cnf) or /etc/my.cnf # Version 2.1 by Jani Tolonen [mysqld_multi] mysqld = /usr/local/bin/safe_mysqld mysqladmin = /usr/local/bin/mysqladmin user = multi_admin password = multipass [mysqld2] socket = /tmp/mysql.sock2 port = 3307 pid-file = /usr/local/mysql/var2/hostname.pid2 datadir = /usr/local/mysql/var2 language = /usr/local/share/mysql/english user = john [mysqld3] socket = /tmp/mysql.sock3 port = 3308 pid-file = /usr/local/mysql/var3/hostname.pid3 datadir = /usr/local/mysql/var3 language = /usr/local/share/mysql/swedish user = monty [mysqld4] socket = /tmp/mysql.sock4 port = 3309 pid-file = /usr/local/mysql/var4/hostname.pid4 datadir = /usr/local/mysql/var4 language = /usr/local/share/mysql/estonia user = tonu [mysqld6] socket = /tmp/mysql.sock6 port = 3311 pid-file = /usr/local/mysql/var6/hostname.pid6 datadir = /usr/local/mysql/var6 language = /usr/local/share/mysql/japanese user = jani @end example @xref{Option files}. @node myisampack, mysqld-max, mysqld_multi, Server-Side Scripts @subsection myisampack, The MySQL Compressed Read-only Table Generator @cindex compressed tables @cindex tables, compressed @cindex MyISAM, compressed tables @cindex @code{myisampack} @cindex @code{pack_isam} @code{myisampack} is used to compress MyISAM tables, and @code{pack_isam} is used to compress ISAM tables. Because ISAM tables are deprecated, we will only discuss @code{myisampack} here, but everything said about @code{myisampack} should also be true for @code{pack_isam}. @code{myisampack} works by compressing each column in the table separately. The information needed to decompress columns is read into memory when the table is opened. This results in much better performance when accessing individual records, because you only have to uncompress exactly one record, not a much larger disk block as when using Stacker on MS-DOS. Usually, @code{myisampack} packs the data file 40%-70%. MySQL uses memory mapping (@code{mmap()}) on compressed tables and falls back to normal read/write file usage if @code{mmap()} doesn't work. There are currently two limitations with @code{myisampack}: @itemize @bullet @item After packing, the table is read-only. @item @code{myisampack} can also pack @code{BLOB} or @code{TEXT} columns. The older @code{pack_isam} could not do this. @end itemize Fixing these limitations is on our TODO list but with low priority. @code{myisampack} is invoked like this: @example shell> myisampack [options] filename ... @end example Each filename should be the name of an index (@file{.MYI}) file. If you are not in the database directory, you should specify the pathname to the file. It is permissible to omit the @file{.MYI} extension. @code{myisampack} supports the following options: @table @code @item -b, --backup Make a backup of the table as @code{tbl_name.OLD}. @item -#, --debug=debug_options Output debug log. The @code{debug_options} string often is @code{'d:t:o,filename'}. @item -f, --force Force packing of the table even if it becomes bigger or if the temporary file exists. @code{myisampack} creates a temporary file named @file{tbl_name.TMD} while it compresses the table. If you kill @code{myisampack}, the @file{.TMD} file may not be deleted. Normally, @code{myisampack} exits with an error if it finds that @file{tbl_name.TMD} exists. With @code{--force}, @code{myisampack} packs the table anyway. @item -?, --help Display a help message and exit. @item -j big_tbl_name, --join=big_tbl_name Join all tables named on the command line into a single table @code{big_tbl_name}. All tables that are to be combined MUST be identical (same column names and types, same indexes, etc.). @item -p #, --packlength=# Specify the record length storage size, in bytes. The value should be 1, 2, or 3. (@code{myisampack} stores all rows with length pointers of 1, 2, or 3 bytes. In most normal cases, @code{myisampack} can determine the right length value before it begins packing the file, but it may notice during the packing process that it could have used a shorter length. In this case, @code{myisampack} will print a note that the next time you pack the same file, you could use a shorter record length.) @item -s, --silent Silent mode. Write output only when errors occur. @item -t, --test Don't actually pack table, just test packing it. @item -T dir_name, --tmp_dir=dir_name Use the named directory as the location in which to write the temporary table. @item -v, --verbose Verbose mode. Write information about progress and packing result. @item -V, --version Display version information and exit. @item -w, --wait Wait and retry if table is in use. If the @code{mysqld} server was invoked with the @code{--skip-locking} option, it is not a good idea to invoke @code{myisampack} if the table might be updated during the packing process. @end table @cindex examples, compressed tables The sequence of commands shown below illustrates a typical table compression session: @example shell> ls -l station.* -rw-rw-r-- 1 monty my 994128 Apr 17 19:00 station.MYD -rw-rw-r-- 1 monty my 53248 Apr 17 19:00 station.MYI -rw-rw-r-- 1 monty my 5767 Apr 17 19:00 station.frm shell> myisamchk -dvv station MyISAM file: station Isam-version: 2 Creation time: 1996-03-13 10:08:58 Recover time: 1997-02-02 3:06:43 Data records: 1192 Deleted blocks: 0 Datafile: Parts: 1192 Deleted data: 0 Datafile pointer (bytes): 2 Keyfile pointer (bytes): 2 Max datafile length: 54657023 Max keyfile length: 33554431 Recordlength: 834 Record format: Fixed length table description: Key Start Len Index Type Root Blocksize Rec/key 1 2 4 unique unsigned long 1024 1024 1 2 32 30 multip. text 10240 1024 1 Field Start Length Type 1 1 1 2 2 4 3 6 4 4 10 1 5 11 20 6 31 1 7 32 30 8 62 35 9 97 35 10 132 35 11 167 4 12 171 16 13 187 35 14 222 4 15 226 16 16 242 20 17 262 20 18 282 20 19 302 30 20 332 4 21 336 4 22 340 1 23 341 8 24 349 8 25 357 8 26 365 2 27 367 2 28 369 4 29 373 4 30 377 1 31 378 2 32 380 8 33 388 4 34 392 4 35 396 4 36 400 4 37 404 1 38 405 4 39 409 4 40 413 4 41 417 4 42 421 4 43 425 4 44 429 20 45 449 30 46 479 1 47 480 1 48 481 79 49 560 79 50 639 79 51 718 79 52 797 8 53 805 1 54 806 1 55 807 20 56 827 4 57 831 4 shell> myisampack station.MYI Compressing station.MYI: (1192 records) - Calculating statistics normal: 20 empty-space: 16 empty-zero: 12 empty-fill: 11 pre-space: 0 end-space: 12 table-lookups: 5 zero: 7 Original trees: 57 After join: 17 - Compressing file 87.14% shell> ls -l station.* -rw-rw-r-- 1 monty my 127874 Apr 17 19:00 station.MYD -rw-rw-r-- 1 monty my 55296 Apr 17 19:04 station.MYI -rw-rw-r-- 1 monty my 5767 Apr 17 19:00 station.frm shell> myisamchk -dvv station MyISAM file: station Isam-version: 2 Creation time: 1996-03-13 10:08:58 Recover time: 1997-04-17 19:04:26 Data records: 1192 Deleted blocks: 0 Datafile: Parts: 1192 Deleted data: 0 Datafilepointer (bytes): 3 Keyfile pointer (bytes): 1 Max datafile length: 16777215 Max keyfile length: 131071 Recordlength: 834 Record format: Compressed table description: Key Start Len Index Type Root Blocksize Rec/key 1 2 4 unique unsigned long 10240 1024 1 2 32 30 multip. text 54272 1024 1 Field Start Length Type Huff tree Bits 1 1 1 constant 1 0 2 2 4 zerofill(1) 2 9 3 6 4 no zeros, zerofill(1) 2 9 4 10 1 3 9 5 11 20 table-lookup 4 0 6 31 1 3 9 7 32 30 no endspace, not_always 5 9 8 62 35 no endspace, not_always, no empty 6 9 9 97 35 no empty 7 9 10 132 35 no endspace, not_always, no empty 6 9 11 167 4 zerofill(1) 2 9 12 171 16 no endspace, not_always, no empty 5 9 13 187 35 no endspace, not_always, no empty 6 9 14 222 4 zerofill(1) 2 9 15 226 16 no endspace, not_always, no empty 5 9 16 242 20 no endspace, not_always 8 9 17 262 20 no endspace, no empty 8 9 18 282 20 no endspace, no empty 5 9 19 302 30 no endspace, no empty 6 9 20 332 4 always zero 2 9 21 336 4 always zero 2 9 22 340 1 3 9 23 341 8 table-lookup 9 0 24 349 8 table-lookup 10 0 25 357 8 always zero 2 9 26 365 2 2 9 27 367 2 no zeros, zerofill(1) 2 9 28 369 4 no zeros, zerofill(1) 2 9 29 373 4 table-lookup 11 0 30 377 1 3 9 31 378 2 no zeros, zerofill(1) 2 9 32 380 8 no zeros 2 9 33 388 4 always zero 2 9 34 392 4 table-lookup 12 0 35 396 4 no zeros, zerofill(1) 13 9 36 400 4 no zeros, zerofill(1) 2 9 37 404 1 2 9 38 405 4 no zeros 2 9 39 409 4 always zero 2 9 40 413 4 no zeros 2 9 41 417 4 always zero 2 9 42 421 4 no zeros 2 9 43 425 4 always zero 2 9 44 429 20 no empty 3 9 45 449 30 no empty 3 9 46 479 1 14 4 47 480 1 14 4 48 481 79 no endspace, no empty 15 9 49 560 79 no empty 2 9 50 639 79 no empty 2 9 51 718 79 no endspace 16 9 52 797 8 no empty 2 9 53 805 1 17 1 54 806 1 3 9 55 807 20 no empty 3 9 56 827 4 no zeros, zerofill(2) 2 9 57 831 4 no zeros, zerofill(1) 2 9 @end example The information printed by @code{myisampack} is described below: @table @code @item normal The number of columns for which no extra packing is used. @item empty-space The number of columns containing values that are only spaces; these will occupy 1 bit. @item empty-zero The number of columns containing values that are only binary 0's; these will occupy 1 bit. @item empty-fill The number of integer columns that don't occupy the full byte range of their type; these are changed to a smaller type (for example, an @code{INTEGER} column may be changed to @code{MEDIUMINT}). @item pre-space The number of decimal columns that are stored with leading spaces. In this case, each value will contain a count for the number of leading spaces. @item end-space The number of columns that have a lot of trailing spaces. In this case, each value will contain a count for the number of trailing spaces. @item table-lookup The column had only a small number of different values, which were converted to an @code{ENUM} before Huffman compression. @item zero The number of columns for which all values are zero. @item Original trees The initial number of Huffman trees. @item After join The number of distinct Huffman trees left after joining trees to save some header space. @end table After a table has been compressed, @code{myisamchk -dvv} prints additional information about each field: @table @code @item Type The field type may contain the following descriptors: @table @code @item constant All rows have the same value. @item no endspace Don't store endspace. @item no endspace, not_always Don't store endspace and don't do end space compression for all values. @item no endspace, no empty Don't store endspace. Don't store empty values. @item table-lookup The column was converted to an @code{ENUM}. @item zerofill(n) The most significant @code{n} bytes in the value are always 0 and are not stored. @item no zeros Don't store zeros. @item always zero 0 values are stored in 1 bit. @end table @item Huff tree The Huffman tree associated with the field. @item Bits The number of bits used in the Huffman tree. @end table After you have run @code{pack_isam}/@code{myisampack} you must run @code{isamchk}/@code{myisamchk} to re-create the index. At this time you can also sort the index blocks and create statistics needed for the MySQL optimizer to work more efficiently: @example myisamchk -rq --analyze --sort-index table_name.MYI isamchk -rq --analyze --sort-index table_name.ISM @end example After you have installed the packed table into the MySQL database directory you should do @code{mysqladmin flush-tables} to force @code{mysqld} to start using the new table. If you want to unpack a packed table, you can do this with the @code{--unpack} option to @code{isamchk} or @code{myisamchk}. @node mysqld-max, , myisampack, Server-Side Scripts @subsection mysqld-max, An extended mysqld server @cindex @code{mysqld-max} @code{mysqld-max} is the MySQL server (@code{mysqld}) configured with the following configure options: @multitable @columnfractions .3 .7 @item @strong{Option} @tab @strong{Comment} @item --with-server-suffix=-max @tab Add a suffix to the @code{mysqld} version string. @item --with-bdb @tab Support for Berkeley DB (BDB) tables @item --with-innodb @tab Support for InnoDB tables. @item CFLAGS=-DUSE_SYMDIR @tab Symbolic links support for Windows. @end multitable You can find the MySQL-max binaries at @uref{http://www.mysql.com/downloads/mysql-max-3.23.html}. The Windows MySQL 3.23 binary distribution includes both the standard @code{mysqld.exe} binary and the @code{mysqld-max.exe} binary. @uref{http://www.mysql.com/downloads/mysql-3.23.html}. @xref{Windows installation}. Note that as Berkeley DB and InnoDB are not available for all platforms, some of the @code{Max} binaries may not have support for both of these. You can check which table types are supported by doing the following query: @example mysql> show variables like "have_%"; +---------------+-------+ | Variable_name | Value | +---------------+-------+ | have_bdb | YES | | have_innodb | NO | | have_isam | YES | | have_raid | NO | | have_ssl | NO | +---------------+-------+ @end example The meaning of the values are: @multitable @columnfractions .3 .7 @item @strong{Value} @tab @strong{Meaning}. @item YES @tab The option is activated and usable. @item NO @tab MySQL is not compiled with support for this option. @item DISABLED @tab The xxxx option is disabled because one started @code{mysqld} with @code{--skip-xxxx} or because one didn't start @code{mysqld} with all needed options to enable the option. In this case the @code{hostname.err} file should contain a reason for why the option is disabled. @end multitable @strong{NOTE}: To be able to create InnoDB tables you @strong{MUST} edit your startup options to include at least the @code{innodb_data_file_path} option. @xref{InnoDB start}. To get better performance for BDB tables, you should add some configuration options for these too. @xref{BDB start}. @code{safe_mysqld} will automatically try to start any @code{mysqld} binary with the @code{-max} prefix. This makes it very easy to test out a another @code{mysqld} binary in an existing installation. Just run @code{configure} with the options you want and then install the new @code{mysqld} binary as @code{mysqld-max} in the same directory where your old @code{mysqld} binary is. @xref{safe_mysqld, , @code{safe_mysqld}}. The @code{mysqld-max} RPM uses the above mentioned @code{safe_mysqld} feature. It just installs the @code{mysqld-max} executable and @code{safe_mysqld} will automatically use this executable when @code{safe_mysqld} is restarted. The following table shows which table types our standard @strong{MySQL-Max} binaries includes: @multitable @columnfractions .4 .3 .3 @item @strong{System} @tab @strong{BDB} @tab @strong{InnoDB} @item AIX 4.3 @tab N @tab Y @item HP-UX 11.0 @tab N @tab Y @item Linux-Alpha @tab N @tab Y @item Linux-Intel @tab Y @tab Y @item Linux-Ia64 @tab N @tab Y @item Solaris-intel @tab N @tab Y @item Solaris-sparc @tab Y @tab Y @item SCO OSR5 @tab Y @tab Y @item UnixWare @tab Y @tab Y @item Windows/NT @tab Y @tab Y @end multitable @node Client-Side Scripts, Log Files, Server-Side Scripts, MySQL Database Administration @section MySQL Client-Side Scripts and Utilities @menu * Client-Side Overview:: Overview of the Client-Side Scripts and Utilities * mysql:: The Command-line Tool * mysqladmin:: mysqladmin, Administrating a MySQL Server * Using mysqlcheck:: Using @code{mysqlcheck} for Table Maintenance and Crash Recovery * mysqldump:: mysqldump, Dumping Table Structure and Data * mysqlhotcopy:: mysqlhotcopy, Copying MySQL Databases and Tables * mysqlimport:: mysqlimport, Importing Data from Text Files * mysqlshow:: Showing Databases, Tables, and Columns * perror:: perror, Explaining Error Codes * Batch Commands:: How to Run SQL Commands from a Text File @end menu @node Client-Side Overview, mysql, Client-Side Scripts, Client-Side Scripts @subsection Overview of the Client-Side Scripts and Utilities @cindex environment variables @cindex programs, list of All MySQL clients that communicate with the server using the @code{mysqlclient} library use the following environment variables: @tindex @code{MYSQL_UNIX_PORT} environment variable @tindex environment variable, @code{MYSQL_UNIX_PORT} @tindex @code{MYSQL_TCP_PORT} environment variable @tindex environment variable, @code{MYSQL_TCP_PORT} @tindex @code{MYSQL_PWD} environment variable @tindex environment variable, @code{MYSQL_PWD} @tindex @code{MYSQL_DEBUG} environment variable @tindex environment variable, @code{MYSQL_DEBUG} @multitable @columnfractions .25 .75 @item @strong{Name} @tab @strong{Description} @item @code{MYSQL_UNIX_PORT} @tab The default socket; used for connections to @code{localhost} @item @code{MYSQL_TCP_PORT} @tab The default TCP/IP port @item @code{MYSQL_PWD} @tab The default password @item @code{MYSQL_DEBUG} @tab Debug-trace options when debugging @item @code{TMPDIR} @tab The directory where temporary tables/files are created @end multitable Use of @code{MYSQL_PWD} is insecure. @xref{Connecting}. @tindex @code{MYSQL_HISTFILE} environment variable @tindex environment variable, @code{MYSQL_HISTFILE} @tindex @code{HOME} environment variable @tindex environment variable, @code{HOME} @cindex history file @cindex command line history @tindex .mysql_history file The @file{mysql} client uses the file named in the @code{MYSQL_HISTFILE} environment variable to save the command-line history. The default value for the history file is @file{$HOME/.mysql_history}, where @code{$HOME} is the value of the @code{HOME} environment variable. @xref{Environment variables}. All MySQL programs take many different options. However, every MySQL program provides a @code{--help} option that you can use to get a full description of the program's different options. For example, try @code{mysql --help}. You can override default options for all standard client programs with an option file. @ref{Option files}. The list below briefly describes the MySQL programs: @table @code @cindex @code{myisamchk} @item myisamchk Utility to describe, check, optimize, and repair MySQL tables. Because @code{myisamchk} has many functions, it is described in its own chapter. @xref{MySQL Database Administration}. @cindex @code{make_binary_distribution} @item make_binary_distribution Makes a binary release of a compiled MySQL. This could be sent by FTP to @file{/pub/mysql/Incoming} on @code{support.mysql.com} for the convenience of other MySQL users. @cindex @code{msql2mysql} @item msql2mysql A shell script that converts @code{mSQL} programs to MySQL. It doesn't handle all cases, but it gives a good start when converting. @cindex @code{mysqlaccess} @item mysqlaccess A script that checks the access privileges for a host, user, and database combination. @cindex @code{mysqladmin} @item mysqladmin Utility for performing administrative operations, such as creating or dropping databases, reloading the grant tables, flushing tables to disk, and reopening log files. @code{mysqladmin} can also be used to retrieve version, process, and status information from the server. @xref{mysqladmin, , @code{mysqladmin}}. @cindex @code{mysqlbug} @item mysqlbug The MySQL bug report script. This script should always be used when filing a bug report to the MySQL list. @cindex @code{mysqld} @item mysqld The SQL daemon. This should always be running. @cindex @code{mysqldump} @item mysqldump Dumps a MySQL database into a file as SQL statements or as tab-separated text files. Enhanced freeware originally by Igor Romanenko. @xref{mysqldump, , @code{mysqldump}}. @cindex @code{mysqlimport} @item mysqlimport Imports text files into their respective tables using @code{LOAD DATA INFILE}. @xref{mysqlimport, , @code{mysqlimport}}. @cindex @code{mysqlshow} @item mysqlshow Displays information about databases, tables, columns, and indexes. @cindex @code{mysql_install_db} @item mysql_install_db Creates the MySQL grant tables with default privileges. This is usually executed only once, when first installing MySQL on a system. @cindex @code{replace} @item replace A utility program that is used by @code{msql2mysql}, but that has more general applicability as well. @code{replace} changes strings in place in files or on the standard input. Uses a finite state machine to match longer strings first. Can be used to swap strings. For example, this command swaps @code{a} and @code{b} in the given files: @example shell> replace a b b a -- file1 file2 ... @end example @end table @node mysql, mysqladmin, Client-Side Overview, Client-Side Scripts @subsection The Command-line Tool @cindex command line tool @cindex tools, command line @cindex scripts @cindex @code{mysql} @code{mysql} is a simple SQL shell (with GNU @code{readline} capabilities). It supports interactive and non-interactive use. When used interactively, query results are presented in an ASCII-table format. When used non-interactively (for example, as a filter), the result is presented in tab-separated format. (The output format can be changed using command-line options.) You can run scripts simply like this: @example shell> mysql database < script.sql > output.tab @end example If you have problems due to insufficient memory in the client, use the @code{--quick} option! This forces @code{mysql} to use @code{mysql_use_result()} rather than @code{mysql_store_result()} to retrieve the result set. Using @code{mysql} is very easy. Just start it as follows: @code{mysql database} or @code{mysql --user=user_name --password=your_password database}. Type a SQL statement, end it with @samp{;}, @samp{\g}, or @samp{\G} and press RETURN/ENTER. @cindex @code{mysql} command line options @cindex command line options, @code{mysql} @cindex options, command line, @code{mysql} @cindex startup parameters, @code{mysql} @code{mysql} supports the following options: @table @code @cindex @code{help}, @code{mysql} option @item -?, --help Display this help and exit. @cindex @code{no-auto-rehash}, @code{mysql} option @item -A, --no-auto-rehash No automatic rehashing. One has to use 'rehash' to get table and field completion. This gives a quicker start of mysql. @cindex @code{batch}, @code{mysql} option @item -B, --batch Print results with a tab as separator, each row on a new line. Doesn't use history file. @cindex @code{character-sets-dir}, @code{mysql} option @item --character-sets-dir=... Directory where character sets are located. @cindex @code{compress}, @code{mysql} option @item -C, --compress Use compression in server/client protocol. @cindex @code{debug}, @code{mysql} option @item -#, --debug[=...] Debug log. Default is 'd:t:o,/tmp/mysql.trace'. @cindex @code{database}, @code{mysql} option @item -D, --database=... Database to use. This is mainly useful in the @code{my.cnf} file. @cindex @code{default-character-set}, @code{mysql} option @item --default-character-set=... Set the default character set. @cindex @code{execute}, @code{mysql} option @item -e, --execute=... Execute command and quit. (Output like with --batch) @cindex @code{vertical}, @code{mysql} option @item -E, --vertical Print the output of a query (rows) vertically. Without this option you can also force this output by ending your statements with @code{\G}. @cindex @code{force}, @code{mysql} option @item -f, --force Continue even if we get a SQL error. @cindex @code{no-named-commands}, @code{mysql} option @item -g, --no-named-commands Named commands are disabled. Use \* form only, or use named commands only in the beginning of a line ending with a semicolon (;). Since Version 10.9, the client now starts with this option ENABLED by default! With the -g option, long format commands will still work from the first line, however. @cindex @code{enable-named-commands}, @code{mysql} option @item -G, --enable-named-commands Named commands are @strong{enabled}. Long format commands are allowed as well as shortened \* commands. @cindex @code{ignore-space}, @code{mysql} option @item -i, --ignore-space Ignore space after function names. @cindex @code{host}, @code{mysql} option @item -h, --host=... Connect to the given host. @cindex @code{html}, @code{mysql} option @item -H, --html Produce HTML output. @cindex @code{skip-line-numbers}, @code{mysql} option @item -L, --skip-line-numbers Don't write line number for errors. Useful when one wants to compare result files that includes error messages @cindex @code{no-pager}, @code{mysql} option @item --no-pager Disable pager and print to stdout. See interactive help (\h) also. @cindex @code{no-tee}, @code{mysql} option @item --no-tee Disable outfile. See interactive help (\h) also. @cindex @code{unbuffered}, @code{mysql} option @item -n, --unbuffered Flush buffer after each query. @cindex @code{skip-column-names}, @code{mysql} option @item -N, --skip-column-names Don't write column names in results. @cindex @code{set-variable}, @code{mysql} option @item -O, --set-variable var=option Give a variable a value. @code{--help} lists variables. @cindex @code{one-database}, @code{mysql} option @item -o, --one-database Only update the default database. This is useful for skipping updates to other database in the update log. @cindex @code{pager}, @code{mysql} option @item @code{--pager[=...]} Output type. Default is your @code{ENV} variable @code{PAGER}. Valid pagers are less, more, cat [> filename], etc. See interactive help (\h) also. This option does not work in batch mode. Pager works only in UNIX. @cindex @code{password}, @code{mysql} option @item -p[password], --password[=...] Password to use when connecting to server. If a password is not given on the command line, you will be prompted for it. Note that if you use the short form @code{-p} you can't have a space between the option and the password. @cindex @code{port}, @code{mysql} option @item -P --port=... TCP/IP port number to use for connection. @cindex @code{quick}, @code{mysql} option @item -q, --quick Don't cache result, print it row-by-row. This may slow down the server if the output is suspended. Doesn't use history file. @cindex @code{raw}, @code{mysql} option @item -r, --raw Write column values without escape conversion. Used with @code{--batch} @cindex @code{silent}, @code{mysql} option @item -s, --silent Be more silent. @cindex @code{socket}, @code{mysql} option @item -S --socket=... Socket file to use for connection. @cindex @code{table}, @code{mysql} option @item -t --table Output in table format. This is default in non-batch mode. @cindex @code{debug-info}, @code{mysql} option @item -T, --debug-info Print some debug information at exit. @cindex @code{tee}, @code{mysql} option @item --tee=... Append everything into outfile. See interactive help (\h) also. Does not work in batch mode. @cindex @code{user}, @code{mysql} option @item -u, --user=# User for login if not current user. @cindex @code{safe-updates}, @code{mysql} option @item -U, --safe-updates[=#], --i-am-a-dummy[=#] Only allow @code{UPDATE} and @code{DELETE} that uses keys. See below for more information about this option. You can reset this option if you have it in your @code{my.cnf} file by using @code{--safe-updates=0}. @cindex @code{verbose}, @code{mysql} option @item -v, --verbose More verbose output (-v -v -v gives the table output format). @cindex @code{version}, @code{mysql} option @item -V, --version Output version information and exit. @cindex @code{wait}, @code{mysql} option @item -w, --wait Wait and retry if connection is down instead of aborting. @end table You can also set the following variables with @code{-O} or @code{--set-variable}: @cindex timeout, @code{connect_timeout} variable @cindex @code{connect_timeout} variable @cindex @code{max_allowed_packet} @cindex @code{net_buffer_length} @cindex @code{select_limit} @cindex @code{max_join_size} @multitable @columnfractions .3 .2 .5 @item Variable Name @tab Default @tab Description @item connect_timeout @tab 0 @tab Number of seconds before timeout connection. @item max_allowed_packet @tab 16777216 @tab Max packetlength to send/receive from to server @item net_buffer_length @tab 16384 @tab Buffer for TCP/IP and socket communication @item select_limit @tab 1000 @tab Automatic limit for SELECT when using --i-am-a-dummy @item max_join_size @tab 1000000 @tab Automatic limit for rows in a join when using --i-am-a-dummy. @end multitable If you type 'help' on the command line, @code{mysql} will print out the commands that it supports: @cindex commands, list of @example mysql> help MySQL commands: help (\h) Display this text. ? (\h) Synonym for `help'. clear (\c) Clear command. connect (\r) Reconnect to the server. Optional arguments are db and host. edit (\e) Edit command with $EDITOR. ego (\G) Send command to mysql server, display result vertically. exit (\q) Exit mysql. Same as quit. go (\g) Send command to mysql server. nopager (\n) Disable pager, print to stdout. notee (\t) Don't write into outfile. pager (\P) Set PAGER [to_pager]. Print the query results via PAGER. print (\p) Print current command. quit (\q) Quit mysql. rehash (\#) Rebuild completion hash. source (\.) Execute a SQL script file. Takes a file name as an argument. status (\s) Get status information from the server. tee (\T) Set outfile [to_outfile]. Append everything into given outfile. use (\u) Use another database. Takes database name as argument. @end example From the above, pager only works in UNIX. @cindex status command The @code{status} command gives you some information about the connection and the server you are using. If you are running in the @code{--safe-updates} mode, @code{status} will also print the values for the @code{mysql} variables that affect your queries. @cindex @code{safe-mode} command A useful startup option for beginners (introduced in MySQL Version 3.23.11) is @code{--safe-updates} (or @code{--i-am-a-dummy} for users that has at some time done a @code{DELETE FROM table_name} but forgot the @code{WHERE} clause). When using this option, @code{mysql} sends the following command to the MySQL server when opening the connection: @example SET SQL_SAFE_UPDATES=1,SQL_SELECT_LIMIT=#select_limit#, SQL_MAX_JOIN_SIZE=#max_join_size#" @end example where @code{#select_limit#} and @code{#max_join_size#} are variables that can be set from the @code{mysql} command line. @xref{SET OPTION, @code{SET}}. The effect of the above is: @itemize @bullet @item You are not allowed to do an @code{UPDATE} or @code{DELETE} statement if you don't have a key constraint in the @code{WHERE} part. One can, however, force an @code{UPDATE/DELETE} by using @code{LIMIT}: @example UPDATE table_name SET not_key_column=# WHERE not_key_column=# LIMIT 1; @end example @item All big results are automatically limited to @code{#select_limit#} rows. @item @code{SELECT}'s that will probably need to examine more than @code{#max_join_size} row combinations will be aborted. @end itemize Some useful hints about the @code{mysql} client: Some data is much more readable when displayed vertically, instead of the usual horizontal box type output. For example longer text, which includes new lines, is often much easier to be read with vertical output. @example mysql> select * from mails where length(txt) < 300 limit 300,1\G *************************** 1. row *************************** msg_nro: 3068 date: 2000-03-01 23:29:50 time_zone: +0200 mail_from: Monty reply: monty@@no.spam.com mail_to: "Thimble Smith" <tim@@no.spam.com> sbj: UTF-8 txt: >>>>> "Thimble" == Thimble Smith writes: Thimble> Hi. I think this is a good idea. Is anyone familiar with UTF-8 Thimble> or Unicode? Otherwise I'll put this on my TODO list and see what Thimble> happens. Yes, please do that. Regards, Monty file: inbox-jani-1 hash: 190402944 1 row in set (0.09 sec) @end example @itemize @bullet @item For logging, you can use the @code{tee} option. The @code{tee} can be started with option @code{--tee=...}, or from the command line interactively with command @code{tee}. All the data displayed on the screen will also be appended into a given file. This can be very useful for debugging purposes also. The @code{tee} can be disabled from the command line with command @code{notee}. Executing @code{tee} again starts logging again. Without a parameter the previous file will be used. Note that @code{tee} will flush the results into the file after each command, just before the command line appears again waiting for the next command. @item Browsing, or searching the results in the interactive mode in UNIX less, more, or any other similar program, is now possible with option @code{--pager[=...]}. Without argument, @code{mysql} client will look for environment variable PAGER and set @code{pager} to that. @code{pager} can be started from the interactive command line with command @code{pager} and disabled with command @code{nopager}. The command takes an argument optionally and the @code{pager} will be set to that. Command @code{pager} can be called without an argument, but this requires that the option @code{--pager} was used, or the @code{pager} will default to stdout. @code{pager} works only in UNIX, since it uses the popen() function, which doesn't exist in Windows. In Windows, the @code{tee} option can be used instead, although it may not be as handy as @code{pager} can be in some situations. @item A few tips about @code{pager}: You can use it to write to a file: @example mysql> pager cat > /tmp/log.txt @end example and the results will only go to a file. You can also pass any options for the programs that you want to use with the @code{pager}: @example mysql> pager less -n -i -S @end example From the above do note the option '-S'. You may find it very useful when browsing the results; try the option with horizontal output (end commands with '\g', or ';') and with vertical output (end commands with '\G'). Sometimes a very wide result set is hard to be read from the screen, with option -S to less you can browse the results within the interactive less from left to right, preventing lines longer than your screen from being continued to the next line. This can make the result set much more readable. You can swith the mode between on and off within the interactive less with '-S'. See the 'h' for more help about less. @item Last (unless you already understood this from the above examples ;) you can combine very complex ways to handle the results, for example the following would send the results to two files in two different directories, on two different hard-disks mounted on /dr1 and /dr2, yet let the results still be seen on the screen via less: @example mysql> pager cat | tee /dr1/tmp/res.txt | tee /dr2/tmp/res2.txt | less -n -i -S @end example @item You can also combine the two functions above; have the @code{tee} enabled, @code{pager} set to 'less' and you will be able to browse the results in unix 'less' and still have everything appended into a file the same time. The difference between @code{UNIX tee} used with the @code{pager} and the @code{mysql} client in-built @code{tee}, is that the in-built @code{tee} works even if you don't have the @code{UNIX tee} available. The in-built @code{tee} also logs everything that is printed on the screen, where the @code{UNIX tee} used with @code{pager} doesn't log quite that much. Last, but not least, the interactive @code{tee} is more handy to switch on and off, when you want to log something into a file, but want to be able to turn the feature off sometimes. @end itemize @node mysqladmin, Using mysqlcheck, mysql, Client-Side Scripts @subsection mysqladmin, Administrating a MySQL Server @cindex administration, server @cindex server administration @cindex @code{mysladmn} A utility for performing administrative operations. The syntax is: @example shell> mysqladmin [OPTIONS] command [command-option] command ... @end example You can get a list of the options your version of @code{mysqladmin} supports by executing @code{mysqladmin --help}. The current @code{mysqladmin} supports the following commands: @table @code @item create databasename Create a new database. @item drop databasename Delete a database and all its tables. @item extended-status Gives an extended status message from the server. @item flush-hosts Flush all cached hosts. @item flush-logs Flush all logs. @item flush-tables Flush all tables. @item flush-privileges Reload grant tables (same as reload). @item kill id,id,... Kill mysql threads. @item password Set a new password. Change old password to new-password. @item ping Check if mysqld is alive. @item processlist Show list of active threads in server. @item reload Reload grant tables. @item refresh Flush all tables and close and open logfiles. @item shutdown Take server down. @item slave-start Start slave replication thread. @item slave-stop Stop slave replication thread. @item status Gives a short status message from the server. @item variables Prints variables available. @item version Get version info from server. @end table All commands can be shortened to their unique prefix. For example: @example shell> mysqladmin proc stat +----+-------+-----------+----+-------------+------+-------+------+ | Id | User | Host | db | Command | Time | State | Info | +----+-------+-----------+----+-------------+------+-------+------+ | 6 | monty | localhost | | Processlist | 0 | | | +----+-------+-----------+----+-------------+------+-------+------+ Uptime: 10077 Threads: 1 Questions: 9 Slow queries: 0 Opens: 6 Flush tables: 1 Open tables: 2 Memory in use: 1092K Max memory used: 1116K @end example @cindex status command, results The @code{mysqladmin status} command result has the following columns: @cindex uptime @multitable @columnfractions .3 .7 @item Uptime @tab Number of seconds the MySQL server has been up. @cindex threads @item Threads @tab Number of active threads (clients). @cindex questions @item Questions @tab Number of questions from clients since @code{mysqld} was started. @cindex slow queries @item Slow queries @tab Queries that have taken more than @code{long_query_time} seconds. @xref{Slow query log}. @cindex opens @item Opens @tab How many tables @code{mysqld} has opened. @cindex flush tables @cindex tables, flush @item Flush tables @tab Number of @code{flush ...}, @code{refresh}, and @code{reload} commands. @cindex open tables @item Open tables @tab Number of tables that are open now. @cindex memory use @item Memory in use @tab Memory allocated directly by the @code{mysqld} code (only available when MySQL is compiled with --with-debug=full). @cindex max memory used @item Max memory used @tab Maximum memory allocated directly by the @code{mysqld} code (only available when MySQL is compiled with --with-debug=full). @end multitable If you do @code{myslqadmin shutdown} on a socket (in other words, on a the computer where @code{mysqld} is running), @code{mysqladmin} will wait until the MySQL @code{pid-file} is removed to ensure that the @code{mysqld} server has stopped properly. @node Using mysqlcheck, mysqldump, mysqladmin, Client-Side Scripts @subsection Using @code{mysqlcheck} for Table Maintenance and Crash Recovery Since MySQL version 3.23.38 you will be able to use a new checking and repairing tool for @code{MyISAM} tables. The difference to @code{myisamchk} is that @code{mysqlcheck} should be used when the @code{mysqld} server is running, where as @code{myisamchk} should be used when it is not. The benefit is that you no longer have to take the server down for checking or repairing your tables. @code{mysqlcheck} uses MySQL server commands @code{CHECK}, @code{REPAIR}, @code{ANALYZE} and @code{OPTIMIZE} in a convenient way for the user. There are three alternative ways to invoke @code{mysqlcheck}: @example shell> mysqlcheck [OPTIONS] database [tables] shell> mysqlcheck [OPTIONS] --databases DB1 [DB2 DB3...] shell> mysqlcheck [OPTIONS] --all-databases @end example So it can be used in a similar way as @code{mysqldump} when it comes to what databases and tables you want to choose. @code{mysqlcheck} does have a special feature compared to the other clients; the default behavior, checking tables (-c), can be changed by renaming the binary. So if you want to have a tool that repairs tables by default, you should just copy @code{mysqlcheck} to your harddrive with a new name, @code{mysqlrepair}, or alternatively make a symbolic link to @code{mysqlrepair} and name the symbolic link as @code{mysqlrepair}. If you invoke @code{mysqlrepair} now, it will repair tables by default. The names that you can use to change @code{mysqlcheck} default behavior are here: @example mysqlrepair: The default option will be -r mysqlanalyze: The default option will be -a mysqloptimize: The default option will be -o @end example The options available for @code{mysqlcheck} are listed here, please check what your version supports with @code{mysqlcheck --help}. @table @code @item -A, --all-databases Check all the databases. This will be same as --databases with all databases selected @item -1, --all-in-1 Instead of making one query for each table, execute all queries in 1 query separately for each database. Table names will be in a comma separated list. @item -a, --analyze Analyze given tables. @item --auto-repair If a checked table is corrupted, automatically fix it. Repairing will be done after all tables have been checked, if corrupted ones were found. @item -#, --debug=... Output debug log. Often this is 'd:t:o,filename' @item --character-sets-dir=... Directory where character sets are @item -c, --check Check table for errors @item -C, --check-only-changed Check only tables that have changed since last check or haven't been closed properly. @item --compress Use compression in server/client protocol. @item -?, --help Display this help message and exit. @item -B, --databases To check several databases. Note the difference in usage; In this case no tables are given. All name arguments are regarded as database names. @item --default-character-set=... Set the default character set @item -F, --fast Check only tables that hasn't been closed properly @item -f, --force Continue even if we get an sql-error. @item -e, --extended If you are using this option with CHECK TABLE, it will ensure that the table is 100 percent consistent, but will take a long time. If you are using this option with REPAIR TABLE, it will run an extended repair on the table, which may not only take a long time to execute, but may produce a lot of garbage rows also! @item -h, --host=... Connect to host. @item -m, --medium-check Faster than extended-check, but only finds 99.99 percent of all errors. Should be good enough for most cases. @item -o, --optimize Optimize table @item -p, --password[=...] Password to use when connecting to server. If password is not given it's solicited on the tty. @item -P, --port=... Port number to use for connection. @item -q, --quick If you are using this option with CHECK TABLE, it prevents the check from scanning the rows to check for wrong links. This is the fastest check. If you are using this option with REPAIR TABLE, it will try to repair only the index tree. This is the fastest repair method for a table. @item -r, --repair Can fix almost anything except unique keys that aren't unique. @item -s, --silent Print only error messages. @item -S, --socket=... Socket file to use for connection. @item --tables Overrides option --databases (-B). @item -u, --user=# User for login if not current user. @item -v, --verbose Print info about the various stages. @item -V, --version Output version information and exit. @end table @node mysqldump, mysqlhotcopy, Using mysqlcheck, Client-Side Scripts @subsection mysqldump, Dumping Table Structure and Data @cindex dumping, databases @cindex databases, dumping @cindex tables, dumping @cindex backing up, databases @cindex @code{mysqldump} Utility to dump a database or a collection of database for backup or for transferring the data to another SQL server (not necessarily a MySQL server). The dump will contain SQL statements to create the table and/or populate the table. If you are doing a backup on the server, you should consider using the @code{mysqlhotcopy} instead. @xref{mysqlhotcopy, , @code{mysqlhotcopy}}. @example shell> mysqldump [OPTIONS] database [tables] OR mysqldump [OPTIONS] --databases [OPTIONS] DB1 [DB2 DB3...] OR mysqldump [OPTIONS] --all-databases [OPTIONS] @end example If you don't give any tables or use the @code{--databases} or @code{--all-databases}, the whole database(s) will be dumped. You can get a list of the options your version of @code{mysqldump} supports by executing @code{mysqldump --help}. Note that if you run @code{mysqldump} without @code{--quick} or @code{--opt}, @code{mysqldump} will load the whole result set into memory before dumping the result. This will probably be a problem if you are dumping a big database. Note that if you are using a new copy of the @code{mysqldump} program and you are going to do a dump that will be read into a very old MySQL server, you should not use the @code{--opt} or @code{-e} options. @code{mysqldump} supports the following options: @table @code @item --add-locks Add @code{LOCK TABLES} before and @code{UNLOCK TABLE} after each table dump. (To get faster inserts into MySQL.) @item --add-drop-table Add a @code{drop table} before each create statement. @item -A, --all-databases Dump all the databases. This will be same as @code{--databases} with all databases selected. @item -a, --all Include all MySQL-specific create options. @item --allow-keywords Allow creation of column names that are keywords. This works by prefixing each column name with the table name. @item -c, --complete-insert Use complete insert statements (with column names). @item -C, --compress Compress all information between the client and the server if both support compression. @item -B, --databases To dump several databases. Note the difference in usage. In this case no tables are given. All name arguments are regarded as database names. @code{USE db_name;} will be included in the output before each new database. @item --delayed Insert rows with the @code{INSERT DELAYED} command. @item -e, --extended-insert Use the new multiline @code{INSERT} syntax. (Gives more compact and faster inserts statements.) @item -#, --debug[=option_string] Trace usage of the program (for debugging). @item --help Display a help message and exit. @item --fields-terminated-by=... @itemx --fields-enclosed-by=... @itemx --fields-optionally-enclosed-by=... @itemx --fields-escaped-by=... @itemx --lines-terminated-by=... These options are used with the @code{-T} option and have the same meaning as the corresponding clauses for @code{LOAD DATA INFILE}. @xref{LOAD DATA, , @code{LOAD DATA}}. @item -F, --flush-logs Flush log file in the MySQL server before starting the dump. @item -f, --force, Continue even if we get a SQL error during a table dump. @item -h, --host=.. Dump data from the MySQL server on the named host. The default host is @code{localhost}. @item -l, --lock-tables. Lock all tables before starting the dump. The tables are locked with @code{READ LOCAL} to allow concurrent inserts in the case of @code{MyISAM} tables. @item -n, --no-create-db 'CREATE DATABASE /*!32312 IF NOT EXISTS*/ db_name;' will not be put in the output. The above line will be added otherwise, if --databases or --all-databases option was given. @item -t, --no-create-info Don't write table creation information (The @code{CREATE TABLE} statement.) @item -d, --no-data Don't write any row information for the table. This is very useful if you just want to get a dump of the structure for a table! @item --opt Same as @code{--quick --add-drop-table --add-locks --extended-insert --lock-tables}. Should give you the fastest possible dump for reading into a MySQL server. @item -pyour_pass, --password[=your_pass] The password to use when connecting to the server. If you specify no @samp{=your_pass} part, @code{mysqldump} you will be prompted for a password. @item -P port_num, --port=port_num The TCP/IP port number to use for connecting to a host. (This is used for connections to hosts other than @code{localhost}, for which Unix sockets are used.) @item -q, --quick Don't buffer query, dump directly to stdout. Uses @code{mysql_use_result()} to do this. @item -r, --result-file=... Direct output to a given file. This option should be used in MSDOS, because it prevents new line '\n' from being converted to '\n\r' (new line + carriage return). @item -S /path/to/socket, --socket=/path/to/socket The socket file to use when connecting to @code{localhost} (which is the default host). @item --tables Overrides option --databases (-B). @item -T, --tab=path-to-some-directory Creates a @code{table_name.sql} file, that contains the SQL CREATE commands, and a @code{table_name.txt} file, that contains the data, for each give table. @strong{NOTE}: This only works if @code{mysqldump} is run on the same machine as the @code{mysqld} daemon. The format of the @code{.txt} file is made according to the @code{--fields-xxx} and @code{--lines--xxx} options. @item -u user_name, --user=user_name The MySQL user name to use when connecting to the server. The default value is your Unix login name. @item -O var=option, --set-variable var=option Set the value of a variable. The possible variables are listed below. @item -v, --verbose Verbose mode. Print out more information on what the program does. @item -V, --version Print version information and exit. @item -w, --where='where-condition' Dump only selected records. Note that QUOTES are mandatory: @example "--where=user='jimf'" "-wuserid>1" "-wuserid<1" @end example @item -O net_buffer_length=#, where # < 16M When creating multi-row-insert statements (as with option @code{--extended-insert} or @code{--opt}), @code{mysqldump} will create rows up to @code{net_buffer_length} length. If you increase this variable, you should also ensure that the @code{max_allowed_packet} variable in the MySQL server is bigger than the @code{net_buffer_length}. @end table The most normal use of @code{mysqldump} is probably for making a backup of whole databases. @xref{Backup}. @example mysqldump --opt database > backup-file.sql @end example You can read this back into MySQL with: @example mysql database < backup-file.sql @end example or @example mysql -e "source /patch-to-backup/backup-file.sql" database @end example However, it's also very useful to populate another MySQL server with information from a database: @example mysqldump --opt database | mysql ---host=remote-host -C database @end example It is possible to dump several databases with one command: @example mysqldump --databases database1 [database2 database3...] > my_databases.sql @end example If all the databases are wanted, one can use: @example mysqldump --all-databases > all_databases.sql @end example @node mysqlhotcopy, mysqlimport, mysqldump, Client-Side Scripts @subsection mysqlhotcopy, Copying MySQL Databases and Tables @cindex dumping, databases @cindex databases, dumping @cindex tables, dumping @cindex backing up, databases @code{mysqlhotcopy} is a perl script that uses @code{LOCK TABLES}, @code{FLUSH TABLES} and @code{cp} or @code{scp} to quickly make a backup of a database. It's the fastest way to make a backup of the database, of single tables but it can only be run on the same machine where the database directories are. @example mysqlhotcopy db_name [/path/to/new_directory] mysqlhotcopy db_name_1 ... db_name_n /path/to/new_directory mysqlhotcopy db_name./regex/ @end example @code{mysqlhotcopy} supports the following options: @table @code @item -?, --help Display a help screen and exit @item -u, --user=# User for database login @item -p, --password=# Password to use when connecting to server @item -P, --port=# Port to use when connecting to local server @item -S, --socket=# Socket to use when connecting to local server @item --allowold Don't abort if target already exists (rename it _old) @item --keepold Don't delete previous (now renamed) target when done @item --noindices Don't include full index files in copy to make the backup smaller and faster The indexes can later be reconstructed with @code{myisamchk -rq.}. @item --method=# Method for copy (@code{cp} or @code{scp}). @item -q, --quiet Be silent except for errors @item --debug Enable debug @item -n, --dryrun Report actions without doing them @item --regexp=# Copy all databases with names matching regexp @item --suffix=# Suffix for names of copied databases @item --checkpoint=# Insert checkpoint entry into specified db.table @item --flushlog Flush logs once all tables are locked. @item --tmpdir=# Temporary directory (instead of /tmp). @end table You can use @code{perldoc mysqlhotcopy} to get a more complete documentation for @code{mysqlhotcopy}. @code{mysqlhotcopy} reads the groups @code{[client]} and @code{[mysqlhotcopy]} from the option files. To be able to execute @code{mysqlhotcopy} you need write access to the backup directory, @code{SELECT} privilege to the tables you are about to copy and the MySQL @code{Reload} privilege (to be able to execute @code{FLUSH TABLES}). @node mysqlimport, mysqlshow, mysqlhotcopy, Client-Side Scripts @subsection mysqlimport, Importing Data from Text Files @cindex importing, data @cindex data, importing @cindex files, text @cindex text files, importing @cindex @code{mysqlimport} @code{mysqlimport} provides a command-line interface to the @code{LOAD DATA INFILE} SQL statement. Most options to @code{mysqlimport} correspond directly to the same options to @code{LOAD DATA INFILE}. @xref{LOAD DATA, , @code{LOAD DATA}}. @code{mysqlimport} is invoked like this: @example shell> mysqlimport [options] database textfile1 [textfile2....] @end example For each text file named on the command line, @code{mysqlimport} strips any extension from the filename and uses the result to determine which table to import the file's contents into. For example, files named @file{patient.txt}, @file{patient.text}, and @file{patient} would all be imported into a table named @code{patient}. @code{mysqlimport} supports the following options: @table @code @item -c, --columns=... This option takes a comma-separated list of field names as an argument. The field list is used to create a proper @code{LOAD DATA INFILE} command, which is then passed to MySQL. @xref{LOAD DATA, , @code{LOAD DATA}}. @item -C, --compress Compress all information between the client and the server if both support compression. @item -#, --debug[=option_string] Trace usage of the program (for debugging). @item -d, --delete Empty the table before importing the text file. @item --fields-terminated-by=... @itemx --fields-enclosed-by=... @itemx --fields-optionally-enclosed-by=... @itemx --fields-escaped-by=... @itemx --lines-terminated-by=... These options have the same meaning as the corresponding clauses for @code{LOAD DATA INFILE}. @xref{LOAD DATA, , @code{LOAD DATA}}. @item -f, --force Ignore errors. For example, if a table for a text file doesn't exist, continue processing any remaining files. Without @code{--force}, @code{mysqlimport} exits if a table doesn't exist. @item --help Display a help message and exit. @item -h host_name, --host=host_name Import data to the MySQL server on the named host. The default host is @code{localhost}. @item -i, --ignore See the description for the @code{--replace} option. @item -l, --lock-tables Lock @strong{ALL} tables for writing before processing any text files. This ensures that all tables are synchronized on the server. @item -L, --local Read input files from the client. By default, text files are assumed to be on the server if you connect to @code{localhost} (which is the default host). @item -pyour_pass, --password[=your_pass] The password to use when connecting to the server. If you specify no @samp{=your_pass} part, @code{mysqlimport} you will be prompted for a password. @item -P port_num, --port=port_num The TCP/IP port number to use for connecting to a host. (This is used for connections to hosts other than @code{localhost}, for which Unix sockets are used.) @item -r, --replace The @code{--replace} and @code{--ignore} options control handling of input records that duplicate existing records on unique key values. If you specify @code{--replace}, new rows replace existing rows that have the same unique key value. If you specify @code{--ignore}, input rows that duplicate an existing row on a unique key value are skipped. If you don't specify either option, an error occurs when a duplicate key value is found, and the rest of the text file is ignored. @item -s, --silent Silent mode. Write output only when errors occur. @item -S /path/to/socket, --socket=/path/to/socket The socket file to use when connecting to @code{localhost} (which is the default host). @item -u user_name, --user=user_name The MySQL user name to use when connecting to the server. The default value is your Unix login name. @item -v, --verbose Verbose mode. Print out more information what the program does. @item -V, --version Print version information and exit. @end table Here is a sample run using @code{mysqlimport}: @example $ mysql --version mysql Ver 9.33 Distrib 3.22.25, for pc-linux-gnu (i686) $ uname -a Linux xxx.com 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown $ mysql -e 'CREATE TABLE imptest(id INT, n VARCHAR(30))' test $ ed a 100 Max Sydow 101 Count Dracula . w imptest.txt 32 q $ od -c imptest.txt 0000000 1 0 0 \t M a x S y d o w \n 1 0 0000020 1 \t C o u n t D r a c u l a \n 0000040 $ mysqlimport --local test imptest.txt test.imptest: Records: 2 Deleted: 0 Skipped: 0 Warnings: 0 $ mysql -e 'SELECT * FROM imptest' test +------+---------------+ | id | n | +------+---------------+ | 100 | Max Sydow | | 101 | Count Dracula | +------+---------------+ @end example @node mysqlshow, perror, mysqlimport, Client-Side Scripts @subsection Showing Databases, Tables, and Columns @cindex databases, displaying @cindex displaying, database information @cindex tables, displaying @cindex columns, displaying @cindex showing, database information @code{mysqlshow} can be used to quickly look at which databases exist, their tables, and the table's columns. With the @code{mysql} program you can get the same information with the @code{SHOW} commands. @xref{SHOW}. @code{mysqlshow} is invoked like this: @example shell> mysqlshow [OPTIONS] [database [table [column]]] @end example @itemize @bullet @item If no database is given, all matching databases are shown. @item If no table is given, all matching tables in the database are shown. @item If no column is given, all matching columns and column types in the table are shown. @end itemize Note that in newer MySQL versions, you only see those database/tables/columns for which you have some privileges. If the last argument contains a shell or SQL wild-card (@code{*}, @code{?}, @code{%} or @code{_}) then only what's matched by the wild card is shown. This may cause some confusion when you try to display the columns for a table with a @code{_} as in this case @code{mysqlshow} only shows you the table names that match the pattern. This is easily fixed by adding an extra @code{%} last on the command line (as a separate argument). @node perror, Batch Commands, mysqlshow, Client-Side Scripts @subsection perror, Explaining Error Codes @cindex error messages, displaying @cindex perror @code{perror} can be used to print error message(s). @code{perror} can be invoked like this: @example shell> perror [OPTIONS] [ERRORCODE [ERRORCODE...]] For example: shell> perror 64 79 Error code 64: Machine is not on the network Error code 79: Can not access a needed shared library @end example @code{perror} can be used to display a description for a system error code, or an MyISAM/ISAM table handler error code. The error messages are mostly system dependent. @node Batch Commands, , perror, Client-Side Scripts @subsection How to Run SQL Commands from a Text File @c FIX add the 'source' command The @code{mysql} client typically is used interactively, like this: @example shell> mysql database @end example However, it's also possible to put your SQL commands in a file and tell @code{mysql} to read its input from that file. To do so, create a text file @file{text_file} that contains the commands you wish to execute. Then invoke @code{mysql} as shown below: @example shell> mysql database < text_file @end example You can also start your text file with a @code{USE db_name} statement. In this case, it is unnecessary to specify the database name on the command line: @example shell> mysql < text_file @end example @xref{Client-Side Scripts}. @node Log Files, Replication, Client-Side Scripts, MySQL Database Administration @section The MySQL Log Files @cindex Log files MySQL has several different log files that can help you find out what's going on inside @code{mysqld}: @multitable @columnfractions .3 .7 @item The error log @tab Problems encountering starting, running or stopping @code{mysqld}. @item The isam log @tab Logs all changes to the ISAM tables. Used only for debugging the isam code. @item The query log @tab Established connections and executed queries. @item The update log @tab Deprecated: Stores all statements that changes data @item The binary log @tab Stores all statements that changes something. Used also for replication @item The slow log @tab Stores all queries that took more than @code{long_query_time} to execute or didn't use indexes. @end multitable All logs can be found in the @code{mysqld} data directory. You can force @code{mysqld} to reopen the log files (or in some cases switch to a new log) by executing @code{FLUSH LOGS}. @xref{FLUSH}. @menu * Error log:: The Error Log * Query log:: The General Query Log * Update log:: The Update Log * Binary log:: The Binary Update Log * Slow query log:: The Slow Query Log * Log file maintenance:: Log File Maintenance @end menu @node Error log, Query log, Log Files, Log Files @subsection The Error Log @code{mysqld} writes all errors to the stderr, which the @code{safe_mysqld} script redirects to a file called @code{'hostname'.err}. (On Windows, @code{mysqld} writes this directly to @file{\mysql\data\mysql.err}). This contains information indicating when @code{mysqld} was started and stopped and also any critical errors found when running. If @code{mysqld} dies unexpectedly and @code{safe_mysqld} needs to restart @code{mysqld}, @code{safe_mysqld} will write a @code{restarted mysqld} row in this file. This log also holds a warning if @code{mysqld} notices a table that needs to be automatically checked or repaired. On some operating systems, the error log will contain a stack trace for where @code{mysqld} died. This can be used to find out where @code{mysqld} died. @xref{Using stack trace}. @node Query log, Update log, Error log, Log Files @subsection The General Query Log @cindex query log @cindex files, query log If you want to know what happens within @code{mysqld}, you should start it with @code{--log[=file]}. This will log all connections and queries to the log file (by default named @file{'hostname'.log}). This log can be very useful when you suspect an error in a client and want to know exactly what @code{mysqld} thought the client sent to it. By default, the @code{mysql.server} script starts the MySQL server with the @code{-l} option. If you need better performance when you start using MySQL in a production environment, you can remove the @code{-l} option from @code{mysql.server} or change it to @code{--log-bin}. The entries in this log are written as @code{mysqld} receives the questions. This may be different than the order in which the statements are executed. This is in contrast to the update log and the binary log which are written after the query is executed, but before any locks are released. @node Update log, Binary log, Query log, Log Files @subsection The Update Log @cindex update log @cindex files, update log @strong{NOTE}: The update log is replaced by the binary log. @xref{Binary log}. With this you can do anything that you can do with the update log. When started with the @code{--log-update[=file_name]} option, @code{mysqld} writes a log file containing all SQL commands that update data. If no filename is given, it defaults to the name of the host machine. If a filename is given, but it doesn't contain a path, the file is written in the data directory. If @file{file_name} doesn't have an extension, @code{mysqld} will create log file names like so: @file{file_name.###}, where @code{###} is a number that is incremented each time you execute @code{mysqladmin refresh}, execute @code{mysqladmin flush-logs}, execute the @code{FLUSH LOGS} statement, or restart the server. @strong{NOTE:} For the above scheme to work, you should NOT create your own files with the same filename as the update log + some extensions that may be regarded as a number, in the directory used by the update log! If you use the @code{--log} or @code{-l} options, @code{mysqld} writes a general log with a filename of @file{hostname.log}, and restarts and refreshes do not cause a new log file to be generated (although it is closed and reopened). In this case you can copy it (on Unix) by doing: @example mv hostname.log hostname-old.log mysqladmin flush-logs cp hostname-old.log to-backup-directory rm hostname-old.log @end example Update logging is smart because it logs only statements that really update data. So an @code{UPDATE} or a @code{DELETE} with a @code{WHERE} that finds no rows is not written to the log. It even skips @code{UPDATE} statements that set a column to the value it already has. The update logging is done immediately after a query completes but before any locks are released or any commit is done. This ensures that the log will be logged in the execution order. If you want to update a database from update log files, you could do the following (assuming your update logs have names of the form @file{file_name.###}): @example shell> ls -1 -t -r file_name.[0-9]* | xargs cat | mysql @end example @code{ls} is used to get all the log files in the right order. This can be useful if you have to revert to backup files after a crash and you want to redo the updates that occurred between the time of the backup and the crash. @node Binary log, Slow query log, Update log, Log Files @subsection The Binary Update Log @cindex binary log @cindex files, binary log The intention is that the binary log should replace the update log, so we recommend you to switch to this log format as soon as possible! The binary log contains all information that is available in the update log in a more efficient format. It also contains information about how long every query that updated the database took. The binary log is also used when you are replicating a slave from a master. @xref{Replication}. When started with the @code{--log-bin[=file_name]} option, @code{mysqld} writes a log file containing all SQL commands that update data. If no file name is given, it defaults to the name of the host machine followed by @code{-bin}. If file name is given, but it doesn't contain a path, the file is written in the data directory. If you supply an extension to @code{--log-bin=filename.extension}, the extension will be silenty removed. To the binary log filename @code{mysqld} will append an extension that is a number that is incremented each time you execute @code{mysqladmin refresh}, execute @code{mysqladmin flush-logs}, execute the @code{FLUSH LOGS} statement or restart the server. You can use the following options to @code{mysqld} to affect what is logged to the binary log: @multitable @columnfractions .4 .6 @item @code{binlog-do-db=database_name} @tab Tells the master it should log updates for the specified database, and exclude all others not explicitly mentioned. (Example: @code{binlog-do-db=some_database}) @item @code{binlog-ignore-db=database_name} @tab Tells the master that updates to the given database should not be logged to the binary log (Example: @code{binlog-ignore-db=some_database}) @end multitable To be able to know which different binary log files have been used, @code{mysqld} will also create a binary log index file that contains the name of all used binary log files. By default this has the same name as the binary log file, with the extension @code{'.index'}. You can change the name of the binary log index file with the @code{--log-bin-index=[filename]} option. If you are using replication, you should not delete old binary log files until you are sure that no slave will ever need to use them. One way to do this is to do @code{mysqladmin flush-logs} once a day and then remove any logs that are more than 3 days old. You can examine the binary log file with the @code{mysqlbinlog} command. For example, you can update a MySQL server from the binary log as follows: @example mysqlbinlog log-file | mysql -h server_name @end example You can also use the @code{mysqlbinlog} program to read the binary log directly from a remote MySQL server! @code{mysqlbinlog --help} will give you more information of how to use this program! If you are using @code{BEGIN [WORK]} or @code{SET AUTOCOMMIT=0}, you must use the MySQL binary log for backups instead of the old update log. The binary logging is done immediately after a query completes but before any locks are released or any commit is done. This ensures that the log will be logged in the execution order. All updates (@code{UPDATE}, @code{DELETE} or @code{INSERT}) that change a transactional table (like BDB tables) are cached until a @code{COMMIT}. Any updates to a non-transactional table are stored in the binary log at once. Every thread will, on start, allocate a buffer of @code{binlog_cache_size} to buffer queries. If a query is bigger than this, the thread will open a temporary file to handle the bigger cache. The temporary file will be deleted when the thread ends. The @code{max_binlog_cache_size} can be used to restrict the total size used to cache a multi-transaction query. If you are using the update or binary log, concurrent inserts will not work together with @code{CREATE ... INSERT} and @code{INSERT ... SELECT}. This is to ensure that you can recreate an exact copy of your tables by applying the log on a backup. @node Slow query log, Log file maintenance, Binary log, Log Files @subsection The Slow Query Log @cindex slow query log @cindex files, slow query log When started with the @code{--log-slow-queries[=file_name]} option, @code{mysqld} writes a log file containing all SQL commands that took more than @code{long_query_time} to execute. The time to get the initial table locks are not counted as execution time. The slow query log is logged after the query is executed and after all locks has been released. This may be different than the order in which the statements are executed. If no file name is given, it defaults to the name of the host machine suffixed with @code{-slow.log}. If a filename is given, but doesn't contain a path, the file is written in the data directory. The slow query log can be used to find queries that take a long time to execute and are thus candidates for optimization. With a large log, that can become a difficult task. You can pipe the slow query log through the @code{mysqldumpslow} command to get a summary of the queries which appear in the log. You are using @code{--log-long-format} then also queries that are not using indexes are printed. @xref{Command-line options}. @node Log file maintenance, , Slow query log, Log Files @subsection Log File Maintenance @cindex files, log @cindex maintaining, log files @cindex log files, maintaining MySQL has a lot of log files which make it easy to see what is going. @xref{Log Files}. One must however from time to time clean up after @code{MysQL} to ensure that the logs don't take up too much disk space. When using MySQL with log files, you will, from time to time, want to remove/backup old log files and tell MySQL to start logging on new files. @xref{Backup}. On a Linux (@code{Redhat}) installation, you can use the @code{mysql-log-rotate} script for this. If you installed MySQL from an RPM distribution, the script should have been installed automatically. Note that you should be careful with this if you are using the log for replication! On other systems you must install a short script yourself that you start from @code{cron} to handle log files. You can force MySQL to start using new log files by using @code{mysqladmin flush-logs} or by using the SQL command @code{FLUSH LOGS}. If you are using MySQL Version 3.21 you must use @code{mysqladmin refresh}. The above command does the following: @itemize @bullet @item If standard logging (@code{--log}) or slow query logging (@code{--log-slow-queries}) is used, closes and reopens the log file. (@file{mysql.log} and @file{`hostname`-slow.log} as default). @item If update logging (@code{--log-update}) is used, closes the update log and opens a new log file with a higher sequence number. @end itemize If you are using only an update log, you only have to flush the logs and then move away the old update log files to a backup. If you are using the normal logging, you can do something like: @example shell> cd mysql-data-directory shell> mv mysql.log mysql.old shell> mysqladmin flush-logs @end example and then take a backup and remove @file{mysql.old}. @node Replication, , Log Files, MySQL Database Administration @section Replication in MySQL @cindex replication @cindex increasing, speed @cindex speed, increasing @cindex databases, replicating @menu * Replication Intro:: Introduction * Replication Implementation:: Replication Implementation Overview * Replication HOWTO:: How To Set Up Replication * Replication Features:: Replication Features and Known Problems * Replication Options:: Replication Options in my.cnf * Replication SQL:: SQL Commands Related to Replication * Replication FAQ:: Replication FAQ * Replication Problems:: Troubleshooting Replication @end menu This chapter describes the various replication features in MySQL. It serves as a reference to the options available with replication. You will be introduced to replication and learn how to implement it. Towards the end, there are some frequently asked questions and descriptions of problems and how to solve them. @node Replication Intro, Replication Implementation, Replication, Replication @subsection Introduction One way replication can be used is to increase both robustness and speed. For robustness you can have two systems and can switch to the backup if you have problems with the master. The extra speed is achieved by sending a part of the non-updating queries to the replica server. Of course this only works if non-updating queries dominate, but that is the normal case. Starting in Version 3.23.15, MySQL supports one-way replication internally. One server acts as the master, while the other acts as the slave. Note that one server could play the roles of master in one pair and slave in the other. The master server keeps a binary log of updates (@xref{Binary log}.) and an index file to binary logs to keep track of log rotation. The slave, upon connecting, informs the master where it left off since the last successfully propagated update, catches up on the updates, and then blocks and waits for the master to notify it of the new updates. Note that if you are replicating a database, all updates to this database should be done through the master! Another benefit of using replication is that one can get live backups of the system by doing a backup on a slave instead of doing it on the master. @xref{Backup}. @node Replication Implementation, Replication HOWTO, Replication Intro, Replication @subsection Replication Implementation Overview @cindex master-slave setup MySQL replication is based on the server keeping track of all changes to your database (updates, deletes, etc) in the binary log. (@xref{Binary log}.) and the slave server(s) reading the saved queries from the master server's binary log so that the slave can execute the same queries on its copy of the data. It is @strong{very important} to realize that the binary log is simply a record starting from a fixed point in time (the moment you enable binary logging). Any slaves which you set up will need copies of all the data from your master as it existed the moment that you enabled binary logging on the master. If you start your slaves with data that doesn't agree with what was on the master @strong{when the binary log was started}, your slaves may fail. A future version (4.0) of MySQL will remove the need to keep a (possibly large) snapshot of data for new slaves that you might wish to set up through the live backup functionality with no locking required. However, at this time, it is necessary to block all writes either with a global read lock or by shutting down the master while taking a snapshot. Once a slave is properly configured and running, it will simply connect to the master and wait for updates to process. If the master goes away or the slave loses connectivity with your master, it will keep trying to connect every @code{master-connect-retry} seconds until it is able to reconnect and resume listening for updates. Each slave keeps track of where it left off. The master server has no knowledge of how many slaves there are or which ones are up-to-date at any given time. The next section explains the master/slave setup process in more detail. @node Replication HOWTO, Replication Features, Replication Implementation, Replication @subsection How To Set Up Replication Below is a quick description of how to set up complete replication on your current MySQL server. It assumes you want to replicate all your databases and have not configured replication before. You will need to shutdown your master server briefly to complete the steps outlined below. @enumerate @item Make sure you have a recent version of MySQL installed on the master and slave(s). Use Version 3.23.29 or higher. Previous releases used a different binary log format and had bugs which have been fixed in newer releases. Please, do not report bugs until you have verified that the problem is present in the latest release. @item Set up special a replication user on the master with the @code{FILE} privilege and permission to connect from all the slaves. If the user is only doing replication (which is recommended), you don't need to grant any additional privileges. For example, to create a user named @code{repl} which can access your master from any host, you might use this command: @example GRANT FILE ON *.* TO repl@@"%" IDENTIFIED BY '<password>'; @end example @item Shut down MySQL on the master. @example mysqladmin -u root -p<password> shutdown @end example @item Snapshot all the data on your master server. The easiest way to do this (on Unix) is to simply use @strong{tar} to produce an archive of your entire data directory. The exact data directory location depends on your installation. @example tar -cvf /tmp/mysql-snapshot.tar /path/to/data-dir @end example Windows users can use WinZip or similar software to create an archive of the data directory. @item In @code{my.cnf} on the master add @code{log-bin} and @code{server-id=unique number} to the @code{[mysqld]} section and restart it. It is very important that the id of the slave is different from the id of the master. Think of @code{server-id} as something similar to the IP address - it uniquely identifies the server instance in the community of replication partners. @example [mysqld] log-bin server-id=1 @end example @item Restart MySQL on the master. @item Add the following to @code{my.cnf} on the slave(s): @example master-host=<hostname of the master> master-user=<replication user name> master-password=<replication user password> master-port=<TCP/IP port for master> server-id=<some unique number between 2 and 2^32-1> @end example replacing the values in <> with what is relevant to your system. @code{server-id} must be different for each server participating in replication. If you don't specify a server-id, it will be set to 1 if you have not defined @code{master-host}, else it will be set to 2. Note that in the case of @code{server-id} omission the master will refuse connections from all slaves, and the slave will refuse to connect to a master. Thus, omitting @code{server-id} is only good for backup with a binary log. @item Copy the snapshot data into your data directory on your slave(s). Make sure that the privileges on the files and directories are correct. The user which MySQL runs as needs to be able to read and write to them, just as on the master. @item Restart the slave(s). @end enumerate After you have done the above, the slave(s) should connect to the master and catch up on any updates which happened since the snapshot was taken. If you have forgotten to set @code{server-id} for the slave you will get the following error in the error log file: @example Warning: one should set server_id to a non-0 value if master_host is set. The server will not act as a slave. @end example If you have forgot to do this for the master, the slaves will not be able to connect to the master. If a slave is not able to replicate for any reason, you will find error messages in the error log on the slave. Once a slave is replicating, you will find a file called @code{master.info} in the same directory as your error log. The @code{master.info} file is used by the slave to keep track of how much of the master's binary log is has processed. @strong{Do not} remove or edit the file, unless you really know what you are doing. Even in that case, it is preferred that you use @code{CHANGE MASTER TO} command. @menu * Replication Features:: Replication Features and Known Problems * Replication Options:: Replication Options in my.cnf * Replication SQL:: SQL Commands Related to Replication * Replication FAQ:: Replication FAQ * Replication Problems:: Troubleshooting Replication @end menu @node Replication Features, Replication Options, Replication HOWTO, Replication @subsection Replication Features and Known Problems @cindex options, replication @cindex @code{my.cnf} file @cindex files,@code{my.cnf} Below is an explanation of what is supported and what is not: @itemize @bullet @item Replication will be done correctly with @code{AUTO_INCREMENT}, @code{LAST_INSERT_ID}, and @code{TIMESTAMP} values. @item @code{RAND()} in updates does not replicate properly. Use @code{RAND(some_non_rand_expr)} if you are replicating updates with @code{RAND()}. You can, for example, use @code{UNIX_TIMESTAMP()} for the argument to @code{RAND()}. @item You have to use the same character set (@code{--default-character-set}) on the master and the slave. If not, you may get duplicate key errors on the slave, because a key that is regarded as unique on the master may not be that in the other character set. @item @code{LOAD DATA INFILE} will be handled properly as long as the file still resides on the master server at the time of update propagation. @code{LOAD LOCAL DATA INFILE} will be skipped. @item Update queries that use user variables are not replication-safe (yet). @item @code{FLUSH} commands are not stored in the binary log and are because of this not replicated to the slaves. This is not normally a problem as @code{FLUSH} doesn't change anything. This does however mean that if you update the @code{MySQL} privilege tables directly without using @code{GRANT} statement and you replicate the @code{MySQL} privilege database, you must do a @code{FLUSH PRIVILEGES} on your slaves to put the new privileges into effect. @item Temporary tables starting in 3.23.29 are replicated properly with the exception of the case when you shut down slave server ( not just slave thread), you have some temporary tables open, and the are used in subsequent updates. To deal with this problem, to shut down the slave, do @code{SLAVE STOP}, then check @code{Slave_open_temp_tables} variable to see if it is 0, then issue @code{mysqladmin shutdown}. If the number is not 0, restart the slave thread with @code{SLAVE START} and see if you have better luck next time. There will be a cleaner solution, but it has to wait until version 4.0. In earlier versions temporary tables are not being replicated properly - we recommend that you either upgrade, or execute @code{SET SQL_LOG_BIN=0} on your clients before all queries with temp tables. @item MySQL only supports one master and many slaves. We will in 4.x add a voting algorithm to automatically change master if something goes wrong with the current master. We will also introduce 'agent' processes to help doing load balancing by sending select queries to different slaves. @item Starting in Version 3.23.26, it is safe to connect servers in a circular master-slave relationship with @code{log-slave-updates} enabled. Note, however, that many queries will not work right in this kind of setup unless your client code is written to take care of the potential problems that can happen from updates that occur in different sequence on different servers. This means that you can do a setup like the following: @example A -> B -> C -> A @end example This setup will only works if you only do non conflicting updates between the tables. In other words, if you insert data in A and C, you should never insert a row in A that may have a conflicting key with a row insert in C. You should also not update the sam rows on two servers if the order in which the updates are applied matters. Note that the log format has changed in Version 3.23.26 so that pre-3.23.26 slaves will not be able to read it. @item If the query on the slave gets an error, the slave thread will terminate, and a message will appear in the @code{.err} file. You should then connect to the slave manually, fix the cause of the error (for example, non-existent table), and then run @code{SLAVE START} sql command (available starting in Version 3.23.16). In Version 3.23.15, you will have to restart the server. @item If connection to the master is lost, the slave will retry immediately, and then in case of failure every @code{master-connect-retry} (default 60) seconds. Because of this, it is safe to shut down the master, and then restart it after a while. The slave will also be able to deal with network connectivity outages. @item Shutting down the slave (cleanly) is also safe, as it keeps track of where it left off. Unclean shutdowns might produce problems, especially if disk cache was not synced before the system died. Your system fault tolerance will be greatly increased if you have a good UPS. @item If the master is listening on a non-standard port, you will also need to specify this with @code{master-port} parameter in @code{my.cnf} . @item In Version 3.23.15, all of the tables and databases will be replicated. Starting in Version 3.23.16, you can restrict replication to a set of databases with @code{replicate-do-db} directives in @code{my.cnf} or just exclude a set of databases with @code{replicate-ignore-db}. Note that up until Version 3.23.23, there was a bug that did not properly deal with @code{LOAD DATA INFILE} if you did it in a database that was excluded from replication. @item Starting in Version 3.23.16, @code{SET SQL_LOG_BIN = 0} will turn off replication (binary) logging on the master, and @code{SET SQL_LOG_BIN = 1} will turn in back on - you must have the process privilege to do this. @item Starting in Version 3.23.19, you can clean up stale replication leftovers when something goes wrong and you want a clean start with @code{FLUSH MASTER} and @code{FLUSH SLAVE} commands. In Version 3.23.26 we have renamed them to @code{RESET MASTER} and @code{RESET SLAVE} respectively to clarify what they do. The old @code{FLUSH} variants still work, though, for compatibility. @item Starting in Version 3.23.21, you can use @code{LOAD TABLE FROM MASTER} for network backup and to set up replication initially. We have recently received a number of bug reports concerning it that we are investigating, so we recommend that you use it only in testing until we make it more stable. @item Starting in Version 3.23.23, you can change masters and adjust log position with @code{CHANGE MASTER TO}. @item Starting in Version 3.23.23, you tell the master that updates in certain databases should not be logged to the binary log with @code{binlog-ignore-db}. @item Starting in Version 3.23.26, you can use @code{replicate-rewrite-db} to tell the slave to apply updates from one database on the master to the one with a different name on the slave. @item Starting in Version 3.23.28, you can use @code{PURGE MASTER LOGS TO 'log-name'} to get rid of old logs while the slave is running. @end itemize @node Replication Options, Replication SQL, Replication Features, Replication @subsection Replication Options in my.cnf If you are using replication, we recommend you to use MySQL Version 3.23.30 or later. Older versions work, but they do have some bugs and are missing some features. On both master and slave you need to use the @code{server-id} option. This sets an unique replication id. You should pick a unique value in the range between 1 to 2^32-1 for each master and slave. Example: @code{server-id=3} The following table has the options you can use for the @strong{MASTER}: @multitable @columnfractions .3 .7 @item @strong{Option} @tab @strong{Description} @item @code{log-bin=filename} @tab Write to a binary update log to the specified location. Note that if you give it a parameter with an extension (for example, @code{log-bin=/mysql/logs/replication.log} ) versions up to 3.23.24 will not work right during replication if you do @code{FLUSH LOGS} . The problem is fixed in Version 3.23.25. If you are using this kind of log name, @code{FLUSH LOGS} will be ignored on binlog. To clear the log, run @code{FLUSH MASTER}, and do not forget to run @code{FLUSH SLAVE} on all slaves. In Version 3.23.26 and in later versions you should use @code{RESET MASTER} and @code{RESET SLAVE} @item @code{log-bin-index=filename} @tab Because the user could issue the @code{FLUSH LOGS} command, we need to know which log is currently active and which ones have been rotated out and in what sequence. This information is stored in the binary log index file. The default is `hostname`.index. You can use this option if you want to be a rebel. Example: @code{log-bin-index=db.index}. @item @code{sql-bin-update-same} @tab If set, setting @code{SQL_LOG_BIN} to a value will automatically set @code{SQL_LOG_UPDATE} to the same value and vice versa. @item @code{binlog-do-db=database_name} @tab Tells the master that it should log updates to the binary log if the current database is 'database_name'. All others database are ignored. Note that if you use this you should ensure that you only do updates in the current database. Example: @code{binlog-do-db=some_database}. @item @code{binlog-ignore-db=database_name} @tab Tells the master that updates where the current database is 'database_name' should not be stored in the binary log. Note that if you use this you should ensure that you only do updates in the current database. Example: @code{binlog-ignore-db=some_database} @end multitable The following table has the options you can use for the @strong{SLAVE}: @multitable @columnfractions .3 .7 @item @strong{Option} @tab @strong{Description} @item @code{master-host=host} @tab Master hostname or IP address for replication. If not set, the slave thread will not be started. Example: @code{master-host=db-master.mycompany.com}. @item @code{master-user=username} @tab The user the slave thread will us for authentication when connecting to the master. The user must have @code{FILE} privilege. If the master user is not set, user @code{test} is assumed. Example: @code{master-user=scott}. @item @code{master-password=password} @tab The password the slave thread will authenticate with when connecting to the master. If not set, an empty password is assumed. Example: @code{master-password=tiger}. @item @code{master-port=portnumber} @tab The port the master is listening on. If not set, the compiled setting of @code{MYSQL_PORT} is assumed. If you have not tinkered with @code{configure} options, this should be 3306. Example: @code{master-port=3306}. @item @code{master-connect-retry=seconds} @tab The number of seconds the slave thread will sleep before retrying to connect to the master in case the master goes down or the connection is lost. Default is 60. Example: @code{master-connect-retry=60}. @item @code{master-info-file=filename} @tab The location of the file that remembers where we left off on the master during the replication process. The default is master.info in the data directory. Sasha: The only reason I see for ever changing the default is the desire to be rebelious. Example: @code{master-info-file=master.info}. @item @code{replicate-do-table=db_name.table_name} @tab Tells the slave thread to restrict replication to the specified table. To specify more than one table, use the directive multiple times, once for each table. This will work for cross-database updates, in contrast to @code{replicate-do-db}. Example: @code{replicate-do-table=some_db.some_table}. @item @code{replicate-ignore-table=db_name.table_name} @tab Tells the slave thread to not replicate to the specified table. To specify more than one table to ignore, use the directive multiple times, once for each table. This will work for cross-datbase updates, in contrast to @code{replicate-ignore-db}. Example: @code{replicate-ignore-table=db_name.some_table}. @item @code{replicate-wild-do-table=db_name.table_name} @tab Tells the slave thread to restrict replication to the tables that match the specified wildcard pattern. To specify more than one table, use the directive multiple times, once for each table. This will work for cross-database updates. Example: @code{replicate-wild-do-table=foo%.bar%} will replicate only updates to tables in all databases that start with foo and whose table names start with bar. @item @code{replicate-wild-ignore-table=db_name.table_name} @tab Tells the slave thread to not replicate to the tables that match the given wild card pattern. To specify more than one table to ignore, use the directive multiple times, once for each table. This will work for cross-database updates. Example: @code{replicate-wild-ignore-table=foo%.bar%} will not do updates to tables in databases that start with foo and whose table names start with bar. @item @code{replicate-ignore-db=database_name} @tab Tells the slave thread to not replicate to the specified database. To specify more than one database to ignore, use the directive multiple times, once for each database. This option will not work if you use cross database updates. If you need cross database updates to work, make sure you have 3.23.28 or later, and use @code{replicate-wild-ignore-table=db_name.%} Example: @code{replicate-ignore-db=some_db}. @item @code{replicate-do-db=database_name} @tab Tells the slave thread to restrict replication to the specified database. To specify more than one database, use the directive multiple times, once for each database. Note that this will only work if you do not use cross-database queries such as @code{UPDATE some_db.some_table SET foo='bar'} while having selected a different or no database. If you need cross database updates to work, make sure you have 3.23.28 or later, and use @code{replicate-wild-do-table=db_name.%} Example: @code{replicate-do-db=some_db}. @item @code{log-slave-updates} @tab Tells the slave to log the updates from the slave thread to the binary log. Off by default. You will need to turn it on if you plan to daisy-chain the slaves. @item @code{replicate-rewrite-db=from_name->to_name} @tab Updates to a database with a different name than the original Example: @code{replicate-rewrite-db=master_db_name->slave_db_name}. @item @code{slave-skip-errors=err_code1,err_code2,..} @tab Available only in 3.23.47 and later. Tells the slave thread to continue replication when a query returns an error from the provided list. Normally, replication will discontinue when an error is encountered giving the user a chance to resolve the inconsistency in the data manually. Do not use this option unless you fully understand why you are getting the errors. If there are no bugs in your replication setup and client programs, and no bugs in MySQL itself, you should never get an abort with error.Indiscriminate use of this option will result in slaves being hopelessly out of sync with the master and you having no idea how the problem happened. For error codes, you should use the numbers provided by the error message in your slave error log and in the output of @code{SHOW SLAVE STATUS}. Full list of error messages can be found in the source distribution in @code{Docs/mysqld_error.txt}. You can ( but should not) also use a very non-recommended value of @code{all} which will ignore all error messages and keep barging along regardless. Needless to say, if you use it, we make no promises regarding your data integrity. Please do not complain if your data on the slave is not anywhere close to what it is on the master in this case - you have been warned. Example: @code{slave-skip-errors=1062,1053} or @code{slave-skip-errors=all} @item @code{skip-slave-start} @tab Tells the slave server not to start the slave on the startup. The user can start it later with @code{SLAVE START}. @item @code{slave_read_timeout=#} @tab Number of seconds to wait for more data from the master before aborting the read. @end multitable @node Replication SQL, Replication FAQ, Replication Options, Replication @subsection SQL Commands Related to Replication @cindex SQL commands, replication @cindex commands, replication @cindex replication, commands Replication can be controlled through the SQL interface. Below is the summary of commands: @multitable @columnfractions .30 .70 @item @strong{Command} @tab @strong{Description} @item @code{SLAVE START} @tab Starts the slave thread. (Slave) @item @code{SLAVE STOP} @tab Stops the slave thread. (Slave) @item @code{SET SQL_LOG_BIN=0} @tab Disables update logging if the user has process privilege. Ignored otherwise. (Master) @item @code{SET SQL_LOG_BIN=1} @tab Re-enables update logging if the user has process privilege. Ignored otherwise. (Master) @item @code{SET SQL_SLAVE_SKIP_COUNTER=n} @tab Skip the next @code{n} events from the master. Only valid when the slave thread is not running, otherwise, gives an error. Useful for recovering from replication glitches. @item @code{RESET MASTER} @tab Deletes all binary logs listed in the index file, resetting the binlog index file to be empty. In pre-3.23.26 versions, @code{FLUSH MASTER} (Master) @item @code{RESET SLAVE} @tab Makes the slave forget its replication position in the master logs. In pre 3.23.26 versions the command was called @code{FLUSH SLAVE}(Slave) @item @code{LOAD TABLE tblname FROM MASTER} @tab Downloads a copy of the table from master to the slave. (Slave) @item @code{CHANGE MASTER TO master_def_list} @tab Changes the master parameters to the values specified in @code{master_def_list} and restarts the slave thread. @code{master_def_list} is a comma-separated list of @code{master_def} where @code{master_def} is one of the following: @code{MASTER_HOST}, @code{MASTER_USER}, @code{MASTER_PASSWORD}, @code{MASTER_PORT}, @code{MASTER_CONNECT_RETRY}, @code{MASTER_LOG_FILE}, @code{MASTER_LOG_POS}. Example: @example CHANGE MASTER TO MASTER_HOST='master2.mycompany.com', MASTER_USER='replication', MASTER_PASSWORD='bigs3cret', MASTER_PORT=3306, MASTER_LOG_FILE='master2-bin.001', MASTER_LOG_POS=4; @end example You only need to specify the values that need to be changed. The values that you omit will stay the same with the exception of when you change the host or the port. In that case, the slave will assume that since you are connecting to a different host or a different port, the master is different. Therefore, the old values of log and position are not applicable anymore, and will automatically be reset to an empty string and 0, respectively (the start values). Note that if you restart the slave, it will remember its last master. If this is not desirable, you should delete the @file{master.info} file before restarting, and the slave will read its master from @code{my.cnf} or the command line. (Slave) @item @code{SHOW MASTER STATUS} @tab Provides status information on the binlog of the master. (Master) @item @code{SHOW SLAVE STATUS} @tab Provides status information on essential parameters of the slave thread. (Slave) @item @code{SHOW MASTER LOGS} @tab Only available starting in Version 3.23.28. Lists the binary logs on the master. You should use this command prior to @code{PURGE MASTER LOGS TO} to find out how far you should go. @item @code{PURGE MASTER LOGS TO 'logname'} @tab Available starting in Version 3.23.28. Deletes all the replication logs that are listed in the log index as being prior to the specified log, and removed them from the log index, so that the given log now becomes first. Example: @example PURGE MASTER LOGS TO 'mysql-bin.010' @end example This command will do nothing and fail with an error if you have an active slave that is currently reading one of the logs you are trying to delete. However, if you have a dormant slave, and happen to purge one of the logs it wants to read, the slave will be unable to replicate once it comes up. The command is safe to run while slaves are replicating - you do not need to stop them. You must first check all the slaves with @code{SHOW SLAVE STATUS} to see which log they are on, then do a listing of the logs on the master with @code{SHOW MASTER LOGS}, find the earliest log among all the slaves (if all the slaves are up to date, this will be the last log on the list), backup all the logs you are about to delete (optional) and purge up to the target log. @end multitable @node Replication FAQ, Replication Problems, Replication SQL, Replication @subsection Replication FAQ @cindex @code{Binlog_Dump} @strong{Q}: Why do I sometimes see more than one @code{Binlog_Dump} thread on the master after I have restarted the slave? @strong{A}: @code{Binlog_Dump} is a continuous process that is handled by the server in the following way: @itemize @bullet @item Catch up on the updates. @item Once there are no more updates left, go into @code{pthread_cond_wait()}, from which we can be awakened either by an update or a kill. @item On wake up, check the reason. If we are not supposed to die, continue the @code{Binlog_dump} loop. @item If there is some fatal error, such as detecting a dead client, terminate the loop. @end itemize So if the slave thread stops on the slave, the corresponding @code{Binlog_Dump} thread on the master will not notice it until after at least one update to the master (or a kill), which is needed to wake it up from @code{pthread_cond_wait()}. In the meantime, the slave could have opened another connection, which resulted in another @code{Binlog_Dump} thread. The above problem should not be present in Version 3.23.26 and later versions. In Version 3.23.26 we added @code{server-id} to each replication server, and now all the old zombie threads are killed on the master when a new replication thread connects from the same slave @strong{Q}: How do I rotate replication logs? @strong{A}: In Version 3.23.28 you should use @code{PURGE MASTER LOGS TO} command after determining which logs can be deleted, and optionally backing them up first. In earlier versions the process is much more painful, and cannot be safely done without stopping all the slaves in the case that you plan to re-use log names. You will need to stop the slave threads, edit the binary log index file, delete all the old logs, restart the master, start slave threads, and then remove the old log files. @strong{Q}: How do I upgrade on a hot replication setup? @strong{A}: If you are upgrading pre-3.23.26 versions, you should just lock the master tables, let the slave catch up, then run @code{FLUSH MASTER} on the master, and @code{FLUSH SLAVE} on the slave to reset the logs, then restart new versions of the master and the slave. Note that the slave can stay down for some time - since the master is logging all the updates, the slave will be able to catch up once it is up and can connect. After 3.23.26, we have locked the replication protocol for modifications, so you can upgrade masters and slave on the fly to a newer 3.23 version and you can have different versions of MySQL running on the slave and the master, as long as they are both newer than 3.23.26. @cindex replication, two-way @strong{Q}: What issues should I be aware of when setting up two-way replication? @strong{A}: MySQL replication currently does not support any locking protocol between master and slave to guarantee the atomicity of a distributed (cross-server) update. In in other words, it is possible for client A to make an update to co-master 1, and in the meantime, before it propagates to co-master 2, client B could make an update to co-master 2 that will make the update of client A work differently than it did on co-master 1. Thus when the update of client A will make it to co-master 2, it will produce tables that will be different than what you have on co-master 1, even after all the updates from co-master 2 have also propagated. So you should not co-chain two servers in a two-way replication relationship, unless you are sure that you updates can safely happen in any order, or unless you take care of mis-ordered updates somehow in the client code. You must also realize that two-way replication actually does not improve performance very much, if at all, as far as updates are concerned. Both servers need to do the same amount of updates each, as you would have one server do. The only difference is that there will be a little less lock contention, because the updates originating on another server will be serialized in one slave thread. This benefit, though, might be offset by network delays. @cindex performance, improving @cindex increasing, performance @strong{Q}: How can I use replication to improve performance of my system? @strong{A}: You should set up one server as the master, and direct all writes to it, and configure as many slaves as you have the money and rackspace for, distributing the reads among the master and the slaves. You can also start the slaves with @code{--skip-bdb}, @code{--low-priority-updates} and @code{--delay-key-write-for-all-tables} to get speed improvements for the slave. In this case the slave will use non-transactional @code{MyISAM} tables instead of @code{BDB} tables to get more speed. @strong{Q}: What should I do to prepare my client code to use performance-enhancing replication? @strong{A}: If the part of your code that is responsible for database access has been properly abstracted/modularized, converting it to run with the replicated setup should be very smooth and easy - just change the implementation of your database access to read from some slave or the master, and to always write to the master. If your code does not have this level of abstraction, setting up a replicated system will give you an opportunity/motivation to it clean up. You should start by creating a wrapper library /module with the following functions: @itemize @bullet @item @code{safe_writer_connect()} @item @code{safe_reader_connect()} @item @code{safe_reader_query()} @item @code{safe_writer_query()} @end itemize @code{safe_} means that the function will take care of handling all the error conditions. You should then convert your client code to use the wrapper library. It may be a painful and scary process at first, but it will pay off in the long run. All applications that follow the above pattern will be able to take advantage of one-master/many slaves solution. The code will be a lot easier to maintain, and adding troubleshooting options will be trivial. You will just need to modify one or two functions, for example, to log how long each query took, or which query, among your many thousands, gave you an error. If you have written a lot of code already, you may want to automate the conversion task by using Monty's @code{replace} utility, which comes with the standard distribution of MySQL, or just write your own Perl script. Hopefully, your code follows some recognizable pattern. If not, then you are probably better off re-writing it anyway, or at least going through and manually beating it into a pattern. Note that, of course, you can use different names for the functions. What is important is having unified interface for connecting for reads, connecting for writes, doing a read, and doing a write. @strong{Q}: When and how much can MySQL replication improve the performance of my system? @strong{A}: MySQL replication is most beneficial for a system with frequent reads and not so frequent writes. In theory, by using a one master/many slaves setup you can scale by adding more slaves until you either run out of network bandwidth, or your update load grows to the point that the master cannot handle it. In order to determine how many slaves you can get before the added benefits begin to level out, and how much you can improve performance of your site, you need to know your query patterns, and empirically (by benchmarking) determine the relationship between the throughput on reads (reads per second, or @code{max_reads}) and on writes @code{max_writes}) on a typical master and a typical slave. The example below will show you a rather simplified calculation of what you can get with replication for our imagined system. Let's say our system load consists of 10% writes and 90% reads, and we have determined that @code{max_reads} = 1200 - 2 * @code{max_writes}, or in other words, our system can do 1200 reads per second with no writes, our average write is twice as slow as average read, and the relationship is linear. Let us suppose that our master and slave are of the same capacity, and we have N slaves and 1 master. Then we have for each server (master or slave): @code{reads = 1200 - 2 * writes} (from bencmarks) @code{reads = 9* writes / (N + 1) } (reads split, but writes go to all servers) @code{9*writes/(N+1) + 2 * writes = 1200} @code{writes = 1200/(2 + 9/(N+1)} So if N = 0, which means we have no replication, our system can handle 1200/11, about 109 writes per second (which means we will have 9 times as many reads due to the nature of our application). If N = 1, we can get up to 184 writes per second. If N = 8, we get up to 400. If N = 17, 480 writes. Eventually as N approaches infinity (and our budget negative infinity), we can get very close to 600 writes per second, increasing system throughput about 5.5 times. However, with only 8 servers, we increased it almost 4 times already. Note that our computations assumed infinite network bandwidth, and neglected several other factors that could turn out to be significant on your system. In many cases, you may not be able to make a computation similar to the one above that will accurately predict what will happen on your system if you add N replication slaves. However, answering the following questions should help you decided whether and how much, if at all, the replication will improve the performance of your system: @itemize @bullet @item What is the read/write ratio on your system? @item How much more write load can one server handle if you reduce the reads? @item How many slaves do you have bandwidth for on your network? @end itemize @strong{Q}: How can I use replication to provide redundancy/high availability? @strong{A}: With the currently available features, you would have to set up a master and a slave (or several slaves), and write a script that will monitor the master to see if it is up, and instruct your applications and the slaves of the master change in case of failure. Some suggestions: @itemize @bullet @item To tell a slave to change the master use the @code{CHANGE MASTER TO} command. @item A good way to keep your applications informed where the master is by having a dynamic DNS entry for the master. With @strong{bind} you can use @code{nsupdate} to dynamically update your DNS. @item You should run your slaves with the @code{log-bin} option and without @code{log-slave-updates}. This way the slave will be ready to become a master as soon as you issue @code{STOP SLAVE}; @code{RESET MASTER}, and @code{CHANGE MASTER TO} on the other slaves. It will also help you catch spurious updates that may happen because of misconfiguration of the slave (ideally, you want to configure access rights so that no client can update the slave, except for the slave thread) combined with the bugs in your client programs (they should never update the slave directly). @end itemize We are currently working on integrating an automatic master election system into MySQL, but until it is ready, you will have to create your own monitoring tools. @node Replication Problems, , Replication FAQ, Replication @subsection Troubleshooting Replication If you have followed the instructions, and your replication setup is not working, first eliminate the user error factor by checking the following: @itemize @bullet @item Is the master logging to the binary log? Check with @code{SHOW MASTER STATUS}. If it is, @code{Position} will be non-zero. If not, verify that you have given the master @code{log-bin} option and have set @code{server-id}. @item Is the slave running? Check with @code{SHOW SLAVE STATUS}. The answer is found in @code{Slave_running} column. If not, verify slave options and check the error log for messages. @item If the slave is running, did it establish connection with the master? Do @code{SHOW PROCESSLIST}, find the thread with @code{system user} value in @code{User} column and @code{none} in the @code{Host} column, and check the @code{State} column. If it says @code{connecting to master}, verify the privileges for the replication user on the master, master host name, your DNS setup, whether the master is actually running, whether it is reachable from the slave, and if all that seems ok, read the error logs. @item If the slave was running, but then stopped, look at SHOW SLAVE STATUS output and check the error logs. It usually happens when some query that succeeded on the master fails on the slave. This should never happen if you have taken a proper snapshot of the master, and never modify the data on the slave outside of the slave thread. If it does, it is a bug, read below on how to report it. @item If a query on that succeeded on the master refuses to run on the slave, and a full database resync ( the proper thing to do ) does not seem feasible, try the following: @itemize @minus @item First see if there is some stray record in the way. Understand how it got there, then delete it and run @code{SLAVE START} @item If the above does not work or does not apply, try to understand if it would be safe to make the update manually ( if needed) and then ignore the next query from the master. @item If you have decided you can skip the next query, do @code{SET SQL_SLAVE_SKIP_COUNTER=1; SLAVE START;} to skip a query that does not use auto_increment, or last_insert_id or @code{SET SQL_SLAVE_SKIP_COUNTER=2; SLAVE START;} otherwise. The reason auto_increment/last_insert_id queries are different is that they take two events in the binary log of the master. @item If you are sure the slave started out perfectly in sync with the master, and no one has updated the tables involved outside of slave thread, report the bug, so you will not have to do the above tricks again. @end itemize @item Make sure you are not running into an old bug by upgrading to the most recent version. @item If all else fails, read the error logs. If they are big, @code{grep -i slave /path/to/your-log.err} on the slave. There is no generic pattern to search for on the master, as the only errors it logs are general system errors - if it can, it will send the error to the slave when things go wrong. @end itemize When you have determined that there is no user error involved, and replication still either does not work at all or is unstable, it is time to start working on a bug report. We need to get as much info as possible from you to be able to track down the bug. Please do spend some time and effort preparing a good bug report. Ideally, we would like to have a test case in the format found in @code{mysql-test/t/rpl*} directory of the source tree. If you submit a test case like that, you can expect a patch within a day or two in most cases, although, of course, you mileage may vary depending on a number of factors. Second best option is a just program with easily configurable connection arguments for the master and the slave that will demonstrate the problem on our systems. You can write one in Perl or in C, depending on which language you know better. If you have one of the above ways to demonstrate the bug, use @code{mysqlbug} to prepare a bug report and send it to @email{bugs@@lists.mysql.com}. If you have a phantom - a problem that does occur but you cannot duplicate "at will": @itemize @bullet @item Verify that there is no user error involved. For example, if you update the slave outside of the slave thread, the data will be out of sync, and you can have unique key violations on updates, in which case the slave thread will stop and wait for you to clean up the tables manually to bring them in sync. @item Run slave with @code{log-slave-updates} and @code{log-bin} - this will keep a log of all updates on the slave. @item Save all evidence before resetting the replication. If we have no or only sketchy information, it would take us a while to track down the problem. The evidence you should collect is: @itemize @minus @item All binary logs on the master @item All binary log on the slave @item The output of @code{SHOW MASTER STATUS} on the master at the time you have discovered the problem @item The output of @code{SHOW SLAVE STATUS} on the master at the time you have discovered the problem @item Error logs on the master and on the slave @end itemize @item Use @code{mysqlbinlog} to examine the binary logs. The following should be helpful to find the trouble query, for example: @example mysqlbinlog -j pos_from_slave_status /path/to/log_from_slave_status | head @end example @end itemize Once you have collected the evidence on the phantom problem, try hard to isolate it into a separate test case first. Then report the problem to @email{bugs@@lists.mysql.com} with as much info as possible. @node MySQL Optimization, Reference, MySQL Database Administration, Top @chapter MySQL Optimization @menu * Optimize Overview:: Optimization Overview * Query Speed:: Optimizing @code{SELECT}s and Other Queries * Locking Issues:: Locking Issues * Optimizing Database Structure:: Optimizing Database Structure * Optimizing the Server:: Optimizing the MySQL Server * Disk issues:: Disk Issues @end menu Optimization is a complicated task because it ultimately requires understanding of the whole system. While it may be possible to do some local optimizations with small knowledge of your system or application, the more optimal you want your system to become the more you will have to know about it. This chapter will try to explain and give some examples of different ways to optimize MySQL. Remember, however, that there are always some (increasingly harder) additional ways to make the system even faster. @node Optimize Overview, Query Speed, MySQL Optimization, MySQL Optimization @section Optimization Overview The most important part for getting a system fast is of course the basic design. You also need to know what kinds of things your system will be doing, and what your bottlenecks are. The most common bottlenecks are: @itemize @bullet @item Disk seeks. It takes time for the disk to find a piece of data. With modern disks in 1999, the mean time for this is usually lower than 10ms, so we can in theory do about 1000 seeks a second. This time improves slowly with new disks and is very hard to optimize for a single table. The way to optimize this is to spread the data on more than one disk. @item Disk reading/writing. When the disk is at the correct position we need to read the data. With modern disks in 1999, one disk delivers something like 10-20Mb/s. This is easier to optimize than seeks because you can read in parallel from multiple disks. @item CPU cycles. When we have the data in main memory (or if it already were there) we need to process it to get to our result. Having small tables compared to the memory is the most common limiting factor. But then, with small tables speed is usually not the problem. @item Memory bandwidth. When the CPU needs more data than can fit in the CPU cache the main memory bandwidth becomes a bottleneck. This is an uncommon bottleneck for most systems, but one should be aware of it. @end itemize @menu * Design Limitations:: MySQL Design Limitations/Tradeoffs * Portability:: Portability * Internal use:: What Have We Used MySQL For? * MySQL Benchmarks:: The MySQL Benchmark Suite * Custom Benchmarks:: Using Your Own Benchmarks @end menu @node Design Limitations, Portability, Optimize Overview, Optimize Overview @subsection MySQL Design Limitations/Tradeoffs @cindex design, limitations @cindex limitations, design Because MySQL uses extremely fast table locking (multiple readers / single writers) the biggest remaining problem is a mix of a steady stream of inserts and slow selects on the same table. We believe that for a huge number of systems the extremely fast performance in other cases make this choice a win. This case is usually also possible to solve by having multiple copies of the table, but it takes more effort and hardware. We are also working on some extensions to solve this problem for some common application niches. @node Portability, Internal use, Design Limitations, Optimize Overview @subsection Portability @cindex portability @cindex crash-me program @cindex programs, crash-me Because all SQL servers implement different parts of SQL, it takes work to write portable SQL applications. For very simple selects/inserts it is very easy, but the more you need the harder it gets. If you want an application that is fast with many databases it becomes even harder! To make a complex application portable you need to choose a number of SQL servers that it should work with. You can use the MySQL crash-me program/web-page @uref{http://www.mysql.com/information/crash-me.php} to find functions, types, and limits you can use with a selection of database servers. Crash-me now tests far from everything possible, but it is still comprehensive with about 450 things tested. For example, you shouldn't have column names longer than 18 characters if you want to be able to use Informix or DB2. Both the MySQL benchmarks and crash-me programs are very database-independent. By taking a look at how we have handled this, you can get a feeling for what you have to do to write your application database-independent. The benchmarks themselves can be found in the @file{sql-bench} directory in the MySQL source distribution. They are written in Perl with DBI database interface (which solves the access part of the problem). See @uref{http://www.mysql.com/information/benchmarks.html} for the results from this benchmark. As you can see in these results, all databases have some weak points. That is, they have different design compromises that lead to different behavior. If you strive for database independence, you need to get a good feeling for each SQL server's bottlenecks. MySQL is VERY fast in retrieving and updating things, but will have a problem in mixing slow readers/writers on the same table. Oracle, on the other hand, has a big problem when you try to access rows that you have recently updated (until they are flushed to disk). Transaction databases in general are not very good at generating summary tables from log tables, as in this case row locking is almost useless. To get your application @emph{really} database-independent, you need to define an easy extendable interface through which you manipulate your data. As C++ is available on most systems, it makes sense to use a C++ classes interface to the databases. If you use some specific feature for some database (like the @code{REPLACE} command in MySQL), you should code a method for the other SQL servers to implement the same feature (but slower). With MySQL you can use the @code{/*! */} syntax to add MySQL-specific keywords to a query. The code inside @code{/**/} will be treated as a comment (ignored) by most other SQL servers. If REAL high performance is more important than exactness, as in some Web applications, a possibility is to create an application layer that caches all results to give you even higher performance. By letting old results 'expire' after a while, you can keep the cache reasonably fresh. This is quite nice in case of extremely high load, in which case you can dynamically increase the cache and set the expire timeout higher until things get back to normal. In this case the table creation information should contain information of the initial size of the cache and how often the table should normally be refreshed. @node Internal use, MySQL Benchmarks, Portability, Optimize Overview @subsection What Have We Used MySQL For? @cindex uses, of MySQL @cindex customers, of MySQL During MySQL initial development, the features of MySQL were made to fit our largest customer. They handle data warehousing for a couple of the biggest retailers in Sweden. From all stores, we get weekly summaries of all bonus card transactions, and we are expected to provide useful information for the store owners to help them find how their advertisement campaigns are affecting their customers. The data is quite huge (about 7 million summary transactions per month), and we have data for 4-10 years that we need to present to the users. We got weekly requests from the customers that they want to get 'instant' access to new reports from this data. We solved this by storing all information per month in compressed 'transaction' tables. We have a set of simple macros (script) that generates summary tables grouped by different criteria (product group, customer id, store ...) from the transaction tables. The reports are Web pages that are dynamically generated by a small Perl script that parses a Web page, executes the SQL statements in it, and inserts the results. We would have used PHP or mod_perl instead but they were not available at that time. For graphical data we wrote a simple tool in @code{C} that can produce GIFs based on the result of a SQL query (with some processing of the result). This is also dynamically executed from the Perl script that parses the @code{HTML} files. In most cases a new report can simply be done by copying an existing script and modifying the SQL query in it. In some cases, we will need to add more fields to an existing summary table or generate a new one, but this is also quite simple, as we keep all transactions tables on disk. (Currently we have at least 50G of transactions tables and 200G of other customer data.) We also let our customers access the summary tables directly with ODBC so that the advanced users can themselves experiment with the data. We haven't had any problems handling this with quite modest Sun Ultra SPARCstation (2x200 Mhz). We recently upgraded one of our servers to a 2 CPU 400 Mhz UltraSPARC, and we are now planning to start handling transactions on the product level, which would mean a ten-fold increase of data. We think we can keep up with this by just adding more disk to our systems. We are also experimenting with Intel-Linux to be able to get more CPU power cheaper. Now that we have the binary portable database format (new in Version 3.23), we will start to use this for some parts of the application. Our initial feelings are that Linux will perform much better on low-to-medium load and Solaris will perform better when you start to get a high load because of extreme disk IO, but we don't yet have anything conclusive about this. After some discussion with a Linux Kernel developer, this might be a side effect of Linux giving so much resources to the batch job that the interactive performance gets very low. This makes the machine feel very slow and unresponsive while big batches are going. Hopefully this will be better handled in future Linux Kernels. @node MySQL Benchmarks, Custom Benchmarks, Internal use, Optimize Overview @subsection The MySQL Benchmark Suite @cindex benchmark suite @cindex crash-me program This should contain a technical description of the MySQL benchmark suite (and @code{crash-me}), but that description is not written yet. Currently, you can get a good idea of the benchmark by looking at the code and results in the @file{sql-bench} directory in any MySQL source distributions. This benchmark suite is meant to be a benchmark that will tell any user what things a given SQL implementation performs well or poorly at. Note that this benchmark is single threaded, so it measures the minimum time for the operations. We plan to in the future add a lot of multi-threaded tests to the benchmark suite. For example, (run on the same NT 4.0 machine): @multitable @columnfractions .6 .2 .2 @strong{Reading 2000000 rows by index} @tab @strong{Seconds} @tab @strong{Seconds} @item mysql @tab 367 @tab 249 @item mysql_odbc @tab 464 @item db2_odbc @tab 1206 @item informix_odbc @tab 121126 @item ms-sql_odbc @tab 1634 @item oracle_odbc @tab 20800 @item solid_odbc @tab 877 @item sybase_odbc @tab 17614 @end multitable @multitable @columnfractions .6 .2 .2 @strong{Inserting (350768) rows} @tab @strong{Seconds} @tab @strong{Seconds} @item mysql @tab 381 @tab 206 @item mysql_odbc @tab 619 @item db2_odbc @tab 3460 @item informix_odbc @tab 2692 @item ms-sql_odbc @tab 4012 @item oracle_odbc @tab 11291 @item solid_odbc @tab 1801 @item sybase_odbc @tab 4802 @end multitable In the above test MySQL was run with a 8M index cache. We have gather some more benchmark results at @uref{http://www.mysql.com/information/benchmarks.html}. Note that Oracle is not included because they asked to be removed. All Oracle benchmarks have to be passed by Oracle! We believe that makes Oracle benchmarks @strong{VERY} biased because the above benchmarks are supposed to show what a standard installation can do for a single client. To run the benchmark suite, you have to download a MySQL source distribution, install the perl DBI driver, the perl DBD driver for the database you want to test and then do: @example cd sql-bench perl run-all-tests --server=# @end example where # is one of supported servers. You can get a list of all options and supported servers by doing @code{run-all-tests --help}. @cindex crash-me @code{crash-me} tries to determine what features a database supports and what its capabilities and limitations are by actually running queries. For example, it determines: @itemize @bullet @item What column types are supported @item How many indexes are supported @item What functions are supported @item How big a query can be @item How big a @code{VARCHAR} column can be @end itemize We can find the result from crash-me on a lot of different databases at @uref{http://www.mysql.com/information/crash-me.php}. @node Custom Benchmarks, , MySQL Benchmarks, Optimize Overview @subsection Using Your Own Benchmarks @cindex benchmarks @cindex performance, benchmarks You should definitely benchmark your application and database to find out where the bottlenecks are. By fixing it (or by replacing the bottleneck with a 'dummy module') you can then easily identify the next bottleneck (and so on). Even if the overall performance for your application is sufficient, you should at least make a plan for each bottleneck, and decide how to solve it if someday you really need the extra performance. For an example of portable benchmark programs, look at the MySQL benchmark suite. @xref{MySQL Benchmarks, , MySQL Benchmarks}. You can take any program from this suite and modify it for your needs. By doing this, you can try different solutions to your problem and test which is really the fastest solution for you. It is very common that some problems only occur when the system is very heavily loaded. We have had many customers who contact us when they have a (tested) system in production and have encountered load problems. In every one of these cases so far, it has been problems with basic design (table scans are NOT good at high load) or OS/Library issues. Most of this would be a @strong{LOT} easier to fix if the systems were not already in production. To avoid problems like this, you should put some effort into benchmarking your whole application under the worst possible load! You can use Super Smack for this, and it is available at: @uref{http://www.mysql.com/Downloads/super-smack/super-smack-1.0.tar.gz}. As the name suggests, it can bring your system down to its knees if you ask it, so make sure to use it only on your development systems. @node Query Speed, Locking Issues, Optimize Overview, MySQL Optimization @section Optimizing @code{SELECT}s and Other Queries @cindex queries, speed of @cindex permission checks, effect on speed @cindex speed, of queries First, one thing that affects all queries: The more complex permission system setup you have, the more overhead you get. If you do not have any @code{GRANT} statements done, MySQL will optimize the permission checking somewhat. So if you have a very high volume it may be worth the time to avoid grants. Otherwise more permission check results in a larger overhead. If your problem is with some explicit MySQL function, you can always time this in the MySQL client: @example mysql> select benchmark(1000000,1+1); +------------------------+ | benchmark(1000000,1+1) | +------------------------+ | 0 | +------------------------+ 1 row in set (0.32 sec) @end example The above shows that MySQL can execute 1,000,000 @code{+} expressions in 0.32 seconds on a @code{PentiumII 400MHz}. All MySQL functions should be very optimized, but there may be some exceptions, and the @code{benchmark(loop_count,expression)} is a great tool to find out if this is a problem with your query. @menu * EXPLAIN:: @code{EXPLAIN} Syntax (Get Information About a @code{SELECT}) * Estimating performance:: Estimating query performance * SELECT speed:: Speed of @code{SELECT} queries * Where optimizations:: How MySQL optimizes @code{WHERE} clauses * DISTINCT optimization:: How MySQL Optimizes @code{DISTINCT} * LEFT JOIN optimization:: How MySQL optimizes @code{LEFT JOIN} * LIMIT optimization:: How MySQL optimizes @code{LIMIT} * Insert speed:: Speed of @code{INSERT} queries * Update speed:: Speed of @code{UPDATE} queries * Delete speed:: Speed of @code{DELETE} queries * Tips:: Other Optimization Tips @end menu @node EXPLAIN, Estimating performance, Query Speed, Query Speed @subsection @code{EXPLAIN} Syntax (Get Information About a @code{SELECT}) @findex EXPLAIN @findex SELECT, optimizing @example EXPLAIN tbl_name or EXPLAIN SELECT select_options @end example @code{EXPLAIN tbl_name} is a synonym for @code{DESCRIBE tbl_name} or @code{SHOW COLUMNS FROM tbl_name}. When you precede a @code{SELECT} statement with the keyword @code{EXPLAIN}, MySQL explains how it would process the @code{SELECT}, providing information about how tables are joined and in which order. With the help of @code{EXPLAIN}, you can see when you must add indexes to tables to get a faster @code{SELECT} that uses indexes to find the records. You can also see if the optimizer joins the tables in an optimal order. To force the optimizer to use a specific join order for a @code{SELECT} statement, add a @code{STRAIGHT_JOIN} clause. For non-simple joins, @code{EXPLAIN} returns a row of information for each table used in the @code{SELECT} statement. The tables are listed in the order they would be read. MySQL resolves all joins using a single-sweep multi-join method. This means that MySQL reads a row from the first table, then finds a matching row in the second table, then in the third table and so on. When all tables are processed, it outputs the selected columns and backtracks through the table list until a table is found for which there are more matching rows. The next row is read from this table and the process continues with the next table. Output from @code{EXPLAIN} includes the following columns: @table @code @item table The table to which the row of output refers. @item type The join type. Information about the various types is given below. @item possible_keys The @code{possible_keys} column indicates which indexes MySQL could use to find the rows in this table. Note that this column is totally independent of the order of the tables. That means that some of the keys in possible_keys may not be usable in practice with the generated table order. If this column is empty, there are no relevant indexes. In this case, you may be able to improve the performance of your query by examining the @code{WHERE} clause to see if it refers to some column or columns that would be suitable for indexing. If so, create an appropriate index and check the query with @code{EXPLAIN} again. @xref{ALTER TABLE}. To see what indexes a table has, use @code{SHOW INDEX FROM tbl_name}. @item key The @code{key} column indicates the key that MySQL actually decided to use. The key is @code{NULL} if no index was chosen. If MySQL chooses the wrong index, you can probably force MySQL to use another index by using @code{myisamchk --analyze}, @xref{myisamchk syntax}, or by using @code{USE INDEX/IGNORE INDEX}. @xref{JOIN}. @item key_len The @code{key_len} column indicates the length of the key that MySQL decided to use. The length is @code{NULL} if the @code{key} is @code{NULL}. Note that this tells us how many parts of a multi-part key MySQL will actually use. @item ref The @code{ref} column shows which columns or constants are used with the @code{key} to select rows from the table. @item rows The @code{rows} column indicates the number of rows MySQL believes it must examine to execute the query. @item Extra This column contains additional information of how MySQL will resolve the query. Here is an explanation of the different text strings that can be found in this column: @table @code @item Distinct MySQL will not continue searching for more rows for the current row combination after it has found the first matching row. @item Not exists MySQL was able to do a @code{LEFT JOIN} optimization on the query and will not examine more rows in this table for the previous row combination after it finds one row that matches the @code{LEFT JOIN} criteria. Here is an example for this: @example SELECT * FROM t1 LEFT JOIN t2 ON t1.id=t2.id WHERE t2.id IS NULL; @end example Assume that @code{t2.id} is defined with @code{NOT NULL}. In this case MySQL will scan @code{t1} and look up the rows in @code{t2} through @code{t1.id}. If MySQL finds a matching row in @code{t2}, it knows that @code{t2.id} can never be @code{NULL}, and will not scan through the rest of the rows in @code{t2} that has the same @code{id}. In other words, for each row in @code{t1}, MySQL only needs to do a single lookup in @code{t2}, independent of how many matching rows there are in @code{t2}. @item @code{range checked for each record (index map: #)} MySQL didn't find a real good index to use. It will, instead, for each row combination in the preceding tables, do a check on which index to use (if any), and use this index to retrieve the rows from the table. This isn't very fast but is faster than having to do a join without an index. @item Using filesort MySQL will need to do an extra pass to find out how to retrieve the rows in sorted order. The sort is done by going through all rows according to the @code{join type} and storing the sort key + pointer to the row for all rows that match the @code{WHERE}. Then the keys are sorted. Finally the rows are retrieved in sorted order. @item Using index The column information is retrieved from the table using only information in the index tree without having to do an additional seek to read the actual row. This can be done when all the used columns for the table are part of the same index. @item Using temporary To resolve the query MySQL will need to create a temporary table to hold the result. This typically happens if you do an @code{ORDER BY} on a different column set than you did a @code{GROUP BY} on. @item Where used A @code{WHERE} clause will be used to restrict which rows will be matched against the next table or sent to the client. If you don't have this information and the table is of type @code{ALL} or @code{index}, you may have something wrong in your query (if you don't intend to fetch/examine all rows from the table). @end table If you want to get your queries as fast as possible, you should look out for @code{Using filesort} and @code{Using temporary}. @end table The different join types are listed below, ordered from best to worst type: @cindex system table @cindex tables, system @table @code @item system The table has only one row (= system table). This is a special case of the @code{const} join type. @cindex constant table @cindex tables, constant @item const The table has at most one matching row, which will be read at the start of the query. Because there is only one row, values from the column in this row can be regarded as constants by the rest of the optimizer. @code{const} tables are very fast as they are read only once! @item eq_ref One row will be read from this table for each combination of rows from the previous tables. This is the best possible join type, other than the @code{const} types. It is used when all parts of an index are used by the join and the index is @code{UNIQUE} or a @code{PRIMARY KEY}. @item ref All rows with matching index values will be read from this table for each combination of rows from the previous tables. @code{ref} is used if the join uses only a leftmost prefix of the key, or if the key is not @code{UNIQUE} or a @code{PRIMARY KEY} (in other words, if the join cannot select a single row based on the key value). If the key that is used matches only a few rows, this join type is good. @item range Only rows that are in a given range will be retrieved, using an index to select the rows. The @code{key} column indicates which index is used. The @code{key_len} contains the longest key part that was used. The @code{ref} column will be NULL for this type. @item index This is the same as @code{ALL}, except that only the index tree is scanned. This is usually faster than @code{ALL}, as the index file is usually smaller than the data file. @item ALL A full table scan will be done for each combination of rows from the previous tables. This is normally not good if the table is the first table not marked @code{const}, and usually @strong{very} bad in all other cases. You normally can avoid @code{ALL} by adding more indexes, so that the row can be retrieved based on constant values or column values from earlier tables. @end table You can get a good indication of how good a join is by multiplying all values in the @code{rows} column of the @code{EXPLAIN} output. This should tell you roughly how many rows MySQL must examine to execute the query. This number is also used when you restrict queries with the @code{max_join_size} variable. @xref{Server parameters}. The following example shows how a @code{JOIN} can be optimized progressively using the information provided by @code{EXPLAIN}. Suppose you have the @code{SELECT} statement shown below, that you examine using @code{EXPLAIN}: @example EXPLAIN SELECT tt.TicketNumber, tt.TimeIn, tt.ProjectReference, tt.EstimatedShipDate, tt.ActualShipDate, tt.ClientID, tt.ServiceCodes, tt.RepetitiveID, tt.CurrentProcess, tt.CurrentDPPerson, tt.RecordVolume, tt.DPPrinted, et.COUNTRY, et_1.COUNTRY, do.CUSTNAME FROM tt, et, et AS et_1, do WHERE tt.SubmitTime IS NULL AND tt.ActualPC = et.EMPLOYID AND tt.AssignedPC = et_1.EMPLOYID AND tt.ClientID = do.CUSTNMBR; @end example For this example, assume that: @itemize @bullet @item The columns being compared have been declared as follows: @multitable @columnfractions .1 .2 .7 @item @strong{Table} @tab @strong{Column} @tab @strong{Column type} @item @code{tt} @tab @code{ActualPC} @tab @code{CHAR(10)} @item @code{tt} @tab @code{AssignedPC} @tab @code{CHAR(10)} @item @code{tt} @tab @code{ClientID} @tab @code{CHAR(10)} @item @code{et} @tab @code{EMPLOYID} @tab @code{CHAR(15)} @item @code{do} @tab @code{CUSTNMBR} @tab @code{CHAR(15)} @end multitable @item The tables have the indexes shown below: @multitable @columnfractions .1 .9 @item @strong{Table} @tab @strong{Index} @item @code{tt} @tab @code{ActualPC} @item @code{tt} @tab @code{AssignedPC} @item @code{tt} @tab @code{ClientID} @item @code{et} @tab @code{EMPLOYID} (primary key) @item @code{do} @tab @code{CUSTNMBR} (primary key) @end multitable @item The @code{tt.ActualPC} values aren't evenly distributed. @end itemize Initially, before any optimizations have been performed, the @code{EXPLAIN} statement produces the following information: @example table type possible_keys key key_len ref rows Extra et ALL PRIMARY NULL NULL NULL 74 do ALL PRIMARY NULL NULL NULL 2135 et_1 ALL PRIMARY NULL NULL NULL 74 tt ALL AssignedPC,ClientID,ActualPC NULL NULL NULL 3872 range checked for each record (key map: 35) @end example Because @code{type} is @code{ALL} for each table, this output indicates that MySQL is doing a full join for all tables! This will take quite a long time, as the product of the number of rows in each table must be examined! For the case at hand, this is @code{74 * 2135 * 74 * 3872 = 45,268,558,720} rows. If the tables were bigger, you can only imagine how long it would take. One problem here is that MySQL can't (yet) use indexes on columns efficiently if they are declared differently. In this context, @code{VARCHAR} and @code{CHAR} are the same unless they are declared as different lengths. Because @code{tt.ActualPC} is declared as @code{CHAR(10)} and @code{et.EMPLOYID} is declared as @code{CHAR(15)}, there is a length mismatch. To fix this disparity between column lengths, use @code{ALTER TABLE} to lengthen @code{ActualPC} from 10 characters to 15 characters: @example mysql> ALTER TABLE tt MODIFY ActualPC VARCHAR(15); @end example Now @code{tt.ActualPC} and @code{et.EMPLOYID} are both @code{VARCHAR(15)}. Executing the @code{EXPLAIN} statement again produces this result: @example table type possible_keys key key_len ref rows Extra tt ALL AssignedPC,ClientID,ActualPC NULL NULL NULL 3872 where used do ALL PRIMARY NULL NULL NULL 2135 range checked for each record (key map: 1) et_1 ALL PRIMARY NULL NULL NULL 74 range checked for each record (key map: 1) et eq_ref PRIMARY PRIMARY 15 tt.ActualPC 1 @end example This is not perfect, but is much better (the product of the @code{rows} values is now less by a factor of 74). This version is executed in a couple of seconds. A second alteration can be made to eliminate the column length mismatches for the @code{tt.AssignedPC = et_1.EMPLOYID} and @code{tt.ClientID = do.CUSTNMBR} comparisons: @example mysql> ALTER TABLE tt MODIFY AssignedPC VARCHAR(15), MODIFY ClientID VARCHAR(15); @end example Now @code{EXPLAIN} produces the output shown below: @example table type possible_keys key key_len ref rows Extra et ALL PRIMARY NULL NULL NULL 74 tt ref AssignedPC,ClientID,ActualPC ActualPC 15 et.EMPLOYID 52 where used et_1 eq_ref PRIMARY PRIMARY 15 tt.AssignedPC 1 do eq_ref PRIMARY PRIMARY 15 tt.ClientID 1 @end example This is almost as good as it can get. The remaining problem is that, by default, MySQL assumes that values in the @code{tt.ActualPC} column are evenly distributed, and that isn't the case for the @code{tt} table. Fortunately, it is easy to tell MySQL about this: @example shell> myisamchk --analyze PATH_TO_MYSQL_DATABASE/tt shell> mysqladmin refresh @end example Now the join is perfect, and @code{EXPLAIN} produces this result: @example table type possible_keys key key_len ref rows Extra tt ALL AssignedPC,ClientID,ActualPC NULL NULL NULL 3872 where used et eq_ref PRIMARY PRIMARY 15 tt.ActualPC 1 et_1 eq_ref PRIMARY PRIMARY 15 tt.AssignedPC 1 do eq_ref PRIMARY PRIMARY 15 tt.ClientID 1 @end example Note that the @code{rows} column in the output from @code{EXPLAIN} is an educated guess from the MySQL join optimizer. To optimize a query, you should check if the numbers are even close to the truth. If not, you may get better performance by using @code{STRAIGHT_JOIN} in your @code{SELECT} statement and trying to list the tables in a different order in the @code{FROM} clause. @node Estimating performance, SELECT speed, EXPLAIN, Query Speed @subsection Estimating Query Performance @cindex estimating, query performance @cindex queries, estimating performance @cindex performance, estimating In most cases you can estimate the performance by counting disk seeks. For small tables, you can usually find the row in 1 disk seek (as the index is probably cached). For bigger tables, you can estimate that (using B++ tree indexes) you will need: @code{log(row_count) / log(index_block_length / 3 * 2 / (index_length + data_pointer_length)) + 1} seeks to find a row. In MySQL an index block is usually 1024 bytes and the data pointer is usually 4 bytes. A 500,000 row table with an index length of 3 (medium integer) gives you: @code{log(500,000)/log(1024/3*2/(3+4)) + 1} = 4 seeks. As the above index would require about 500,000 * 7 * 3/2 = 5.2M, (assuming that the index buffers are filled to 2/3, which is typical) you will probably have much of the index in memory and you will probably only need 1-2 calls to read data from the OS to find the row. For writes, however, you will need 4 seek requests (as above) to find where to place the new index and normally 2 seeks to update the index and write the row. Note that the above doesn't mean that your application will slowly degenerate by N log N! As long as everything is cached by the OS or SQL server things will only go marginally slower while the table gets bigger. After the data gets too big to be cached, things will start to go much slower until your applications is only bound by disk-seeks (which increase by N log N). To avoid this, increase the index cache as the data grows. @xref{Server parameters}. @node SELECT speed, Where optimizations, Estimating performance, Query Speed @subsection Speed of @code{SELECT} Queries @findex SELECT speed @cindex speed, of queries In general, when you want to make a slow @code{SELECT ... WHERE} faster, the first thing to check is whether or not you can add an index. @xref{MySQL indexes, , MySQL indexes}. All references between different tables should usually be done with indexes. You can use the @code{EXPLAIN} command to determine which indexes are used for a @code{SELECT}. @xref{EXPLAIN, , @code{EXPLAIN}}. Some general tips: @itemize @bullet @item To help MySQL optimize queries better, run @code{myisamchk --analyze} on a table after it has been loaded with relevant data. This updates a value for each index part that indicates the average number of rows that have the same value. (For unique indexes, this is always 1, of course.). MySQL will use this to decide which index to choose when you connect two tables with 'a non-constant expression'. You can check the result from the @code{analyze} run by doing @code{SHOW INDEX FROM table_name} and examining the @code{Cardinality} column. @item To sort an index and data according to an index, use @code{myisamchk --sort-index --sort-records=1} (if you want to sort on index 1). If you have a unique index from which you want to read all records in order according to that index, this is a good way to make that faster. Note, however, that this sorting isn't written optimally and will take a long time for a large table! @end itemize @node Where optimizations, DISTINCT optimization, SELECT speed, Query Speed @subsection How MySQL Optimizes @code{WHERE} Clauses @findex WHERE @cindex optimizations The @code{WHERE} optimizations are put in the @code{SELECT} part here because they are mostly used with @code{SELECT}, but the same optimizations apply for @code{WHERE} in @code{DELETE} and @code{UPDATE} statements. Also note that this section is incomplete. MySQL does many optimizations, and we have not had time to document them all. Some of the optimizations performed by MySQL are listed below: @itemize @bullet @item Removal of unnecessary parentheses: @example ((a AND b) AND c OR (((a AND b) AND (c AND d)))) -> (a AND b AND c) OR (a AND b AND c AND d) @end example @item Constant folding: @example (a<b AND b=c) AND a=5 -> b>5 AND b=c AND a=5 @end example @item Constant condition removal (needed because of constant folding): @example (B>=5 AND B=5) OR (B=6 AND 5=5) OR (B=7 AND 5=6) -> B=5 OR B=6 @end example @item Constant expressions used by indexes are evaluated only once. @item @code{COUNT(*)} on a single table without a @code{WHERE} is retrieved directly from the table information. This is also done for any @code{NOT NULL} expression when used with only one table. @item Early detection of invalid constant expressions. MySQL quickly detects that some @code{SELECT} statements are impossible and returns no rows. @item @code{HAVING} is merged with @code{WHERE} if you don't use @code{GROUP BY} or group functions (@code{COUNT()}, @code{MIN()}...). @item For each sub-join, a simpler @code{WHERE} is constructed to get a fast @code{WHERE} evaluation for each sub-join and also to skip records as soon as possible. @cindex constant table @cindex tables, constant @item All constant tables are read first, before any other tables in the query. A constant table is: @itemize @minus @item An empty table or a table with 1 row. @item A table that is used with a @code{WHERE} clause on a @code{UNIQUE} index, or a @code{PRIMARY KEY}, where all index parts are used with constant expressions and the index parts are defined as @code{NOT NULL}. @end itemize All the following tables are used as constant tables: @example mysql> SELECT * FROM t WHERE primary_key=1; mysql> SELECT * FROM t1,t2 WHERE t1.primary_key=1 AND t2.primary_key=t1.id; @end example @item The best join combination to join the tables is found by trying all possibilities. If all columns in @code{ORDER BY} and in @code{GROUP BY} come from the same table, then this table is preferred first when joining. @item If there is an @code{ORDER BY} clause and a different @code{GROUP BY} clause, or if the @code{ORDER BY} or @code{GROUP BY} contains columns from tables other than the first table in the join queue, a temporary table is created. @item If you use @code{SQL_SMALL_RESULT}, MySQL will use an in-memory temporary table. @item Each table index is queried, and the best index that spans fewer than 30% of the rows is used. If no such index can be found, a quick table scan is used. @item In some cases, MySQL can read rows from the index without even consulting the data file. If all columns used from the index are numeric, then only the index tree is used to resolve the query. @item Before each record is output, those that do not match the @code{HAVING} clause are skipped. @end itemize Some examples of queries that are very fast: @example mysql> SELECT COUNT(*) FROM tbl_name; mysql> SELECT MIN(key_part1),MAX(key_part1) FROM tbl_name; mysql> SELECT MAX(key_part2) FROM tbl_name WHERE key_part_1=constant; mysql> SELECT ... FROM tbl_name ORDER BY key_part1,key_part2,... LIMIT 10; mysql> SELECT ... FROM tbl_name ORDER BY key_part1 DESC,key_part2 DESC,... LIMIT 10; @end example The following queries are resolved using only the index tree (assuming the indexed columns are numeric): @example mysql> SELECT key_part1,key_part2 FROM tbl_name WHERE key_part1=val; mysql> SELECT COUNT(*) FROM tbl_name WHERE key_part1=val1 AND key_part2=val2; mysql> SELECT key_part2 FROM tbl_name GROUP BY key_part1; @end example The following queries use indexing to retrieve the rows in sorted order without a separate sorting pass: @example mysql> SELECT ... FROM tbl_name ORDER BY key_part1,key_part2,... ; mysql> SELECT ... FROM tbl_name ORDER BY key_part1 DESC,key_part2 DESC,... ; @end example @node DISTINCT optimization, LEFT JOIN optimization, Where optimizations, Query Speed @subsection How MySQL Optimizes @code{DISTINCT} @findex DISTINCT @cindex optimizing, DISTINCT @code{DISTINCT} is converted to a @code{GROUP BY} on all columns, @code{DISTINCT} combined with @code{ORDER BY} will in many cases also need a temporary table. When combining @code{LIMIT #} with @code{DISTINCT}, MySQL will stop as soon as it finds @code{#} unique rows. If you don't use columns from all used tables, MySQL will stop the scanning of the not used tables as soon as it has found the first match. @example SELECT DISTINCT t1.a FROM t1,t2 where t1.a=t2.a; @end example In the case, assuming t1 is used before t2 (check with @code{EXPLAIN}), then MySQL will stop reading from t2 (for that particular row in t1) when the first row in t2 is found. @node LEFT JOIN optimization, LIMIT optimization, DISTINCT optimization, Query Speed @subsection How MySQL Optimizes @code{LEFT JOIN} and @code{RIGHT JOIN} @findex LEFT JOIN @cindex optimizing, LEFT JOIN @code{A LEFT JOIN B} in MySQL is implemented as follows: @itemize @bullet @item The table @code{B} is set to be dependent on table @code{A} and all tables that @code{A} is dependent on. @item The table @code{A} is set to be dependent on all tables (except @code{B}) that are used in the @code{LEFT JOIN} condition. @item All @code{LEFT JOIN} conditions are moved to the @code{WHERE} clause. @item All standard join optimizations are done, with the exception that a table is always read after all tables it is dependent on. If there is a circular dependence then MySQL will issue an error. @item All standard @code{WHERE} optimizations are done. @item If there is a row in @code{A} that matches the @code{WHERE} clause, but there wasn't any row in @code{B} that matched the @code{LEFT JOIN} condition, then an extra @code{B} row is generated with all columns set to @code{NULL}. @item If you use @code{LEFT JOIN} to find rows that don't exist in some table and you have the following test: @code{column_name IS NULL} in the @code{WHERE} part, where column_name is a column that is declared as @code{NOT NULL}, then MySQL will stop searching after more rows (for a particular key combination) after it has found one row that matches the @code{LEFT JOIN} condition. @end itemize @code{RIGHT JOIN} is implemented analogously as @code{LEFT JOIN}. The table read order forced by @code{LEFT JOIN} and @code{STRAIGHT JOIN} will help the join optimizer (which calculates in which order tables should be joined) to do its work much more quickly, as there are fewer table permutations to check. Note that the above means that if you do a query of type: @example SELECT * FROM a,b LEFT JOIN c ON (c.key=a.key) LEFT JOIN d (d.key=a.key) WHERE b.key=d.key @end example MySQL will do a full scan on @code{b} as the @code{LEFT JOIN} will force it to be read before @code{d}. The fix in this case is to change the query to: @example SELECT * FROM b,a LEFT JOIN c ON (c.key=a.key) LEFT JOIN d (d.key=a.key) WHERE b.key=d.key @end example @node LIMIT optimization, Insert speed, LEFT JOIN optimization, Query Speed @subsection How MySQL Optimizes @code{LIMIT} @findex LIMIT @cindex optimizing, LIMIT In some cases MySQL will handle the query differently when you are using @code{LIMIT #} and not using @code{HAVING}: @itemize @bullet @item If you are selecting only a few rows with @code{LIMIT}, MySQL will use indexes in some cases when it normally would prefer to do a full table scan. @item If you use @code{LIMIT #} with @code{ORDER BY}, MySQL will end the sorting as soon as it has found the first @code{#} lines instead of sorting the whole table. @item When combining @code{LIMIT #} with @code{DISTINCT}, MySQL will stop as soon as it finds @code{#} unique rows. @item In some cases a @code{GROUP BY} can be resolved by reading the key in order (or do a sort on the key) and then calculate summaries until the key value changes. In this case @code{LIMIT #} will not calculate any unnecessary @code{GROUP BY}'s. @item As soon as MySQL has sent the first @code{#} rows to the client, it will abort the query. @item @code{LIMIT 0} will always quickly return an empty set. This is useful to check the query and to get the column types of the result columns. @item The size of temporary tables uses the @code{LIMIT #} to calculate how much space is needed to resolve the query. @end itemize @node Insert speed, Update speed, LIMIT optimization, Query Speed @subsection Speed of @code{INSERT} Queries @findex INSERT @cindex speed, inserting @cindex inserting, speed of The time to insert a record consists approximately of: @itemize @bullet @item Connect: (3) @item Sending query to server: (2) @item Parsing query: (2) @item Inserting record: (1 x size of record) @item Inserting indexes: (1 x number of indexes) @item Close: (1) @end itemize where the numbers are somewhat proportional to the overall time. This does not take into consideration the initial overhead to open tables (which is done once for each concurrently running query). The size of the table slows down the insertion of indexes by N log N (B-trees). Some ways to speed up inserts: @itemize @bullet @item If you are inserting many rows from the same client at the same time, use multiple value lists @code{INSERT} statements. This is much faster (many times in some cases) than using separate @code{INSERT} statements. @item If you are inserting a lot of rows from different clients, you can get higher speed by using the @code{INSERT DELAYED} statement. @xref{INSERT, , @code{INSERT}}. @item Note that with @code{MyISAM} you can insert rows at the same time @code{SELECT}s are running if there are no deleted rows in the tables. @item When loading a table from a text file, use @code{LOAD DATA INFILE}. This is usually 20 times faster than using a lot of @code{INSERT} statements. @xref{LOAD DATA, , @code{LOAD DATA}}. @item It is possible with some extra work to make @code{LOAD DATA INFILE} run even faster when the table has many indexes. Use the following procedure: @enumerate @item Optionally create the table with @code{CREATE TABLE}. For example, using @code{mysql} or Perl-DBI. @item Execute a @code{FLUSH TABLES} statement or the shell command @code{mysqladmin flush-tables}. @item Use @code{myisamchk --keys-used=0 -rq /path/to/db/tbl_name}. This will remove all usage of all indexes from the table. @item Insert data into the table with @code{LOAD DATA INFILE}. This will not update any indexes and will therefore be very fast. @item If you are going to only read the table in the future, run @code{myisampack} on it to make it smaller. @xref{Compressed format}. @item Re-create the indexes with @code{myisamchk -r -q /path/to/db/tbl_name}. This will create the index tree in memory before writing it to disk, which is much faster because it avoids lots of disk seeks. The resulting index tree is also perfectly balanced. @item Execute a @code{FLUSH TABLES} statement or the shell command @code{mysqladmin flush-tables}. @end enumerate This procedure will be built into @code{LOAD DATA INFILE} in some future version of MySQL. @item You can speed up insertions by locking your tables: @example mysql> LOCK TABLES a WRITE; mysql> INSERT INTO a VALUES (1,23),(2,34),(4,33); mysql> INSERT INTO a VALUES (8,26),(6,29); mysql> UNLOCK TABLES; @end example The main speed difference is that the index buffer is flushed to disk only once, after all @code{INSERT} statements have completed. Normally there would be as many index buffer flushes as there are different @code{INSERT} statements. Locking is not needed if you can insert all rows with a single statement. Locking will also lower the total time of multi-connection tests, but the maximum wait time for some threads will go up (because they wait for locks). For example: @example thread 1 does 1000 inserts thread 2, 3, and 4 does 1 insert thread 5 does 1000 inserts @end example If you don't use locking, 2, 3, and 4 will finish before 1 and 5. If you use locking, 2, 3, and 4 probably will not finish before 1 or 5, but the total time should be about 40% faster. As @code{INSERT}, @code{UPDATE}, and @code{DELETE} operations are very fast in MySQL, you will obtain better overall performance by adding locks around everything that does more than about 5 inserts or updates in a row. If you do very many inserts in a row, you could do a @code{LOCK TABLES} followed by an @code{UNLOCK TABLES} once in a while (about each 1000 rows) to allow other threads access to the table. This would still result in a nice performance gain. Of course, @code{LOAD DATA INFILE} is much faster for loading data. @end itemize To get some more speed for both @code{LOAD DATA INFILE} and @code{INSERT}, enlarge the key buffer. @xref{Server parameters}. @node Update speed, Delete speed, Insert speed, Query Speed @subsection Speed of @code{UPDATE} Queries Update queries are optimized as a @code{SELECT} query with the additional overhead of a write. The speed of the write is dependent on the size of the data that is being updated and the number of indexes that are updated. Indexes that are not changed will not be updated. Also, another way to get fast updates is to delay updates and then do many updates in a row later. Doing many updates in a row is much quicker than doing one at a time if you lock the table. Note that, with dynamic record format, updating a record to a longer total length may split the record. So if you do this often, it is very important to @code{OPTIMIZE TABLE} sometimes. @xref{OPTIMIZE TABLE, , @code{OPTIMIZE TABLE}}. @node Delete speed, Tips, Update speed, Query Speed @subsection Speed of @code{DELETE} Queries If you want to delete all rows in the table, you should use @code{TRUNCATE TABLE table_name}. @xref{TRUNCATE}. The time to delete a record is exactly proportional to the number of indexes. To delete records more quickly, you can increase the size of the index cache. @xref{Server parameters}. @node Tips, , Delete speed, Query Speed @subsection Other Optimization Tips @cindex optimization, tips @cindex tips, optimization Unsorted tips for faster systems: @itemize @bullet @item Use persistent connections to the database to avoid the connection overhead. If you can't use persistent connections and you are doing a lot of new connections to the database, you may want to change the value of the @code{thread_cache_size} variable. @xref{Server parameters}. @item Always check that all your queries really use the indexes you have created in the tables. In MySQL you can do this with the @code{EXPLAIN} command. @xref{EXPLAIN, Explain, Explain, manual}. @item Try to avoid complex @code{SELECT} queries on tables that are updated a lot. This is to avoid problems with table locking. @item The new @code{MyISAM} tables can insert rows in a table without deleted rows at the same time another table is reading from it. If this is important for you, you should consider methods where you don't have to delete rows or run @code{OPTIMIZE TABLE} after you have deleted a lot of rows. @item Use @code{ALTER TABLE ... ORDER BY expr1,expr2...} if you mostly retrieve rows in expr1,expr2.. order. By using this option after big changes to the table, you may be able to get higher performance. @item In some cases it may make sense to introduce a column that is 'hashed' based on information from other columns. If this column is short and reasonably unique it may be much faster than a big index on many columns. In MySQL it's very easy to use this extra column: @code{SELECT * FROM table_name WHERE hash=MD5(concat(col1,col2)) AND col_1='constant' AND col_2='constant'} @item For tables that change a lot you should try to avoid all @code{VARCHAR} or @code{BLOB} columns. You will get dynamic row length as soon as you are using a single @code{VARCHAR} or @code{BLOB} column. @xref{Table types}. @item It's not normally useful to split a table into different tables just because the rows gets 'big'. To access a row, the biggest performance hit is the disk seek to find the first byte of the row. After finding the data most new disks can read the whole row fast enough for most applications. The only cases where it really matters to split up a table is if it's a dynamic row size table (see above) that you can change to a fixed row size, or if you very often need to scan the table and don't need most of the columns. @xref{Table types}. @item If you very often need to calculate things based on information from a lot of rows (like counts of things), it's probably much better to introduce a new table and update the counter in real time. An update of type @code{UPDATE table set count=count+1 where index_column=constant} is very fast! This is really important when you use databases like MySQL that only have table locking (multiple readers / single writers). This will also give better performance with most databases, as the row locking manager in this case will have less to do. @item If you need to collect statistics from big log tables, use summary tables instead of scanning the whole table. Maintaining the summaries should be much faster than trying to do statistics 'live'. It's much faster to regenerate new summary tables from the logs when things change (depending on business decisions) than to have to change the running application! @item If possible, one should classify reports as 'live' or 'statistical', where data needed for statistical reports are only generated based on summary tables that are generated from the actual data. @item Take advantage of the fact that columns have default values. Insert values explicitly only when the value to be inserted differs from the default. This reduces the parsing that MySQL need to do and improves the insert speed. @item In some cases it's convenient to pack and store data into a blob. In this case you have to add some extra code in your application to pack/unpack things in the blob, but this may save a lot of accesses at some stage. This is practical when you have data that doesn't conform to a static table structure. @item Normally you should try to keep all data non-redundant (what is called 3rd normal form in database theory), but you should not be afraid of duplicating things or creating summary tables if you need these to gain more speed. @item Stored procedures or UDF (user-defined functions) may be a good way to get more performance. In this case you should, however, always have a way to do this some other (slower) way if you use some database that doesn't support this. @item You can always gain something by caching queries/answers in your application and trying to do many inserts/updates at the same time. If your database supports lock tables (like MySQL and Oracle), this should help to ensure that the index cache is only flushed once after all updates. @item Use @code{INSERT /*! DELAYED */} when you do not need to know when your data is written. This speeds things up because many records can be written with a single disk write. @item Use @code{INSERT /*! LOW_PRIORITY */} when you want your selects to be more important. @item Use @code{SELECT /*! HIGH_PRIORITY */} to get selects that jump the queue. That is, the select is done even if there is somebody waiting to do a write. @item Use the multi-line @code{INSERT} statement to store many rows with one SQL command (many SQL servers supports this). @item Use @code{LOAD DATA INFILE} to load bigger amounts of data. This is faster than normal inserts and will be even faster when @code{myisamchk} is integrated in @code{mysqld}. @item Use @code{AUTO_INCREMENT} columns to make unique values. @item Use @code{OPTIMIZE TABLE} once in a while to avoid fragmentation when using dynamic table format. @xref{OPTIMIZE TABLE, , @code{OPTIMIZE TABLE}}. @item Use @code{HEAP} tables to get more speed when possible. @xref{Table types}. @item When using a normal Web server setup, images should be stored as files. That is, store only a file reference in the database. The main reason for this is that a normal Web server is much better at caching files than database contents. So it it's much easier to get a fast system if you are using files. @item Use in memory tables for non-critical data that are accessed often (like information about the last shown banner for users that don't have cookies). @item Columns with identical information in different tables should be declared identical and have identical names. Before Version 3.23 you got slow joins otherwise. Try to keep the names simple (use @code{name} instead of @code{customer_name} in the customer table). To make your names portable to other SQL servers you should keep them shorter than 18 characters. @item If you need REALLY high speed, you should take a look at the low-level interfaces for data storage that the different SQL servers support! For example, by accessing the MySQL @code{MyISAM} directly, you could get a speed increase of 2-5 times compared to using the SQL interface. To be able to do this the data must be on the same server as the application, and usually it should only be accessed by one process (because external file locking is really slow). One could eliminate the above problems by introducing low-level @code{MyISAM} commands in the MySQL server (this could be one easy way to get more performance if needed). By carefully designing the database interface, it should be quite easy to support this types of optimization. @item In many cases it's faster to access data from a database (using a live connection) than accessing a text file, just because the database is likely to be more compact than the text file (if you are using numerical data), and this will involve fewer disk accesses. You will also save code because you don't have to parse your text files to find line and column boundaries. @item You can also use replication to speed things up. @xref{Replication}. @item Declaring a table with @code{DELAY_KEY_WRITE=1} will make the updating of indexes faster, as these are not logged to disk until the file is closed. The downside is that you should run @code{myisamchk} on these tables before you start @code{mysqld} to ensure that they are okay if something killed @code{mysqld} in the middle. As the key information can always be generated from the data, you should not lose anything by using @code{DELAY_KEY_WRITE}. @end itemize @node Locking Issues, Optimizing Database Structure, Query Speed, MySQL Optimization @section Locking Issues @menu * Internal locking:: How MySQL Locks Tables * Table locking:: Table Locking Issues @end menu @node Internal locking, Table locking, Locking Issues, Locking Issues @subsection How MySQL Locks Tables @cindex internal locking @cindex locking, tables @cindex tables, locking You can find a discussion about different locking methods in the appendix. @xref{Locking methods}. All locking in MySQL is deadlock-free. This is managed by always requesting all needed locks at once at the beginning of a query and always locking the tables in the same order. The locking method MySQL uses for @code{WRITE} locks works as follows: @itemize @bullet @item If there are no locks on the table, put a write lock on it. @item Otherwise, put the lock request in the write lock queue. @end itemize The locking method MySQL uses for @code{READ} locks works as follows: @itemize @bullet @item If there are no write locks on the table, put a read lock on it. @item Otherwise, put the lock request in the read lock queue. @end itemize When a lock is released, the lock is made available to the threads in the write lock queue, then to the threads in the read lock queue. This means that if you have many updates on a table, @code{SELECT} statements will wait until there are no more updates. To work around this for the case where you want to do many @code{INSERT} and @code{SELECT} operations on a table, you can insert rows in a temporary table and update the real table with the records from the temporary table once in a while. This can be done with the following code: @example mysql> LOCK TABLES real_table WRITE, insert_table WRITE; mysql> insert into real_table select * from insert_table; mysql> TRUNCATE TABLE insert_table; mysql> UNLOCK TABLES; @end example You can use the @code{LOW_PRIORITY} options with @code{INSERT}, @code{UPDATE} or @code{DELETE} or @code{HIGH_PRIORITY} with @code{SELECT} if you want to prioritize retrieval in some specific cases. You can also start @code{mysqld} with @code{--low-priority-updates} to get the same behaveour. Using @code{SQL_BUFFER_RESULT} can also help making table locks shorter. @xref{SELECT}. You could also change the locking code in @file{mysys/thr_lock.c} to use a single queue. In this case, write locks and read locks would have the same priority, which might help some applications. @node Table locking, , Internal locking, Locking Issues @subsection Table Locking Issues @cindex problems, table locking The table locking code in MySQL is deadlock free. MySQL uses table locking (instead of row locking or column locking) on all table types, except @code{BDB} tables, to achieve a very high lock speed. For large tables, table locking is MUCH better than row locking for most applications, but there are, of course, some pitfalls. For @code{BDB} and @code{InnoDB} tables, MySQL only uses table locking if you explicitely lock the table with @code{LOCK TABLES} or execute a command that will modify every row in the table, like @code{ALTER TABLE}. For these table types we recommend you to not use @code{LOCK TABLES} at all. In MySQL Version 3.23.7 and above, you can insert rows into @code{MyISAM} tables at the same time other threads are reading from the table. Note that currently this only works if there are no holes after deleted rows in the table at the time the insert is made. When all holes has been filled with new data, concurrent inserts will automatically be enabled again. Table locking enables many threads to read from a table at the same time, but if a thread wants to write to a table, it must first get exclusive access. During the update, all other threads that want to access this particular table will wait until the update is ready. As updates on tables normally are considered to be more important than @code{SELECT}, all statements that update a table have higher priority than statements that retrieve information from a table. This should ensure that updates are not 'starved' because one issues a lot of heavy queries against a specific table. (You can change this by using LOW_PRIORITY with the statement that does the update or @code{HIGH_PRIORITY} with the @code{SELECT} statement.) Starting from MySQL Version 3.23.7 one can use the @code{max_write_lock_count} variable to force MySQL to temporary give all @code{SELECT} statements, that wait for a table, a higher priority after a specific number of inserts on a table. Table locking is, however, not very good under the following senario: @itemize @bullet @item A client issues a @code{SELECT} that takes a long time to run. @item Another client then issues an @code{UPDATE} on a used table. This client will wait until the @code{SELECT} is finished. @item Another client issues another @code{SELECT} statement on the same table. As @code{UPDATE} has higher priority than @code{SELECT}, this @code{SELECT} will wait for the @code{UPDATE} to finish. It will also wait for the first @code{SELECT} to finish! @item A thread is waiting for something like @code{full disk}, in which case all threads that wants to access the problem table will also be put in a waiting state until more disk space is made available. @end itemize Some possible solutions to this problem are: @itemize @bullet @item Try to get the @code{SELECT} statements to run faster. You may have to create some summary tables to do this. @item Start @code{mysqld} with @code{--low-priority-updates}. This will give all statements that update (modify) a table lower priority than a @code{SELECT} statement. In this case the last @code{SELECT} statement in the previous scenario would execute before the @code{INSERT} statement. @item You can give a specific @code{INSERT}, @code{UPDATE}, or @code{DELETE} statement lower priority with the @code{LOW_PRIORITY} attribute. @item Start @code{mysqld} with a low value for @strong{max_write_lock_count} to give @code{READ} locks after a certain number of @code{WRITE} locks. @item You can specify that all updates from a specific thread should be done with low priority by using the SQL command: @code{SET SQL_LOW_PRIORITY_UPDATES=1}. @xref{SET OPTION, , @code{SET OPTION}}. @item You can specify that a specific @code{SELECT} is very important with the @code{HIGH_PRIORITY} attribute. @xref{SELECT, , @code{SELECT}}. @item If you have problems with @code{INSERT} combined with @code{SELECT}, switch to use the new @code{MyISAM} tables as these support concurrent @code{SELECT}s and @code{INSERT}s. @item If you mainly mix @code{INSERT} and @code{SELECT} statements, the @code{DELAYED} attribute to @code{INSERT} will probably solve your problems. @xref{INSERT, , @code{INSERT}}. @item If you have problems with @code{SELECT} and @code{DELETE}, the @code{LIMIT} option to @code{DELETE} may help. @xref{DELETE, , @code{DELETE}}. @end itemize @node Optimizing Database Structure, Optimizing the Server, Locking Issues, MySQL Optimization @section Optimizing Database Structure @menu * Design:: Design Choices * Data size:: Get Your Data as Small as Possible * MySQL indexes:: How MySQL Uses Indexes * Indexes:: Column Indexes * Multiple-column indexes:: Multiple-Column Indexes * Table cache:: How MySQL Opens and Closes Tables * Creating many tables:: Drawbacks to Creating Large Numbers of Tables in the Same Database * Open tables:: Why So Many Open tables? @end menu @node Design, Data size, Optimizing Database Structure, Optimizing Database Structure @subsection Design Choices @cindex design, choices @cindex database design @cindex storage of data MySQL keeps row data and index data in separate files. Many (almost all) other databases mix row and index data in the same file. We believe that the MySQL choice is better for a very wide range of modern systems. Another way to store the row data is to keep the information for each column in a separate area (examples are SDBM and Focus). This will cause a performance hit for every query that accesses more than one column. Because this degenerates so quickly when more than one column is accessed, we believe that this model is not good for general purpose databases. The more common case is that the index and data are stored together (like in Oracle/Sybase et al). In this case you will find the row information at the leaf page of the index. The good thing with this layout is that it, in many cases, depending on how well the index is cached, saves a disk read. The bad things with this layout are: @itemize @bullet @item Table scanning is much slower because you have to read through the indexes to get at the data. @item You can't use only the index table to retrieve data for a query. @item You lose a lot of space, as you must duplicate indexes from the nodes (as you can't store the row in the nodes). @item Deletes will degenerate the table over time (as indexes in nodes are usually not updated on delete). @item It's harder to cache ONLY the index data. @end itemize @node Data size, MySQL indexes, Design, Optimizing Database Structure @subsection Get Your Data as Small as Possible @cindex data, size @cindex reducing, data size @cindex storage space, minimizing @cindex tables, improving performance @cindex performance, improving One of the most basic optimization is to get your data (and indexes) to take as little space on the disk (and in memory) as possible. This can give huge improvements because disk reads are faster and normally less main memory will be used. Indexing also takes less resources if done on smaller columns. MySQL supports a lot of different table types and row formats. Choosing the right table format may give you a big performance gain. @xref{Table types}. You can get better performance on a table and minimize storage space using the techniques listed below: @itemize @bullet @item Use the most efficient (smallest) types possible. MySQL has many specialized types that save disk space and memory. @item Use the smaller integer types if possible to get smaller tables. For example, @code{MEDIUMINT} is often better than @code{INT}. @item Declare columns to be @code{NOT NULL} if possible. It makes everything faster and you save one bit per column. Note that if you really need @code{NULL} in your application you should definitely use it. Just avoid having it on all columns by default. @item If you don't have any variable-length columns (@code{VARCHAR}, @code{TEXT}, or @code{BLOB} columns), a fixed-size record format is used. This is faster but unfortunately may waste some space. @xref{MyISAM table formats}. @item The primary index of a table should be as short as possible. This makes identification of one row easy and efficient. @item For each table, you have to decide which storage/index method to use. @xref{Table types}. @item Only create the indexes that you really need. Indexes are good for retrieval but bad when you need to store things fast. If you mostly access a table by searching on a combination of columns, make an index on them. The first index part should be the most used column. If you are ALWAYS using many columns, you should use the column with more duplicates first to get better compression of the index. @item If it's very likely that a column has a unique prefix on the first number of characters, it's better to only index this prefix. MySQL supports an index on a part of a character column. Shorter indexes are faster not only because they take less disk space but also because they will give you more hits in the index cache and thus fewer disk seeks. @xref{Server parameters}. @item In some circumstances it can be beneficial to split into two a table that is scanned very often. This is especially true if it is a dynamic format table and it is possible to use a smaller static format table that can be used to find the relevant rows when scanning the table. @end itemize @node MySQL indexes, Indexes, Data size, Optimizing Database Structure @subsection How MySQL Uses Indexes @cindex indexes, use of Indexes are used to find rows with a specific value of one column fast. Without an index MySQL has to start with the first record and then read through the whole table until it finds the relevant rows. The bigger the table, the more this costs. If the table has an index for the columns in question, MySQL can quickly get a position to seek to in the middle of the data file without having to look at all the data. If a table has 1000 rows, this is at least 100 times faster than reading sequentially. Note that if you need to access almost all 1000 rows it is faster to read sequentially because we then avoid disk seeks. All MySQL indexes (@code{PRIMARY}, @code{UNIQUE}, and @code{INDEX}) are stored in B-trees. Strings are automatically prefix- and end-space compressed. @xref{CREATE INDEX, , @code{CREATE INDEX}}. Indexes are used to: @itemize @bullet @item Quickly find the rows that match a @code{WHERE} clause. @item Retrieve rows from other tables when performing joins. @item Find the @code{MAX()} or @code{MIN()} value for a specific indexed column. This is optimized by a preprocessor that checks if you are using @code{WHERE} key_part_# = constant on all key parts < N. In this case MySQL will do a single key lookup and replace the @code{MIN()} expression with a constant. If all expressions are replaced with constants, the query will return at once: @example SELECT MIN(key_part2),MAX(key_part2) FROM table_name where key_part1=10 @end example @item Sort or group a table if the sorting or grouping is done on a leftmost prefix of a usable key (for example, @code{ORDER BY key_part_1,key_part_2 }). The key is read in reverse order if all key parts are followed by @code{DESC}. The index can also be used even if the @code{ORDER BY} doesn't match the index exactly, as long as all the unused index parts and all the extra are @code{ORDER BY} columns are constants in the @code{WHERE} clause. The following queries will use the index to resolve the @code{ORDER BY} part: @example SELECT * FROM foo ORDER BY key_part1,key_part2,key_part3; SELECT * FROM foo WHERE column=constant ORDER BY column, key_part1; SELECT * FROM foo WHERE key_part1=const GROUP BY key_part2; @end example @item In some cases a query can be optimized to retrieve values without consulting the data file. If all used columns for some table are numeric and form a leftmost prefix for some key, the values may be retrieved from the index tree for greater speed: @example SELECT key_part3 FROM table_name WHERE key_part1=1 @end example @end itemize Suppose you issue the following @code{SELECT} statement: @example mysql> SELECT * FROM tbl_name WHERE col1=val1 AND col2=val2; @end example If a multiple-column index exists on @code{col1} and @code{col2}, the appropriate rows can be fetched directly. If separate single-column indexes exist on @code{col1} and @code{col2}, the optimizer tries to find the most restrictive index by deciding which index will find fewer rows and using that index to fetch the rows. @cindex indexes, leftmost prefix of @cindex leftmost prefix of indexes If the table has a multiple-column index, any leftmost prefix of the index can be used by the optimizer to find rows. For example, if you have a three-column index on @code{(col1,col2,col3)}, you have indexed search capabilities on @code{(col1)}, @code{(col1,col2)}, and @code{(col1,col2,col3)}. MySQL can't use a partial index if the columns don't form a leftmost prefix of the index. Suppose you have the @code{SELECT} statements shown below: @example mysql> SELECT * FROM tbl_name WHERE col1=val1; mysql> SELECT * FROM tbl_name WHERE col2=val2; mysql> SELECT * FROM tbl_name WHERE col2=val2 AND col3=val3; @end example If an index exists on @code{(col1,col2,col3)}, only the first query shown above uses the index. The second and third queries do involve indexed columns, but @code{(col2)} and @code{(col2,col3)} are not leftmost prefixes of @code{(col1,col2,col3)}. @findex LIKE, and indexes @findex LIKE, and wildcards @cindex indexes, and @code{LIKE} @cindex wildcards, and @code{LIKE} MySQL also uses indexes for @code{LIKE} comparisons if the argument to @code{LIKE} is a constant string that doesn't start with a wild-card character. For example, the following @code{SELECT} statements use indexes: @example mysql> select * from tbl_name where key_col LIKE "Patrick%"; mysql> select * from tbl_name where key_col LIKE "Pat%_ck%"; @end example In the first statement, only rows with @code{"Patrick" <= key_col < "Patricl"} are considered. In the second statement, only rows with @code{"Pat" <= key_col < "Pau"} are considered. The following @code{SELECT} statements will not use indexes: @example mysql> select * from tbl_name where key_col LIKE "%Patrick%"; mysql> select * from tbl_name where key_col LIKE other_col; @end example In the first statement, the @code{LIKE} value begins with a wild-card character. In the second statement, the @code{LIKE} value is not a constant. @findex IS NULL, and indexes @cindex indexes, and @code{IS NULL} Searching using @code{column_name IS NULL} will use indexes if column_name is an index. MySQL normally uses the index that finds the least number of rows. An index is used for columns that you compare with the following operators: @code{=}, @code{>}, @code{>=}, @code{<}, @code{<=}, @code{BETWEEN}, and a @code{LIKE} with a non-wild-card prefix like @code{'something%'}. Any index that doesn't span all @code{AND} levels in the @code{WHERE} clause is not used to optimize the query. In other words: To be able to use an index, a prefix of the index must be used in every @code{AND} group. The following @code{WHERE} clauses use indexes: @example ... WHERE index_part1=1 AND index_part2=2 AND other_column=3 ... WHERE index=1 OR A=10 AND index=2 /* index = 1 OR index = 2 */ ... WHERE index_part1='hello' AND index_part_3=5 /* optimized like "index_part1='hello'" */ ... WHERE index1=1 and index2=2 or index1=3 and index3=3; /* Can use index on index1 but not on index2 or index 3 */ @end example These @code{WHERE} clauses do @strong{NOT} use indexes: @example ... WHERE index_part2=1 AND index_part3=2 /* index_part_1 is not used */ ... WHERE index=1 OR A=10 /* Index is not used in both AND parts */ ... WHERE index_part1=1 OR index_part2=10 /* No index spans all rows */ @end example Note that in some cases MySQL will not use an index, even if one would be available. Some of the cases where this happens are: @itemize @bullet @item If the use of the index would require MySQL to access more than 30 % of the rows in the table. (In this case a table scan is probably much faster, as this will require us to do much fewer seeks). Note that if such a query uses @code{LIMIT} to only retrieve part of the rows, MySQL will use an index anyway, as it can much more quickly find the few rows to return in the result. @end itemize @menu * Indexes:: Column Indexes * Multiple-column indexes:: Multiple-Column Indexes * Table cache:: How MySQL Opens and Closes Tables * Creating many tables:: Drawbacks to Creating Large Numbers of Tables in the Same Database * Open tables:: Why So Many Open tables? @end menu @node Indexes, Multiple-column indexes, MySQL indexes, Optimizing Database Structure @subsection Column Indexes @cindex indexes, columns @cindex columns, indexes @cindex keys All MySQL column types can be indexed. Use of indexes on the relevant columns is the best way to improve the performance of @code{SELECT} operations. The maximum number of keys and the maximum index length is defined per table handler. @xref{Table types}. You can with all table handlers have at least 16 keys and a total index length of at least 256 bytes. For @code{CHAR} and @code{VARCHAR} columns, you can index a prefix of a column. This is much faster and requires less disk space than indexing the whole column. The syntax to use in the @code{CREATE TABLE} statement to index a column prefix looks like this: @example KEY index_name (col_name(length)) @end example The example below creates an index for the first 10 characters of the @code{name} column: @example mysql> CREATE TABLE test ( name CHAR(200) NOT NULL, KEY index_name (name(10))); @end example For @code{BLOB} and @code{TEXT} columns, you must index a prefix of the column. You cannot index the entire column. In MySQL Version 3.23.23 or later, you can also create special @strong{FULLTEXT} indexes. They are used for full-text search. Only the @code{MyISAM} table type supports @code{FULLTEXT} indexes. They can be created only from @code{VARCHAR} and @code{TEXT} columns. Indexing always happens over the entire column and partial indexing is not supported. See @ref{Fulltext Search} for details. @node Multiple-column indexes, Table cache, Indexes, Optimizing Database Structure @subsection Multiple-Column Indexes @cindex multi-column indexes @cindex indexes, multi-column @cindex keys, multi-column MySQL can create indexes on multiple columns. An index may consist of up to 15 columns. (On @code{CHAR} and @code{VARCHAR} columns you can also use a prefix of the column as a part of an index). A multiple-column index can be considered a sorted array containing values that are created by concatenating the values of the indexed columns. MySQL uses multiple-column indexes in such a way that queries are fast when you specify a known quantity for the first column of the index in a @code{WHERE} clause, even if you don't specify values for the other columns. Suppose a table is created using the following specification: @example mysql> CREATE TABLE test ( id INT NOT NULL, last_name CHAR(30) NOT NULL, first_name CHAR(30) NOT NULL, PRIMARY KEY (id), INDEX name (last_name,first_name)); @end example Then the index @code{name} is an index over @code{last_name} and @code{first_name}. The index will be used for queries that specify values in a known range for @code{last_name}, or for both @code{last_name} and @code{first_name}. Therefore, the @code{name} index will be used in the following queries: @example mysql> SELECT * FROM test WHERE last_name="Widenius"; mysql> SELECT * FROM test WHERE last_name="Widenius" AND first_name="Michael"; mysql> SELECT * FROM test WHERE last_name="Widenius" AND (first_name="Michael" OR first_name="Monty"); mysql> SELECT * FROM test WHERE last_name="Widenius" AND first_name >="M" AND first_name < "N"; @end example However, the @code{name} index will NOT be used in the following queries: @example mysql> SELECT * FROM test WHERE first_name="Michael"; mysql> SELECT * FROM test WHERE last_name="Widenius" OR first_name="Michael"; @end example For more information on the manner in which MySQL uses indexes to improve query performance, see @ref{MySQL indexes, , MySQL indexes}. @node Table cache, Creating many tables, Multiple-column indexes, Optimizing Database Structure @subsection How MySQL Opens and Closes Tables @findex table_cache @cindex tables, opening @cindex tables, closing @cindex opening, tables @cindex closing, tables @cindex table cache @code{table_cache}, @code{max_connections}, and @code{max_tmp_tables} affect the maximum number of files the server keeps open. If you increase one or both of these values, you may run up against a limit imposed by your operating system on the per-process number of open file descriptors. However, you can increase the limit on many systems. Consult your OS documentation to find out how to do this, because the method for changing the limit varies widely from system to system. @code{table_cache} is related to @code{max_connections}. For example, for 200 concurrent running connections, you should have a table cache of at least @code{200 * n}, where @code{n} is the maximum number of tables in a join. You also need to reserve some extra file descriptors for temporary tables and files. The cache of open tables can grow to a maximum of @code{table_cache} (default 64; this can be changed with the @code{-O table_cache=#} option to @code{mysqld}). A table is never closed, except when the cache is full and another thread tries to open a table or if you use @code{mysqladmin refresh} or @code{mysqladmin flush-tables}. When the table cache fills up, the server uses the following procedure to locate a cache entry to use: @itemize @bullet @item Tables that are not currently in use are released, in least-recently-used order. @item If the cache is full and no tables can be released, but a new table needs to be opened, the cache is temporarily extended as necessary. @item If the cache is in a temporarily-extended state and a table goes from in-use to not-in-use state, the table is closed and released from the cache. @end itemize A table is opened for each concurrent access. This means that if you have two threads accessing the same table or access the table twice in the same query (with @code{AS}) the table needs to be opened twice. The first open of any table takes two file descriptors; each additional use of the table takes only one file descriptor. The extra descriptor for the first open is used for the index file; this descriptor is shared among all threads. You can check if your table cache is too small by checking the mysqld variable @code{opened_tables}. If this is quite big, even if you haven't done a lot of @code{FLUSH TABLES}, you should increase your table cache. @xref{SHOW STATUS}. @node Creating many tables, Open tables, Table cache, Optimizing Database Structure @subsection Drawbacks to Creating Large Numbers of Tables in the Same Database @cindex tables, too many If you have many files in a directory, open, close, and create operations will be slow. If you execute @code{SELECT} statements on many different tables, there will be a little overhead when the table cache is full, because for every table that has to be opened, another must be closed. You can reduce this overhead by making the table cache larger. @node Open tables, , Creating many tables, Optimizing Database Structure @subsection Why So Many Open tables? @cindex tables, open @cindex open tables When you run @code{mysqladmin status}, you'll see something like this: @example Uptime: 426 Running threads: 1 Questions: 11082 Reloads: 1 Open tables: 12 @end example This can be somewhat perplexing if you only have 6 tables. MySQL is multithreaded, so it may have many queries on the same table simultaneously. To minimize the problem with two threads having different states on the same file, the table is opened independently by each concurrent thread. This takes some memory and one extra file descriptor for the data file. The index file descriptor is shared between all threads. @node Optimizing the Server, Disk issues, Optimizing Database Structure, MySQL Optimization @section Optimizing the MySQL Server @menu * System:: System/Compile Time and Startup Parameter Tuning * Server parameters:: Tuning Server Parameters * Compile and link options:: How Compiling and Linking Affects the Speed of MySQL * Memory use:: How MySQL Uses Memory * DNS:: How MySQL uses DNS * SET OPTION:: @code{SET} Syntax @end menu @node System, Server parameters, Optimizing the Server, Optimizing the Server @subsection System/Compile Time and Startup Parameter Tuning @cindex compiling, optimizing @cindex system optimization @cindex startup parameters, tuning We start with the system level things since some of these decisions have to be made very early. In other cases a fast look at this part may suffice because it not that important for the big gains. However, it is always nice to have a feeling about how much one could gain by changing things at this level. The default OS to use is really important! To get the most use of multiple CPU machines one should use Solaris (because the threads works really nice) or Linux (because the 2.2 kernel has really good SMP support). Also on 32-bit machines Linux has a 2G file size limit by default. Hopefully this will be fixed soon when new filesystems are released (XFS/Reiserfs). If you have a desperate need for files bigger than 2G on Linux-intel 32 bit, you should get the LFS patch for the ext2 file system. Because we have not run MySQL in production on that many platforms, we advice you to test your intended platform before choosing it, if possible. @cindex locking Other tips: @itemize @bullet @item If you have enough RAM, you could remove all swap devices. Some operating systems will use a swap device in some contexts even if you have free memory. @item Use the @code{--skip-locking} MySQL option to avoid external locking. Note that this will not impact MySQL's functionality as long as you only run one server. Just remember to take down the server (or lock relevant parts) before you run @code{myisamchk}. On some system this switch is mandatory because the external locking does not work in any case. The @code{--skip-locking} option is on by default when compiling with MIT-pthreads, because @code{flock()} isn't fully supported by MIT-pthreads on all platforms. It's also on default for Linux as Linux file locking are not yet safe. The only case when you can't use @code{--skip-locking} is if you run multiple MySQL @emph{servers} (not clients) on the same data, or run @code{myisamchk} on the table without first flushing and locking the @code{mysqld} server tables first. You can still use @code{LOCK TABLES}/@code{UNLOCK TABLES} even if you are using @code{--skip-locking} @end itemize @node Server parameters, Compile and link options, System, Optimizing the Server @subsection Tuning Server Parameters @cindex parameters, server @cindex @code{mysqld} server, buffer sizes @cindex buffer sizes, @code{mysqld} server @cindex startup parameters You can get the default buffer sizes used by the @code{mysqld} server with this command: @example shell> mysqld --help @end example @cindex @code{mysqld} options @cindex variables, @code{mysqld} This command produces a list of all @code{mysqld} options and configurable variables. The output includes the default values and looks something like this: @example Possible variables for option --set-variable (-O) are: back_log current value: 5 bdb_cache_size current value: 1048540 binlog_cache_size current_value: 32768 connect_timeout current value: 5 delayed_insert_timeout current value: 300 delayed_insert_limit current value: 100 delayed_queue_size current value: 1000 flush_time current value: 0 interactive_timeout current value: 28800 join_buffer_size current value: 131072 key_buffer_size current value: 1048540 lower_case_table_names current value: 0 long_query_time current value: 10 max_allowed_packet current value: 1048576 max_binlog_cache_size current_value: 4294967295 max_connections current value: 100 max_connect_errors current value: 10 max_delayed_threads current value: 20 max_heap_table_size current value: 16777216 max_join_size current value: 4294967295 max_sort_length current value: 1024 max_tmp_tables current value: 32 max_write_lock_count current value: 4294967295 myisam_sort_buffer_size current value: 8388608 net_buffer_length current value: 16384 net_retry_count current value: 10 net_read_timeout current value: 30 net_write_timeout current value: 60 query_buffer_size current value: 0 record_buffer current value: 131072 record_rnd_buffer current value: 131072 slow_launch_time current value: 2 sort_buffer current value: 2097116 table_cache current value: 64 thread_concurrency current value: 10 tmp_table_size current value: 1048576 thread_stack current value: 131072 wait_timeout current value: 28800 @end example If there is a @code{mysqld} server currently running, you can see what values it actually is using for the variables by executing this command: @example shell> mysqladmin variables @end example You can find a full description for all variables in the @code{SHOW VARIABLES} section in this manual. @xref{SHOW VARIABLES}. You can also see some statistics from a running server by issuing the command @code{SHOW STATUS}. @xref{SHOW STATUS}. MySQL uses algorithms that are very scalable, so you can usually run with very little memory. If you, however, give MySQL more memory, you will normally also get better performance. When tuning a MySQL server, the two most important variables to use are @code{key_buffer_size} and @code{table_cache}. You should first feel confident that you have these right before trying to change any of the other variables. If you have much memory (>=256M) and many tables and want maximum performance with a moderate number of clients, you should use something like this: @example shell> safe_mysqld -O key_buffer=64M -O table_cache=256 \ -O sort_buffer=4M -O record_buffer=1M & @end example If you have only 128M and only a few tables, but you still do a lot of sorting, you can use something like: @example shell> safe_mysqld -O key_buffer=16M -O sort_buffer=1M @end example If you have little memory and lots of connections, use something like this: @example shell> safe_mysqld -O key_buffer=512k -O sort_buffer=100k \ -O record_buffer=100k & @end example or even: @example shell> safe_mysqld -O key_buffer=512k -O sort_buffer=16k \ -O table_cache=32 -O record_buffer=8k -O net_buffer=1K & @end example If you are doing a @code{GROUP BY} or @code{ORDER BY} on files that are much bigger than your available memory you should increase the value of @code{record_rnd_buffer} to speed up the reading of rows after the sorting is done. When you have installed MySQL, the @file{support-files} directory will contain some different @code{my.cnf} example files, @file{my-huge.cnf}, @file{my-large.cnf}, @file{my-medium.cnf}, and @file{my-small.cnf}, you can use as a base to optimize your system. If there are very many connections, ``swapping problems'' may occur unless @code{mysqld} has been configured to use very little memory for each connection. @code{mysqld} performs better if you have enough memory for all connections, of course. Note that if you change an option to @code{mysqld}, it remains in effect only for that instance of the server. To see the effects of a parameter change, do something like this: @example shell> mysqld -O key_buffer=32m --help @end example Make sure that the @code{--help} option is last; otherwise, the effect of any options listed after it on the command line will not be reflected in the output. @node Compile and link options, Memory use, Server parameters, Optimizing the Server @subsection How Compiling and Linking Affects the Speed of MySQL @cindex linking, speed @cindex compiling, speed @cindex speed, compiling @cindex speed, linking Most of the following tests are done on Linux with the MySQL benchmarks, but they should give some indication for other operating systems and workloads. You get the fastest executable when you link with @code{-static}. On Linux, you will get the fastest code when compiling with @code{pgcc} and @code{-O3}. To compile @file{sql_yacc.cc} with these options, you need about 200M memory because @code{gcc/pgcc} needs a lot of memory to make all functions inline. You should also set @code{CXX=gcc} when configuring MySQL to avoid inclusion of the @code{libstdc++} library (it is not needed). Note that with some versions of @code{pgcc}, the resulting code will only run on true Pentium processors, even if you use the compiler option that you want the resulting code to be working on all x586 type processors (like AMD). By just using a better compiler and/or better compiler options you can get a 10-30 % speed increase in your application. This is particularly important if you compile the SQL server yourself! We have tested both the Cygnus CodeFusion and Fujitsu compilers, but when we tested them, neither was sufficiently bug free to allow MySQL to be compiled with optimizations on. When you compile MySQL you should only include support for the character sets that you are going to use. (Option @code{--with-charset=xxx}). The standard MySQL binary distributions are compiled with support for all character sets. Here is a list of some measurements that we have done: @itemize @bullet @item If you use @code{pgcc} and compile everything with @code{-O6}, the @code{mysqld} server is 1% faster than with @code{gcc} 2.95.2. @item If you link dynamically (without @code{-static}), the result is 13% slower on Linux. Note that you still can use a dynamic linked MySQL library. It is only the server that is critical for performance. @item If you strip your @code{mysqld} binary with @code{strip libexec/mysqld}, the resulting binary can be up to 4 % faster. @item If you connect using TCP/IP rather than Unix sockets, the result is 7.5% slower on the same computer. (If you are connection to @code{localhost}, MySQL will, by default, use sockets). @item If you connect using TCP/IP from another computer over a 100M Ethernet, things will be 8-11 % slower. @item If you compile with @code{--with-debug=full}, then you will loose 20 % for most queries, but some queries may take substantially longer (The MySQL benchmarks ran 35 % slower) If you use @code{--with-debug}, then you will only loose 15 %. By starting a @code{mysqld} version compiled with @code{--with-debug=full} with @code{--skip-safemalloc} the end result should be close to when configuring with @code{--with-debug}. @item On a Sun SPARCstation 20, SunPro C++ 4.2 is 5 % faster than @code{gcc} 2.95.2. @item Compiling with @code{gcc} 2.95.2 for ultrasparc with the option @code{-mcpu=v8 -Wa,-xarch=v8plusa} gives 4 % more performance. @item On Solaris 2.5.1, MIT-pthreads is 8-12% slower than Solaris native threads on a single processor. With more load/CPUs the difference should get bigger. @item Running with @code{--log-bin} makes @strong{[MySQL} 1 % slower. @item Compiling on Linux-x86 using gcc without frame pointers @code{-fomit-frame-pointer} or @code{-fomit-frame-pointer -ffixed-ebp} @code{mysqld} 1-4% faster. @end itemize The MySQL-Linux distribution provided by MySQL AB used to be compiled with @code{pgcc}, but we had to go back to regular gcc because of a bug in @code{pgcc} that would generate the code that does not run on AMD. We will continue using gcc until that bug is resolved. In the meantime, if you have a non-AMD machine, you can get a faster binary by compiling with @code{pgcc}. The standard MySQL Linux binary is linked statically to get it faster and more portable. @node Memory use, DNS, Compile and link options, Optimizing the Server @subsection How MySQL Uses Memory @cindex memory use The list below indicates some of the ways that the @code{mysqld} server uses memory. Where applicable, the name of the server variable relevant to the memory use is given: @itemize @bullet @item The key buffer (variable @code{key_buffer_size}) is shared by all threads; Other buffers used by the server are allocated as needed. @xref{Server parameters}. @item Each connection uses some thread-specific space: A stack (default 64K, variable @code{thread_stack}), a connection buffer (variable @code{net_buffer_length}), and a result buffer (variable @code{net_buffer_length}). The connection buffer and result buffer are dynamically enlarged up to @code{max_allowed_packet} when needed. When a query is running, a copy of the current query string is also allocated. @item All threads share the same base memory. @item Only the compressed ISAM / MyISAM tables are memory mapped. This is because the 32-bit memory space of 4GB is not large enough for most big tables. When systems with a 64-bit address space become more common we may add general support for memory mapping. @item Each request doing a sequential scan over a table allocates a read buffer (variable @code{record_buffer}). @item When reading rows in 'random' order (for example after a sort) a random-read buffer is allocated to avoid disk seeks. (variable @code{record_rnd_buffer}). @item All joins are done in one pass, and most joins can be done without even using a temporary table. Most temporary tables are memory-based (HEAP) tables. Temporary tables with a big record length (calculated as the sum of all column lengths) or that contain @code{BLOB} columns are stored on disk. One problem in MySQL versions before Version 3.23.2 is that if a HEAP table exceeds the size of @code{tmp_table_size}, you get the error @code{The table tbl_name is full}. In newer versions this is handled by automatically changing the in-memory (HEAP) table to a disk-based (MyISAM) table as necessary. To work around this problem, you can increase the temporary table size by setting the @code{tmp_table_size} option to @code{mysqld}, or by setting the SQL option @code{SQL_BIG_TABLES} in the client program. @xref{SET OPTION, , @code{SET OPTION}}. In MySQL Version 3.20, the maximum size of the temporary table was @code{record_buffer*16}, so if you are using this version, you have to increase the value of @code{record_buffer}. You can also start @code{mysqld} with the @code{--big-tables} option to always store temporary tables on disk. However, this will affect the speed of many complicated queries. @item Most requests doing a sort allocates a sort buffer and 0-2 temporary files depending on the result set size. @xref{Temporary files}. @item Almost all parsing and calculating is done in a local memory store. No memory overhead is needed for small items and the normal slow memory allocation and freeing is avoided. Memory is allocated only for unexpectedly large strings (this is done with @code{malloc()} and @code{free()}). @item Each index file is opened once and the data file is opened once for each concurrently running thread. For each concurrent thread, a table structure, column structures for each column, and a buffer of size @code{3 * n} is allocated (where @code{n} is the maximum row length, not counting @code{BLOB} columns). A @code{BLOB} uses 5 to 8 bytes plus the length of the @code{BLOB} data. The @code{ISAM}/@code{MyISAM} table handlers will use one extra row buffer for internal usage. @item For each table having @code{BLOB} columns, a buffer is enlarged dynamically to read in larger @code{BLOB} values. If you scan a table, a buffer as large as the largest @code{BLOB} value is allocated. @item Table handlers for all in-use tables are saved in a cache and managed as a FIFO. Normally the cache has 64 entries. If a table has been used by two running threads at the same time, the cache contains two entries for the table. @xref{Table cache}. @item A @code{mysqladmin flush-tables} command closes all tables that are not in use and marks all in-use tables to be closed when the currently executing thread finishes. This will effectively free most in-use memory. @end itemize @code{ps} and other system status programs may report that @code{mysqld} uses a lot of memory. This may be caused by thread-stacks on different memory addresses. For example, the Solaris version of @code{ps} counts the unused memory between stacks as used memory. You can verify this by checking available swap with @code{swap -s}. We have tested @code{mysqld} with commercial memory-leakage detectors, so there should be no memory leaks. @node DNS, SET OPTION, Memory use, Optimizing the Server @subsection How MySQL uses DNS @cindex DNS @cindex hostname caching When a new thread connects to @code{mysqld}, @code{mysqld} will span a new thread to handle the request. This thread will first check if the hostname is in the hostname cache. If not the thread will call @code{gethostbyaddr_r()} and @code{gethostbyname_r()} to resolve the hostname. If the operating system doesn't support the above thread-safe calls, the thread will lock a mutex and call @code{gethostbyaddr()} and @code{gethostbyname()} instead. Note that in this case no other thread can resolve other hostnames that is not in the hostname cache until the first thread is ready. You can disable DNS host lookup by starting @code{mysqld} with @code{--skip-name-resolve}. In this case you can however only use IP names in the MySQL privilege tables. If you have a very slow DNS and many hosts, you can get more performance by either disabling DNS lookop with @code{--skip-name-resolve} or by increasing the @code{HOST_CACHE_SIZE} define (default: 128) and recompile @code{mysqld}. You can disable the hostname cache with @code{--skip-host-cache}. You can clear the hostname cache with @code{FLUSH HOSTS} or @code{mysqladmin flush-hosts}. If you don't want to allow connections over @code{TCP/IP}, you can do this by starting @code{mysqld} with @code{--skip-networking}. @node SET OPTION, , DNS, Optimizing the Server @subsection @code{SET} Syntax @findex SET OPTION @example SET [OPTION] SQL_VALUE_OPTION= value, ... @end example @code{SET OPTION} sets various options that affect the operation of the server or your client. Any option you set remains in effect until the current session ends, or until you set the option to a different value. @table @code @item CHARACTER SET character_set_name | DEFAULT This maps all strings from and to the client with the given mapping. Currently the only option for @code{character_set_name} is @code{cp1251_koi8}, but you can easily add new mappings by editing the @file{sql/convert.cc} file in the MySQL source distribution. The default mapping can be restored by using a @code{character_set_name} value of @code{DEFAULT}. Note that the syntax for setting the @code{CHARACTER SET} option differs from the syntax for setting the other options. @item PASSWORD = PASSWORD('some password') @cindex passwords, setting Set the password for the current user. Any non-anonymous user can change his own password! @item PASSWORD FOR user = PASSWORD('some password') Set the password for a specific user on the current server host. Only a user with access to the @code{mysql} database can do this. The user should be given in @code{user@@hostname} format, where @code{user} and @code{hostname} are exactly as they are listed in the @code{User} and @code{Host} columns of the @code{mysql.user} table entry. For example, if you had an entry with @code{User} and @code{Host} fields of @code{'bob'} and @code{'%.loc.gov'}, you would write: @example mysql> SET PASSWORD FOR bob@@"%.loc.gov" = PASSWORD("newpass"); or mysql> UPDATE mysql.user SET password=PASSWORD("newpass") where user="bob' and host="%.loc.gov"; @end example @item SQL_AUTO_IS_NULL = 0 | 1 If set to @code{1} (default) then one can find the last inserted row for a table with an auto_increment row with the following construct: @code{WHERE auto_increment_column IS NULL}. This is used by some ODBC programs like Access. @item AUTOCOMMIT= 0 | 1 If set to @code{1} all changes to a table will be done at once. To start a multi-command transaction, you have to use the @code{BEGIN} statement. @xref{COMMIT}. If set to @code{0} you have to use @code{COMMIT} / @code{ROLLBACK} to accept/revoke that transaction. @xref{COMMIT}. Note that when you change from not @code{AUTOCOMMIT} mode to @code{AUTOCOMMIT} mode, MySQL will do an automatic @code{COMMIT} on any open transactions. @item SQL_BIG_TABLES = 0 | 1 @cindex table is full If set to @code{1}, all temporary tables are stored on disk rather than in memory. This will be a little slower, but you will not get the error @code{The table tbl_name is full} for big @code{SELECT} operations that require a large temporary table. The default value for a new connection is @code{0} (that is, use in-memory temporary tables). @item SQL_BIG_SELECTS = 0 | 1 If set to @code{0}, MySQL will abort if a @code{SELECT} is attempted that probably will take a very long time. This is useful when an inadvisable @code{WHERE} statement has been issued. A big query is defined as a @code{SELECT} that probably will have to examine more than @code{max_join_size} rows. The default value for a new connection is @code{1} (which will allow all @code{SELECT} statements). @item SQL_BUFFER_RESULT = 0 | 1 @code{SQL_BUFFER_RESULT} will force the result from @code{SELECT}'s to be put into a temporary table. This will help MySQL free the table locks early and will help in cases where it takes a long time to send the result set to the client. @item SQL_LOW_PRIORITY_UPDATES = 0 | 1 If set to @code{1}, all @code{INSERT}, @code{UPDATE}, @code{DELETE}, and and @code{LOCK TABLE WRITE} statements wait until there is no pending @code{SELECT} or @code{LOCK TABLE READ} on the affected table. @item SQL_MAX_JOIN_SIZE = value | DEFAULT Don't allow @code{SELECT}s that will probably need to examine more than @code{value} row combinations. By setting this value, you can catch @code{SELECT}s where keys are not used properly and that would probably take a long time. Setting this to a value other than @code{DEFAULT} will reset the @code{SQL_BIG_SELECTS} flag. If you set the @code{SQL_BIG_SELECTS} flag again, the @code{SQL_MAX_JOIN_SIZE} variable will be ignored. You can set a default value for this variable by starting @code{mysqld} with @code{-O max_join_size=#}. @item SQL_SAFE_UPDATES = 0 | 1 If set to @code{1}, MySQL will abort if an @code{UPDATE} or @code{DELETE} is attempted that doesn't use a key or @code{LIMIT} in the @code{WHERE} clause. This makes it possible to catch wrong updates when creating SQL commands by hand. @item SQL_SELECT_LIMIT = value | DEFAULT The maximum number of records to return from @code{SELECT} statements. If a @code{SELECT} has a @code{LIMIT} clause, the @code{LIMIT} takes precedence over the value of @code{SQL_SELECT_LIMIT}. The default value for a new connection is ``unlimited.'' If you have changed the limit, the default value can be restored by using a @code{SQL_SELECT_LIMIT} value of @code{DEFAULT}. @item SQL_LOG_OFF = 0 | 1 If set to @code{1}, no logging will be done to the standard log for this client, if the client has the @strong{process} privilege. This does not affect the update log! @item SQL_LOG_UPDATE = 0 | 1 If set to @code{0}, no logging will be done to the update log for the client, if the client has the @strong{process} privilege. This does not affect the standard log! @item SQL_QUOTE_SHOW_CREATE = 0 | 1 If set to @code{1}, @code{SHOW CREATE TABLE} will quote table and column names. This is @strong{on} by default, for replication of tables with fancy column names to work. @ref{SHOW CREATE TABLE, , @code{SHOW CREATE TABLE}}. @item TIMESTAMP = timestamp_value | DEFAULT Set the time for this client. This is used to get the original timestamp if you use the update log to restore rows. @code{timestamp_value} should be a UNIX Epoch timestamp, not a MySQL timestamp. @item LAST_INSERT_ID = # Set the value to be returned from @code{LAST_INSERT_ID()}. This is stored in the update log when you use @code{LAST_INSERT_ID()} in a command that updates a table. @item INSERT_ID = # Set the value to be used by the following @code{INSERT} or @code{ALTER TABLE} command when inserting an @code{AUTO_INCREMENT} value. This is mainly used with the update log. @end table @menu * SET TRANSACTION:: @code{SET TRANSACTION} Syntax @end menu @node Disk issues, , Optimizing the Server, MySQL Optimization @section Disk Issues @cindex disk issues @cindex performance, disk issues @itemize @bullet @item As mentioned before, disks seeks are a big performance bottleneck. This problems gets more and more apparent when the data starts to grow so large that effective caching becomes impossible. For large databases, where you access data more or less randomly, you can be sure that you will need at least one disk seek to read and a couple of disk seeks to write things. To minimize this problem, use disks with low seek times. @item Increase the number of available disk spindles (and thereby reduce the seek overhead) by either symlink files to different disks or striping the disks. @table @strong @item Using symbolic links This means that you symlink the index and/or data file(s) from the normal data directory to another disk (that may also be striped). This makes both the seek and read times better (if the disks are not used for other things). @xref{Symbolic links}. @cindex striping, defined @item Striping Striping means that you have many disks and put the first block on the first disk, the second block on the second disk, and the Nth on the (N mod number_of_disks) disk, and so on. This means if your normal data size is less than the stripe size (or perfectly aligned) you will get much better performance. Note that striping is very dependent on the OS and stripe-size. So benchmark your application with different stripe-sizes. @xref{Custom Benchmarks}. Note that the speed difference for striping is @strong{very} dependent on the parameters. Depending on how you set the striping parameters and number of disks you may get a difference in orders of magnitude. Note that you have to choose to optimize for random or sequential access. @end table @item For reliability you may want to use RAID 0+1 (striping + mirroring), but in this case you will need 2*N drives to hold N drives of data. This is probably the best option if you have the money for it! You may, however, also have to invest in some volume-management software to handle it efficiently. @item A good option is to have semi-important data (that can be regenerated) on RAID 0 disk while storing really important data (like host information and logs) on a RAID 0+1 or RAID N disk. RAID N can be a problem if you have many writes because of the time to update the parity bits. @item You may also set the parameters for the file system that the database uses. One easy change is to mount the file system with the noatime option. That makes it skip the updating of the last access time in the inode and by this will avoid some disk seeks. @item On Linux, you can get much more performance (up to 100 % under load is not uncommon) by using hdpram to configure your disk's interface! The following should be quite good hdparm options for MySQL (and probably many other applications): @example hdparm -m 16 -d 1 @end example Note that the performance/reliability when using the above depends on your hardware, so we strongly suggest that you test your system thoroughly after using @code{hdparm}! Please consult the @code{hdparm} man page for more information! If @code{hdparm} is not used wisely, filesystem corruption may result. Backup everything before experimenting! @item On many operating systems you can mount the disks with the 'async' flag to set the file system to be updated asynchronously. If your computer is reasonable stable, this should give you more performance without sacrificing too much reliability. (This flag is on by default on Linux.) @item If you don't need to know when a file was last accessed (which is not really useful on a database server), you can mount your file systems with the noatime flag. @end itemize @menu * Symbolic links:: Using Symbolic Links @end menu @node Symbolic links, , Disk issues, Disk issues @subsection Using Symbolic Links @cindex symbolic links @cindex links, symbolic You can move tables and databases from the database directory to other locations and replace them with symbolic links to the new locations. You might want to do this, for example, to move a database to a file system with more free space or increase the speed of your system by spreading your tables to different disk. The recommended may to do this, is to just symlink databases to different disk and only symlink tables as a last resort. @cindex databases, symbolic links @menu * Symbolic links to databases:: Using Symbolic Links for Databases * Symbolic links to tables:: Using Symbolic Links for Tables @end menu @node Symbolic links to databases, Symbolic links to tables, Symbolic links, Symbolic links @subsubsection Using Symbolic Links for Databases The way to symlink a database is to first create a directory on some disk where you have free space and then create a symlink to it from the MySQL database directory. @example shell> mkdir /dr1/databases/test shell> ln -s /dr1/databases/test mysqld-datadir @end example MySQL doesn't support that you link one directory to multiple databases. Replacing a database directory with a symbolic link will work fine as long as you don't make a symbolic link between databases. Suppose you have a database @code{db1} under the MySQL data directory, and then make a symlink @code{db2} that points to @code{db1}: @example shell> cd /path/to/datadir shell> ln -s db1 db2 @end example Now, for any table @code{tbl_a} in @code{db1}, there also appears to be a table @code{tbl_a} in @code{db2}. If one thread updates @code{db1.tbl_a} and another thread updates @code{db2.tbl_a}, there will be problems. If you really need this, you must change the following code in @file{mysys/mf_format.c}: @example if (flag & 32 || (!lstat(to,&stat_buff) && S_ISLNK(stat_buff.st_mode))) @end example to @example if (1) @end example On Windows you can use internal symbolic links to directories by compiling MySQL with @code{-DUSE_SYMDIR}. This allows you to put different databases on different disks. @xref{Windows symbolic links}. @node Symbolic links to tables, , Symbolic links to databases, Symbolic links @subsubsection Using Symbolic Links for Tables @cindex databases, symbolic links Before MySQL 4.0 you should not symlink tables, if you are not very carefully with them. The problem is that if you run @code{ALTER TABLE}, @code{REPAIR TABLE} or @code{OPTIMIZE TABLE} on a symlinked table, the symlinks will be removed and replaced by the original files. This happens because the above command works by creating a temporary file in the database directory and when the command is complete, replace the original file with the temporary file. You should not symlink tables on system that doesn't have a fully working @code{realpath()} call. (At least Linux and Solaris support @code{realpath()}) In MySQL 4.0 symlinks is only fully supported for @code{MyISAM} tables. For other table types you will probably get strange problems when doing any of the above mentioned commands. The handling of symbolic links in MySQL 4.0 works the following way (this is mostly relevant only for @code{MyISAM} tables). @itemize @bullet @item In the data directory you will always have the table definition file and the data/index files. @item You can symlink the index file and the data file to different directories independent of the other. @item The symlinking can be done from the operating system (if @code{mysqld} is not running) or with the @code{INDEX/DATA DIRECTORY="path-to-dir"} command in @code{CREATE TABLE}. @xref{CREATE TABLE}. @item @code{myisamchk} will not replace a symlink with the index/file but work directly on the files the symlinks points to. Any temporary files will be created in the same directory where the data/index file is. @item When you drop a table that is using symlinks, both the symlink and the file the symlink points to is dropped. This is a good reason to why you should NOT run @code{mysqld} as root and not allow persons to have write access to the MySQL database directories. @item If you rename a table with @code{ALTER TABLE RENAME} and you don't change database, the symlink in the database directory will be renamed to the new name and the data/index file will be renamed accordingly. @item If you use @code{ALTER TABLE RENAME} to move a table to another database, then the table will be moved to the other database directory and the old symlinks and the files they pointed to will be deleted. @item If you are not using symlinks you should use the @code{--skip-symlink} option to @code{mysqld} to ensure that no one can drop or rename a file outside of the @code{mysqld} data directory. @end itemize Things that are not yet supported: @cindex TODO, symlinks @itemize @bullet @item @code{ALTER TABLE} ignores all @code{INDEX/DATA DIRECTORY="path"} options. @item @code{CREATE TABLE} doesn't report if the table has symbolic links. @item @code{mysqldump} doesn't include the symbolic links information in the output. @item @code{BACKUP TABLE} and @code{RESTORE TABLE} don't respect symbolic links. @end itemize @node Reference, Table types, MySQL Optimization, Top @chapter MySQL Language Reference @menu * Language Structure:: Language Structure * Column types:: Column types * Functions:: Functions * Data Manipulation:: Data Manipulation: @code{SELECT}, @code{INSERT}, @code{UPDATE}, @code{DELETE} * Data Definition:: Data Definition: @code{CREATE}, @code{DROP}, @code{ALTER} * Basic User Commands:: Basic MySQL User Utility Commands * Transactional Commands:: MySQL Transactional and Locking Commands * Fulltext Search:: MySQL Full-text Search @end menu MySQL has a very complex, but intuitive and easy to learn SQL interface. This chapter describes the various commands, types, and functions you will need to know in order to use MySQL efficiently and effectively. This chapter also serves as a reference to all functionality included in MySQL. In order to use this chapter effectively, you may find it useful to refer to the various indexes. @node Language Structure, Column types, Reference, Reference @section Language Structure @menu * Literals:: Literals: How to Write Strings and Numbers * Legal names:: Database, Table, Index, Column, and Alias Names * Name case sensitivity:: Case Sensitivity in Names * Variables:: User Variables * Comments:: Comment Syntax * Reserved words:: Is MySQL Picky About Reserved Words? @end menu @node Literals, Legal names, Language Structure, Language Structure @subsection Literals: How to Write Strings and Numbers @cindex strings, defined @cindex strings, escaping characters @cindex literals @cindex escape characters @cindex backslash, escape character @menu * String syntax:: Strings * Number syntax:: Numbers * Hexadecimal values:: Hexadecimal values * NULL values:: @code{NULL} values @end menu This section describes the various ways to write strings and numbers in MySQL. It also covers the various nuances and ``gotchas'' that you may run into when dealing with these basic types in MySQL. @node String syntax, Number syntax, Literals, Literals @subsubsection Strings A string is a sequence of characters, surrounded by either single quote (@samp{'}) or double quote (@samp{"}) characters (only the single quote if you run in ANSI mode). Examples: @example 'a string' "another string" @end example Within a string, certain sequences have special meaning. Each of these sequences begins with a backslash (@samp{\}), known as the @emph{escape character}. MySQL recognizes the following escape sequences: @c these aren't really functions, but that's probably the most reasonable index @table @code @findex \0 (ASCII 0) @findex NUL @item \0 An ASCII 0 (@code{NUL}) character. @findex \' (single quote) @findex single quote (\') @item \' A single quote (@samp{'}) character. @findex \" (double quote) @findex double quote (\") @item \" A double quote (@samp{"}) character. @findex \b (backspace) @findex backspace (\b) @item \b A backspace character. @findex \n (newline) @findex newline (\n) @item \n A newline character. @findex \r (carriage return) @findex return (\r) @findex carriage return (\r) @item \r A carriage return character. @findex \t (tab) @findex tab (\t) @item \t A tab character. @findex \z (Control-Z) ASCII(26) @findex (Control-Z) \z @item \z ASCII(26) (Control-Z). This character can be encoded to allow you to go around the problem that ASCII(26) stands for END-OF-FILE on Windows. (ASCII(26) will cause problems if you try to use @code{mysql database < filename}). @findex \\ (escape) @findex escape (\\) @item \\ A backslash (@samp{\}) character. @findex % (wild card character) @findex Wild card character (%) @item \% A @samp{%} character. This is used to search for literal instances of @samp{%} in contexts where @samp{%} would otherwise be interpreted as a wild-card character. @xref{String comparison functions}. @findex _ (wild card character) @findex Wild card character (_) @item \_ A @samp{_} character. This is used to search for literal instances of @samp{_} in contexts where @samp{_} would otherwise be interpreted as a wild-card character. @xref{String comparison functions}. @end table Note that if you use @samp{\%} or @samp{\_} in some string contexts, these will return the strings @samp{\%} and @samp{\_} and not @samp{%} and @samp{_}. @cindex quotes, in strings @noindent There are several ways to include quotes within a string: @itemize @bullet @item A @samp{'} inside a string quoted with @samp{'} may be written as @samp{''}. @item A @samp{"} inside a string quoted with @samp{"} may be written as @samp{""}. @item You can precede the quote character with an escape character (@samp{\}). @item A @samp{'} inside a string quoted with @samp{"} needs no special treatment and need not be doubled or escaped. In the same way, @samp{"} inside a string quoted with @samp{'} needs no special treatment. @end itemize The @code{SELECT} statements shown below demonstrate how quoting and escaping work: @example mysql> SELECT 'hello', '"hello"', '""hello""', 'hel''lo', '\'hello'; +-------+---------+-----------+--------+--------+ | hello | "hello" | ""hello"" | hel'lo | 'hello | +-------+---------+-----------+--------+--------+ mysql> SELECT "hello", "'hello'", "''hello''", "hel""lo", "\"hello"; +-------+---------+-----------+--------+--------+ | hello | 'hello' | ''hello'' | hel"lo | "hello | +-------+---------+-----------+--------+--------+ mysql> SELECT "This\nIs\nFour\nlines"; +--------------------+ | This Is Four lines | +--------------------+ @end example @cindex quoting binary data If you want to insert binary data into a @code{BLOB} column, the following characters must be represented by escape sequences: @table @code @item NUL ASCII 0. You should represent this by @samp{\0} (a backslash and an ASCII @samp{0} character). @item \ ASCII 92, backslash. Represent this by @samp{\\}. @item ' ASCII 39, single quote. Represent this by @samp{\'}. @item " ASCII 34, double quote. Represent this by @samp{\"}. @end table @cindex quoting @cindex @code{BLOB}, inserting binary data @findex mysql_escape_string() @findex DBI->quote If you write C code, you can use the C API function @code{mysql_escape_string()} to escape characters for the @code{INSERT} statement. @xref{C API function overview}. In Perl, you can use the @code{quote} method of the @code{DBI} package to convert special characters to the proper escape sequences. @xref{Perl DBI Class, , Perl @code{DBI} Class}. You should use an escape function on any string that might contain any of the special characters listed above! @node Number syntax, Hexadecimal values, String syntax, Literals @subsubsection Numbers @cindex numbers @cindex valid numbers, examples @cindex integers @cindex floats @cindex negative values Integers are represented as a sequence of digits. Floats use @samp{.} as a decimal separator. Either type of number may be preceded by @samp{-} to indicate a negative value. Examples of valid integers: @example 1221 0 -32 @end example Examples of valid floating-point numbers: @example 294.42 -32032.6809e+10 148.00 @end example An integer may be used in a floating-point context; it is interpreted as the equivalent floating-point number. @node Hexadecimal values, NULL values, Number syntax, Literals @subsubsection Hexadecimal Values @tindex hexadecimal values MySQL supports hexadecimal values. In number context these act like an integer (64-bit precision). In string context these act like a binary string where each pair of hex digits is converted to a character: @example mysql> SELECT 0xa+0; -> 10 mysql> select 0x5061756c; -> Paul @end example Hexadecimal strings are often used by ODBC to give values for BLOB columns. @node NULL values, , Hexadecimal values, Literals @subsubsection @code{NULL} Values @tindex NULL value The @code{NULL} value means ``no data'' and is different from values such as @code{0} for numeric types or the empty string for string types. @xref{Problems with NULL, , Problems with @code{NULL}}. @code{NULL} may be represented by @code{\N} when using the text file import or export formats (@code{LOAD DATA INFILE}, @code{SELECT ... INTO OUTFILE}). @xref{LOAD DATA, , @code{LOAD DATA}}. @node Legal names, Name case sensitivity, Literals, Language Structure @subsection Database, Table, Index, Column, and Alias Names @cindex names @cindex legal names @cindex databases, names @cindex tables, names @cindex indexes, names @cindex columns, names @cindex aliases, names @menu * Name case sensitivity:: Case sensitivity in names @end menu Database, table, index, column, and alias names all follow the same rules in MySQL. @tindex identifiers, quoting @tindex quoting of identifiers @tindex ` @tindex " Note that the rules changed starting with MySQL Version 3.23.6 when we introduced quoting of identifiers (database, table, and column names) with @samp{`}. @samp{"} will also work to quote identifiers if you run in ANSI mode. @xref{ANSI mode}. @multitable @columnfractions .15 .15 .70 @item @strong{Identifier} @tab @strong{Max length} @tab @strong{Allowed characters} @item Database @tab 64 @tab Any character that is allowed in a directory name except @samp{/} or @samp{.}. @item Table @tab 64 @tab Any character that is allowed in a file name, except @samp{/} or @samp{.}. @item Column @tab 64 @tab All characters. @item Alias @tab 255 @tab All characters. @end multitable Note that in addition to the above, you can't have ASCII(0) or ASCII(255) or the quoting character in an identifier. Note that if the identifier is a restricted word or contains special characters you must always quote it with @code{`} when you use it: @example SELECT * from `select` where `select`.id > 100; @end example In previous versions of MySQL, the name rules are as follows: @itemize @bullet @item A name may consist of alphanumeric characters from the current character set and also @samp{_} and @samp{$}. The default character set is ISO-8859-1 Latin1; this may be changed with the @code{--default-character-set} option to @code{mysqld}. @xref{Character sets}. @item A name may start with any character that is legal in a name. In particular, a name may start with a number (this differs from many other database systems!). However, a name cannot consist @emph{only} of numbers. @item You cannot use the @samp{.} character in names because it is used to extend the format by which you can refer to columns (see immediately below). @end itemize It is recommended that you do not use names like @code{1e}, because an expression like @code{1e+1} is ambiguous. It may be interpreted as the expression @code{1e + 1} or as the number @code{1e+1}. In MySQL you can refer to a column using any of the following forms: @multitable @columnfractions .35 .65 @item @strong{Column reference} @tab @strong{Meaning} @item @code{col_name} @tab Column @code{col_name} from whichever table used in the query contains a column of that name. @item @code{tbl_name.col_name} @tab Column @code{col_name} from table @code{tbl_name} of the current database. @item @code{db_name.tbl_name.col_name} @tab Column @code{col_name} from table @code{tbl_name} of the database @code{db_name}. This form is available in MySQL Version 3.22 or later. @item @code{`column_name`} @tab A column that is a keyword or contains special characters. @end multitable You need not specify a @code{tbl_name} or @code{db_name.tbl_name} prefix for a column reference in a statement unless the reference would be ambiguous. For example, suppose tables @code{t1} and @code{t2} each contain a column @code{c}, and you retrieve @code{c} in a @code{SELECT} statement that uses both @code{t1} and @code{t2}. In this case, @code{c} is ambiguous because it is not unique among the tables used in the statement, so you must indicate which table you mean by writing @code{t1.c} or @code{t2.c}. Similarly, if you are retrieving from a table @code{t} in database @code{db1} and from a table @code{t} in database @code{db2}, you must refer to columns in those tables as @code{db1.t.col_name} and @code{db2.t.col_name}. @cindex ODBC compatibility @cindex compatibility, with ODBC The syntax @code{.tbl_name} means the table @code{tbl_name} in the current database. This syntax is accepted for ODBC compatibility, because some ODBC programs prefix table names with a @samp{.} character. @node Name case sensitivity, Variables, Legal names, Language Structure @subsection Case Sensitivity in Names @cindex names, case-sensitivity @cindex case-sensitivity, in names @cindex database names, case sensitivity @cindex table names, case sensitivity @cindex column names, case sensitivity @cindex alias names, case sensitivity In MySQL, databases and tables correspond to directories and files within those directories. Consequently, the case sensitivity of the underlying operating system determines the case sensitivity of database and table names. This means database and table names are case sensitive in Unix and case insensitive in Windows. @xref{Extensions to ANSI}. @strong{NOTE:} Although database and table names are case insensitive for Windows, you should not refer to a given database or table using different cases within the same query. The following query would not work because it refers to a table both as @code{my_table} and as @code{MY_TABLE}: @example mysql> SELECT * FROM my_table WHERE MY_TABLE.col=1; @end example Column names are case insensitive in all cases. Aliases on tables are case sensitive. The following query would not work because it refers to the alias both as @code{a} and as @code{A}: @example mysql> SELECT col_name FROM tbl_name AS a WHERE a.col_name = 1 OR A.col_name = 2; @end example Aliases on columns are case insensitive. If you have a problem remembering the used cases for a table names, adopt a consistent convention, such as always creating databases and tables using lowercase names. One way to avoid this problem is to start @code{mysqld} with @code{-O lower_case_table_names=1}. By default this option is 1 on Windows and 0 on Unix. If @code{lower_case_table_names} is 1 MySQL will convert all table names to lower case on storage and lookup. Note that if you change this option, you need to first convert your old table names to lower case before starting @code{mysqld}. @node Variables, Comments, Name case sensitivity, Language Structure @subsection User Variables @cindex variables, user @cindex user variables @cindex names, variables MySQL supports thread-specific variables with the @code{@@variablename} syntax. A variable name may consist of alphanumeric characters from the current character set and also @samp{_}, @samp{$}, and @samp{.} . The default character set is ISO-8859-1 Latin1; this may be changed with the @code{--default-character-set} option to @code{mysqld}. @xref{Character sets}. Variables don't have to be initialized. They contain @code{NULL} by default and can store an integer, real, or string value. All variables for a thread are automatically freed when the thread exits. You can set a variable with the @code{SET} syntax: @example SET @@variable= @{ integer expression | real expression | string expression @} [,@@variable= ...]. @end example You can also set a variable in an expression with the @code{@@variable:=expr} syntax: @example select @@t1:=(@@t2:=1)+@@t3:=4,@@t1,@@t2,@@t3; +----------------------+------+------+------+ | @@t1:=(@@t2:=1)+@@t3:=4 | @@t1 | @@t2 | @@t3 | +----------------------+------+------+------+ | 5 | 5 | 1 | 4 | +----------------------+------+------+------+ @end example (We had to use the @code{:=} syntax here, because @code{=} was reserved for comparisons.) User variables may be used where expressions are allowed. Note that this does not currently include use in contexts where a number is explicitly required, such as in the @code{LIMIT} clause of a @code{SELECT} statement, or the @code{IGNORE number LINES} clause of a @code{LOAD DATA} statement. @strong{NOTE:} In a @code{SELECT} statement, each expression is only evaluated when it's sent to the client. This means that in the @code{HAVING}, @code{GROUP BY}, or @code{ORDER BY} clause, you can't refer to an expression that involves variables that are set in the @code{SELECT} part. For example, the following statement will NOT work as expected: @example SELECT (@@aa:=id) AS a, (@@aa+3) AS b FROM table_name HAVING b=5; @end example The reason is that @code{@@aa} will not contain the value of the current row, but the value of @code{id} for the previous accepted row. @menu * Comments:: Comment Syntax * Reserved words:: Is MySQL Picky About Reserved Words? @end menu @node Comments, Reserved words, Variables, Language Structure @subsection Comment Syntax @findex Comment syntax @cindex comments, adding The MySQL server supports the @code{# to end of line}, @code{-- to end of line} and @code{/* in-line or multiple-line */} comment styles: @example mysql> select 1+1; # This comment continues to the end of line mysql> select 1+1; -- This comment continues to the end of line mysql> select 1 /* this is an in-line comment */ + 1; mysql> select 1+ /* this is a multiple-line comment */ 1; @end example Note that the @code{--} comment style requires you to have at least one space after the @code{--}! Although the server understands the comment syntax just described, there are some limitations on the way that the @code{mysql} client parses @code{/* ... */} comments: @itemize @bullet @item Single-quote and double-quote characters are taken to indicate the beginning of a quoted string, even within a comment. If the quote is not matched by a second quote within the comment, the parser doesn't realize the comment has ended. If you are running @code{mysql} interactively, you can tell that it has gotten confused like this because the prompt changes from @code{mysql>} to @code{'>} or @code{">}. @item A semicolon is taken to indicate the end of the current SQL statement and anything following it to indicate the beginning of the next statement. @end itemize These limitations apply both when you run @code{mysql} interactively and when you put commands in a file and tell @code{mysql} to read its input from that file with @code{mysql < some-file}. MySQL doesn't support the @samp{--} ANSI SQL comment style. @xref{Missing comments}. @node Reserved words, , Comments, Language Structure @subsection Is MySQL Picky About Reserved Words? @cindex keywords @cindex reserved words, exceptions A common problem stems from trying to create a table with column names that use the names of datatypes or functions built into MySQL, such as @code{TIMESTAMP} or @code{GROUP}. You're allowed to do it (for example, @code{ABS} is an allowed column name), but whitespace is not allowed between a function name and the @samp{(} when using functions whose names are also column names. The following words are explicitly reserved in MySQL. Most of them are forbidden by ANSI SQL92 as column and/or table names (for example, @code{group}). A few are reserved because MySQL needs them and is (currently) using a @code{yacc} parser: @c This is fixed by including the symbols table from lex.h here and then running @c fix-mysql-reserved-words in emacs (or let David do it): @c (defun fix-mysql-reserved-words () @c (interactive) @c (let ((cnt 0)) @c (insert "\n@item ") @c (while (looking-at "[ \t]*{ +\"\\([^\"]+\\)\"[ \t]*,.*\n") @c (replace-match "@code{\\1}") @c (incf cnt) @c (if (> cnt 3) @c (progn @c (setf cnt 0) @c (insert "\n@item ")) @c (insert " @tab "))))) @c But remove the non alphanumeric entries by hand first. @c Updated after 3.23.4 990928 by David @multitable @columnfractions .25 .25 .25 .25 @item @code{action} @tab @code{add} @tab @code{aggregate} @tab @code{all} @item @code{alter} @tab @code{after} @tab @code{and} @tab @code{as} @item @code{asc} @tab @code{avg} @tab @code{avg_row_length} @tab @code{auto_increment} @item @code{between} @tab @code{bigint} @tab @code{bit} @tab @code{binary} @item @code{blob} @tab @code{bool} @tab @code{both} @tab @code{by} @item @code{cascade} @tab @code{case} @tab @code{char} @tab @code{character} @item @code{change} @tab @code{check} @tab @code{checksum} @tab @code{column} @item @code{columns} @tab @code{comment} @tab @code{constraint} @tab @code{create} @item @code{cross} @tab @code{current_date} @tab @code{current_time} @tab @code{current_timestamp} @item @code{data} @tab @code{database} @tab @code{databases} @tab @code{date} @item @code{datetime} @tab @code{day} @tab @code{day_hour} @tab @code{day_minute} @item @code{day_second} @tab @code{dayofmonth} @tab @code{dayofweek} @tab @code{dayofyear} @item @code{dec} @tab @code{decimal} @tab @code{default} @tab @code{delayed} @item @code{delay_key_write} @tab @code{delete} @tab @code{desc} @tab @code{describe} @item @code{distinct} @tab @code{distinctrow} @tab @code{double} @tab @code{drop} @item @code{end} @tab @code{else} @tab @code{escape} @tab @code{escaped} @item @code{enclosed} @tab @code{enum} @tab @code{explain} @tab @code{exists} @item @code{fields} @tab @code{file} @tab @code{first} @tab @code{float} @item @code{float4} @tab @code{float8} @tab @code{flush} @tab @code{foreign} @item @code{from} @tab @code{for} @tab @code{full} @tab @code{function} @item @code{global} @tab @code{grant} @tab @code{grants} @tab @code{group} @item @code{having} @tab @code{heap} @tab @code{high_priority} @tab @code{hour} @item @code{hour_minute} @tab @code{hour_second} @tab @code{hosts} @tab @code{identified} @item @code{ignore} @tab @code{in} @tab @code{index} @tab @code{infile} @item @code{inner} @tab @code{insert} @tab @code{insert_id} @tab @code{int} @item @code{integer} @tab @code{interval} @tab @code{int1} @tab @code{int2} @item @code{int3} @tab @code{int4} @tab @code{int8} @tab @code{into} @item @code{if} @tab @code{is} @tab @code{isam} @tab @code{join} @item @code{key} @tab @code{keys} @tab @code{kill} @tab @code{last_insert_id} @item @code{leading} @tab @code{left} @tab @code{length} @tab @code{like} @item @code{lines} @tab @code{limit} @tab @code{load} @tab @code{local} @item @code{lock} @tab @code{logs} @tab @code{long} @tab @code{longblob} @item @code{longtext} @tab @code{low_priority} @tab @code{max} @tab @code{max_rows} @item @code{match} @tab @code{mediumblob} @tab @code{mediumtext} @tab @code{mediumint} @item @code{middleint} @tab @code{min_rows} @tab @code{minute} @tab @code{minute_second} @item @code{modify} @tab @code{month} @tab @code{monthname} @tab @code{myisam} @item @code{natural} @tab @code{numeric} @tab @code{no} @tab @code{not} @item @code{null} @tab @code{on} @tab @code{optimize} @tab @code{option} @item @code{optionally} @tab @code{or} @tab @code{order} @tab @code{outer} @item @code{outfile} @tab @code{pack_keys} @tab @code{partial} @tab @code{password} @item @code{precision} @tab @code{primary} @tab @code{procedure} @tab @code{process} @item @code{processlist} @tab @code{privileges} @tab @code{read} @tab @code{real} @item @code{references} @tab @code{reload} @tab @code{regexp} @tab @code{rename} @item @code{replace} @tab @code{restrict} @tab @code{returns} @tab @code{revoke} @item @code{rlike} @tab @code{row} @tab @code{rows} @tab @code{second} @item @code{select} @tab @code{set} @tab @code{show} @tab @code{shutdown} @item @code{smallint} @tab @code{soname} @tab @code{sql_big_tables} @tab @code{sql_big_selects} @item @code{sql_low_priority_updates} @tab @code{sql_log_off} @tab @code{sql_log_update} @tab @code{sql_select_limit} @item @code{sql_small_result} @tab @code{sql_big_result} @tab @code{sql_warnings} @tab @code{straight_join} @item @code{starting} @tab @code{status} @tab @code{string} @tab @code{table} @item @code{tables} @tab @code{temporary} @tab @code{terminated} @tab @code{text} @item @code{then} @tab @code{time} @tab @code{timestamp} @tab @code{tinyblob} @item @code{tinytext} @tab @code{tinyint} @tab @code{trailing} @tab @code{to} @item @code{type} @tab @code{use} @tab @code{using} @tab @code{unique} @item @code{unlock} @tab @code{unsigned} @tab @code{update} @tab @code{usage} @item @code{values} @tab @code{varchar} @tab @code{variables} @tab @code{varying} @item @code{varbinary} @tab @code{with} @tab @code{write} @tab @code{when} @item @code{where} @tab @code{year} @tab @code{year_month} @tab @code{zerofill} @end multitable The following symbols (from the table above) are disallowed by ANSI SQL but allowed by MySQL as column/table names. This is because some of these names are very natural names and a lot of people have already used them. @itemize @bullet @item @code{ACTION} @item @code{BIT} @item @code{DATE} @item @code{ENUM} @item @code{NO} @item @code{TEXT} @item @code{TIME} @item @code{TIMESTAMP} @end itemize @node Column types, Functions, Language Structure, Reference @section Column Types @cindex columns, types @cindex types, columns MySQL supports a number of column types, which may be grouped into three categories: numeric types, date and time types, and string (character) types. This section first gives an overview of the types available and summarizes the storage requirements for each column type, then provides a more detailed description of the properties of the types in each category. The overview is intentionally brief. The more detailed descriptions should be consulted for additional information about particular column types, such as the allowable formats in which you can specify values. The column types supported by MySQL are listed below. The following code letters are used in the descriptions: @cindex display size @cindex sizes, display @cindex digits @cindex decimal point @cindex brackets, square @cindex square brackets @table @code @item M Indicates the maximum display size. The maximum legal display size is 255. @item D Applies to floating-point types and indicates the number of digits following the decimal point. The maximum possible value is 30, but should be no greater than @code{M}-2. @end table Square brackets (@samp{[} and @samp{]}) indicate parts of type specifiers that are optional. @tindex Types @c The @w{-number} stuff keeps a linebreak from occurring between @c the - and number. Note that if you specify @code{ZEROFILL} for a column, MySQL will automatically add the @code{UNSIGNED} attribute to the column. @table @code @tindex TINYINT @item TINYINT[(M)] [UNSIGNED] [ZEROFILL] A very small integer. The signed range is @code{-128} to @code{127}. The unsigned range is @code{0} to @code{255}. @tindex SMALLINT @item SMALLINT[(M)] [UNSIGNED] [ZEROFILL] A small integer. The signed range is @code{-32768} to @code{32767}. The unsigned range is @code{0} to @code{65535}. @tindex MEDIUMINT @item MEDIUMINT[(M)] [UNSIGNED] [ZEROFILL] A medium-size integer. The signed range is @code{-8388608} to @code{8388607}. The unsigned range is @code{0} to @code{16777215}. @tindex INT @item INT[(M)] [UNSIGNED] [ZEROFILL] A normal-size integer. The signed range is @code{-2147483648} to @code{2147483647}. The unsigned range is @code{0} to @code{4294967295}. @tindex INTEGER @item INTEGER[(M)] [UNSIGNED] [ZEROFILL] This is a synonym for @code{INT}. @tindex BIGINT @item BIGINT[(M)] [UNSIGNED] [ZEROFILL] A large integer. The signed range is @code{-9223372036854775808} to @code{9223372036854775807}. The unsigned range is @code{0} to @code{18446744073709551615}. Some things you should be aware about @code{BIGINT} columns: @itemize @bullet @item @cindex rounding errors As all arithmetic is done using signed @code{BIGINT} or @code{DOUBLE} values, so you shouldn't use unsigned big integers larger than @code{9223372036854775807} (63 bits) except with bit functions! If you do that, some of the last digits in the result may be wrong because of rounding errors when converting the @code{BIGINT} to a @code{DOUBLE}. @item You can always store an exact integer value in a @code{BIGINT} column by storing it as a string, as there is in this case there will be no intermediate double representation. @item @samp{-}, @samp{+}, and @samp{*} will use @code{BIGINT} arithmetic when both arguments are @code{INTEGER} values! This means that if you multiply two big integers (or results from functions that return integers) you may get unexpected results when the result is larger than @code{9223372036854775807}. @end itemize @cindex floating-point number @tindex FLOAT @tindex FLOAT(precision) @item FLOAT(precision) [ZEROFILL] A floating-point number. Cannot be unsigned. @code{precision} can be @code{<=24} for a single-precision floating-point number and between 25 and 53 for a double-precision floating-point number. These types are like the @code{FLOAT} and @code{DOUBLE} types described immediately below. @code{FLOAT(X)} has the same range as the corresponding @code{FLOAT} and @code{DOUBLE} types, but the display size and number of decimals is undefined. In MySQL Version 3.23, this is a true floating-point value. In earlier MySQL versions, @code{FLOAT(precision)} always has 2 decimals. Note that using @code{FLOAT} may give you some unexpected problems as all calculation in MySQL is done with double precision. @xref{No matching rows}. @cindex ODBC compatibility @cindex compatibility, with ODBC This syntax is provided for ODBC compatibility. @tindex FLOAT @tindex FLOAT(M,D) @item FLOAT[(M,D)] [ZEROFILL] A small (single-precision) floating-point number. Cannot be unsigned. Allowable values are @code{@w{-3.402823466E+38}} to @code{@w{-1.175494351E-38}}, @code{0}, and @code{@w{1.175494351E-38}} to @code{3.402823466E+38}. The M is the display width and D is the number of decimals. @code{FLOAT} without an argument or with an argument of <= 24 stands for a single-precision floating-point number. @tindex DOUBLE @tindex FLOAT(precision) @item DOUBLE[(M,D)] [ZEROFILL] A normal-size (double-precision) floating-point number. Cannot be unsigned. Allowable values are @code{@w{-1.7976931348623157E+308}} to @code{@w{-2.2250738585072014E-308}}, @code{0}, and @code{2.2250738585072014E-308} to @code{1.7976931348623157E+308}. The M is the display width and D is the number of decimals. @code{DOUBLE} without an argument or @code{FLOAT(X)} where 25 <= X <= 53 stands for a double-precision floating-point number. @tindex DOUBLE PRECISION @tindex REAL @item DOUBLE PRECISION[(M,D)] [ZEROFILL] @itemx REAL[(M,D)] [ZEROFILL] These are synonyms for @code{DOUBLE}. @tindex DECIMAL @item DECIMAL[(M[,D])] [ZEROFILL] An unpacked floating-point number. Cannot be unsigned. Behaves like a @code{CHAR} column: ``unpacked'' means the number is stored as a string, using one character for each digit of the value. The decimal point and, for negative numbers, the @samp{-} sign, are not counted in M (but space for these are reserved). If @code{D} is 0, values will have no decimal point or fractional part. The maximum range of @code{DECIMAL} values is the same as for @code{DOUBLE}, but the actual range for a given @code{DECIMAL} column may be constrained by the choice of @code{M} and @code{D}. If @code{D} is left out it's set to 0. If @code{M} is left out it's set to 10. Note that in MySQL Version 3.22 the @code{M} argument had to includes the space needed for the sign and the decimal point. @tindex NUMERIC @item NUMERIC(M,D) [ZEROFILL] This is a synonym for @code{DECIMAL}. @tindex DATE @item DATE A date. The supported range is @code{'1000-01-01'} to @code{'9999-12-31'}. MySQL displays @code{DATE} values in @code{'YYYY-MM-DD'} format, but allows you to assign values to @code{DATE} columns using either strings or numbers. @xref{DATETIME}. @tindex DATETIME @item DATETIME A date and time combination. The supported range is @code{'1000-01-01 00:00:00'} to @code{'9999-12-31 23:59:59'}. MySQL displays @code{DATETIME} values in @code{'YYYY-MM-DD HH:MM:SS'} format, but allows you to assign values to @code{DATETIME} columns using either strings or numbers. @xref{DATETIME}. @tindex TIMESTAMP @item TIMESTAMP[(M)] A timestamp. The range is @code{'1970-01-01 00:00:00'} to sometime in the year @code{2037}. MySQL displays @code{TIMESTAMP} values in @code{YYYYMMDDHHMMSS}, @code{YYMMDDHHMMSS}, @code{YYYYMMDD}, or @code{YYMMDD} format, depending on whether @code{M} is @code{14} (or missing), @code{12}, @code{8}, or @code{6}, but allows you to assign values to @code{TIMESTAMP} columns using either strings or numbers. A @code{TIMESTAMP} column is useful for recording the date and time of an @code{INSERT} or @code{UPDATE} operation because it is automatically set to the date and time of the most recent operation if you don't give it a value yourself. You can also set it to the current date and time by assigning it a @code{NULL} value. @xref{Date and time types}. A @code{TIMESTAMP} is always stored in 4 bytes. The @code{M} argument only affects how the @code{TIMESTAMP} column is displayed. Note that @code{TIMESTAMP(X)} columns where X is 8 or 14 are reported to be numbers while other @code{TIMESTAMP(X)} columns are reported to be strings. This is just to ensure that one can reliably dump and restore the table with these types! @xref{DATETIME}. @tindex TIME @item TIME A time. The range is @code{'-838:59:59'} to @code{'838:59:59'}. MySQL displays @code{TIME} values in @code{'HH:MM:SS'} format, but allows you to assign values to @code{TIME} columns using either strings or numbers. @xref{TIME}. @tindex YEAR @item YEAR[(2|4)] A year in 2- or 4-digit format (default is 4-digit). The allowable values are @code{1901} to @code{2155}, @code{0000} in the 4-digit year format, and 1970-2069 if you use the 2-digit format (70-69). MySQL displays @code{YEAR} values in @code{YYYY} format, but allows you to assign values to @code{YEAR} columns using either strings or numbers. (The @code{YEAR} type is new in MySQL Version 3.22.). @xref{YEAR}. @tindex NATIONAL CHAR @tindex NCHAR @tindex CHAR @tindex CHARACTER @item [NATIONAL] CHAR(M) [BINARY] A fixed-length string that is always right-padded with spaces to the specified length when stored. The range of @code{M} is 1 to 255 characters. Trailing spaces are removed when the value is retrieved. @code{CHAR} values are sorted and compared in case-insensitive fashion according to the default character set unless the @code{BINARY} keyword is given. @code{NATIONAL CHAR} (short form @code{NCHAR}) is the ANSI SQL way to define that a CHAR column should use the default CHARACTER set. This is the default in MySQL. @code{CHAR} is a shorthand for @code{CHARACTER}. MySQL allows you to create a column of type @code{CHAR(0)}. This is mainly useful when you have to be compliant with some old applications that depend on the existence of a column but that do not actually use the value. This is also quite nice when you need a column that only can take 2 values: A @code{CHAR(0)}, that is not defined as @code{NOT NULL}, will only occupy one bit and can only take 2 values: @code{NULL} or @code{""}. @xref{CHAR}. @tindex BOOL @tindex BIT @item BIT @itemx BOOL @itemx CHAR These three are synonyms for @code{CHAR(1)}. @tindex CHARACTER VARYING @tindex CHAR VARYING @tindex VARCHAR @item [NATIONAL] VARCHAR(M) [BINARY] A variable-length string. @strong{NOTE:} Trailing spaces are removed when the value is stored (this differs from the ANSI SQL specification). The range of @code{M} is 1 to 255 characters. @code{VARCHAR} values are sorted and compared in case-insensitive fashion unless the @code{BINARY} keyword is given. @xref{Silent column changes}. @code{VARCHAR} is a shorthand for @code{CHARACTER VARYING}. @xref{CHAR}. @tindex TINYBLOB @tindex TINYTEXT @item TINYBLOB @itemx TINYTEXT A @code{BLOB} or @code{TEXT} column with a maximum length of 255 (2^8 - 1) characters. @xref{Silent column changes}. @xref{BLOB}. @tindex BLOB @tindex TEXT @item BLOB @itemx TEXT A @code{BLOB} or @code{TEXT} column with a maximum length of 65535 (2^16 - 1) characters. @xref{Silent column changes}. @xref{BLOB}. @tindex MEDIUMBLOB @tindex MEDIUMTEXT @item MEDIUMBLOB @itemx MEDIUMTEXT A @code{BLOB} or @code{TEXT} column with a maximum length of 16777215 (2^24 - 1) characters. @xref{Silent column changes}. @xref{BLOB}. @tindex LONGBLOB @tindex LONGTEXT @item LONGBLOB @itemx LONGTEXT A @code{BLOB} or @code{TEXT} column with a maximum length of 4294967295 (2^32 - 1) characters. @xref{Silent column changes}. Note that because the server/client protocol and MyISAM tables has currently a limit of 16M per communication packet / table row, you can't yet use this the whole range of this type. @xref{BLOB}. @tindex ENUM @item ENUM('value1','value2',...) An enumeration. A string object that can have only one value, chosen from the list of values @code{'value1'}, @code{'value2'}, @code{...}, @code{NULL} or the special @code{""} error value. An @code{ENUM} can have a maximum of 65535 distinct values. @xref{ENUM}. @tindex SET @item SET('value1','value2',...) A set. A string object that can have zero or more values, each of which must be chosen from the list of values @code{'value1'}, @code{'value2'}, @code{...} A @code{SET} can have a maximum of 64 members. @xref{SET}. @end table @menu * Numeric types:: Numeric types * Date and time types:: Date and time types * String types:: String types * Choosing types:: Choosing the right type for a column * Other-vendor column types:: Using column types from other database engines * Storage requirements:: Column type storage requirements @end menu @node Numeric types, Date and time types, Column types, Column types @subsection Numeric Types MySQL supports all of the ANSI/ISO SQL92 numeric types. These types include the exact numeric data types (@code{NUMERIC}, @code{DECIMAL}, @code{INTEGER}, and @code{SMALLINT}), as well as the approximate numeric data types (@code{FLOAT}, @code{REAL}, and @code{DOUBLE PRECISION}). The keyword @code{INT} is a synonym for @code{INTEGER}, and the keyword @code{DEC} is a synonym for @code{DECIMAL}. The @code{NUMERIC} and @code{DECIMAL} types are implemented as the same type by MySQL, as permitted by the SQL92 standard. They are used for values for which it is important to preserve exact precision, for example with monetary data. When declaring a column of one of these types the precision and scale can be (and usually is) specified; for example: @example salary DECIMAL(9,2) @end example In this example, @code{9} (@code{precision}) represents the number of significant decimal digits that will be stored for values, and @code{2} (@code{scale}) represents the number of digits that will be stored following the decimal point. In this case, therefore, the range of values that can be stored in the @code{salary} column is from @code{-9999999.99} to @code{9999999.99}. (MySQL can actually store numbers up to @code{9999999.99} in this column because it doesn't have to store the sign for positive numbers) In ANSI/ISO SQL92, the syntax @code{DECIMAL(p)} is equivalent to @code{DECIMAL(p,0)}. Similarly, the syntax @code{DECIMAL} is equivalent to @code{DECIMAL(p,0)}, where the implementation is allowed to decide the value of @code{p}. MySQL does not currently support either of these variant forms of the @code{DECIMAL}/@code{NUMERIC} data types. This is not generally a serious problem, as the principal benefits of these types derive from the ability to control both precision and scale explicitly. @code{DECIMAL} and @code{NUMERIC} values are stored as strings, rather than as binary floating-point numbers, in order to preserve the decimal precision of those values. One character is used for each digit of the value, the decimal point (if @code{scale} > 0), and the @samp{-} sign (for negative numbers). If @code{scale} is 0, @code{DECIMAL} and @code{NUMERIC} values contain no decimal point or fractional part. The maximum range of @code{DECIMAL} and @code{NUMERIC} values is the same as for @code{DOUBLE}, but the actual range for a given @code{DECIMAL} or @code{NUMERIC} column can be constrained by the @code{precision} or @code{scale} for a given column. When such a column is assigned a value with more digits following the decimal point than are allowed by the specified @code{scale}, the value is rounded to that @code{scale}. When a @code{DECIMAL} or @code{NUMERIC} column is assigned a value whose magnitude exceeds the range implied by the specified (or defaulted) @code{precision} and @code{scale}, MySQL stores the value representing the corresponding end point of that range. As an extension to the ANSI/ISO SQL92 standard, MySQL also supports the integral types @code{TINYINT}, @code{MEDIUMINT}, and @code{BIGINT} as listed in the tables above. Another extension is supported by MySQL for optionally specifying the display width of an integral value in parentheses following the base keyword for the type (for example, @code{INT(4)}). This optional width specification is used to left-pad the display of values whose width is less than the width specified for the column, but does not constrain the range of values that can be stored in the column, nor the number of digits that will be displayed for values whose width exceeds that specified for the column. When used in conjunction with the optional extension attribute @code{ZEROFILL}, the default padding of spaces is replaced with zeroes. For example, for a column declared as @code{INT(5) ZEROFILL}, a value of @code{4} is retrieved as @code{00004}. Note that if you store larger values than the display width in an integer column, you may experience problems when MySQL generates temporary tables for some complicated joins, as in these cases MySQL trusts that the data did fit into the original column width. All integral types can have an optional (non-standard) attribute @code{UNSIGNED}. Unsigned values can be used when you want to allow only positive numbers in a column and you need a little bigger numeric range for the column. The @code{FLOAT} type is used to represent approximate numeric data types. The ANSI/ISO SQL92 standard allows an optional specification of the precision (but not the range of the exponent) in bits following the keyword @code{FLOAT} in parentheses. The MySQL implementation also supports this optional precision specification. When the keyword @code{FLOAT} is used for a column type without a precision specification, MySQL uses four bytes to store the values. A variant syntax is also supported, with two numbers given in parentheses following the @code{FLOAT} keyword. With this option, the first number continues to represent the storage requirements for the value in bytes, and the second number specifies the number of digits to be stored and displayed following the decimal point (as with @code{DECIMAL} and @code{NUMERIC}). When MySQL is asked to store a number for such a column with more decimal digits following the decimal point than specified for the column, the value is rounded to eliminate the extra digits when the value is stored. The @code{REAL} and @code{DOUBLE PRECISION} types do not accept precision specifications. As an extension to the ANSI/ISO SQL92 standard, MySQL recognizes @code{DOUBLE} as a synonym for the @code{DOUBLE PRECISION} type. In contrast with the standard's requirement that the precision for @code{REAL} be smaller than that used for @code{DOUBLE PRECISION}, MySQL implements both as 8-byte double-precision floating-point values (when not running in ``ANSI mode''). For maximum portability, code requiring storage of approximate numeric data values should use @code{FLOAT} or @code{DOUBLE PRECISION} with no specification of precision or number of decimal points. When asked to store a value in a numeric column that is outside the column type's allowable range, MySQL clips the value to the appropriate endpoint of the range and stores the resulting value instead. For example, the range of an @code{INT} column is @code{-2147483648} to @code{2147483647}. If you try to insert @code{-9999999999} into an @code{INT} column, the value is clipped to the lower endpoint of the range, and @code{-2147483648} is stored instead. Similarly, if you try to insert @code{9999999999}, @code{2147483647} is stored instead. If the @code{INT} column is @code{UNSIGNED}, the size of the column's range is the same but its endpoints shift up to @code{0} and @code{4294967295}. If you try to store @code{-9999999999} and @code{9999999999}, the values stored in the column become @code{0} and @code{4294967296}. Conversions that occur due to clipping are reported as ``warnings'' for @code{ALTER TABLE}, @code{LOAD DATA INFILE}, @code{UPDATE}, and multi-row @code{INSERT} statements. @node Date and time types, String types, Numeric types, Column types @subsection Date and Time Types @cindex types, Date and Time @cindex Date and Time types @menu * Y2K issues:: Y2K issues and date types * DATETIME:: The @code{DATETIME}, @code{DATE} and @code{TIMESTAMP} types * TIME:: The @code{TIME} type * YEAR:: The @code{YEAR} type @end menu The date and time types are @code{DATETIME}, @code{DATE}, @code{TIMESTAMP}, @code{TIME}, and @code{YEAR}. Each of these has a range of legal values, as well as a ``zero'' value that is used when you specify a really illegal value. Note that MySQL allows you to store certain 'not strictly' legal date values, for example @code{1999-11-31}. The reason for this is that we think it's the responsibility of the application to handle date checking, not the SQL servers. To make the date checking 'fast', MySQL only checks that the month is in the range of 0-12 and the day is in the range of 0-31. The above ranges are defined this way because MySQL allows you to store, in a @code{DATE} or @code{DATETIME} column, dates where the day or month-day is zero. This is extremely useful for applications that need to store a birth-date for which you don't know the exact date. In this case you simply store the date like @code{1999-00-00} or @code{1999-01-00}. (You cannot expect to get a correct value from functions like @code{DATE_SUB()} or @code{DATE_ADD} for dates like these.) Here are some general considerations to keep in mind when working with date and time types: @itemize @bullet @item MySQL retrieves values for a given date or time type in a standard format, but it attempts to interpret a variety of formats for values that you supply (for example, when you specify a value to be assigned to or compared to a date or time type). Nevertheless, only the formats described in the following sections are supported. It is expected that you will supply legal values, and unpredictable results may occur if you use values in other formats. @item Although MySQL tries to interpret values in several formats, it always expects the year part of date values to be leftmost. Dates must be given in year-month-day order (for example, @code{'98-09-04'}), rather than in the month-day-year or day-month-year orders commonly used elsewhere (for example, @code{'09-04-98'}, @code{'04-09-98'}). @item MySQL automatically converts a date or time type value to a number if the value is used in a numeric context, and vice versa. @item When MySQL encounters a value for a date or time type that is out of range or otherwise illegal for the type (see the start of this section), it converts the value to the ``zero'' value for that type. (The exception is that out-of-range @code{TIME} values are clipped to the appropriate endpoint of the @code{TIME} range.) The table below shows the format of the ``zero'' value for each type: @multitable @columnfractions .3 .7 @item @strong{Column type} @tab @strong{``Zero'' value} @item @code{DATETIME} @tab @code{'0000-00-00 00:00:00'} @item @code{DATE} @tab @code{'0000-00-00'} @item @code{TIMESTAMP} @tab @code{00000000000000} (length depends on display size) @item @code{TIME} @tab @code{'00:00:00'} @item @code{YEAR} @tab @code{0000} @end multitable @item The ``zero'' values are special, but you can store or refer to them explicitly using the values shown in the table. You can also do this using the values @code{'0'} or @code{0}, which are easier to write. @item ``Zero'' date or time values used through @strong{MyODBC} are converted automatically to @code{NULL} in @strong{MyODBC} Version 2.50.12 and above, because ODBC can't handle such values. @end itemize @node Y2K issues, DATETIME, Date and time types, Date and time types @subsubsection Y2K Issues and Date Types @cindex Year 2000 issues @cindex date types, Y2K issues MySQL itself is Y2K-safe (@pxref{Year 2000 compliance}), but input values presented to MySQL may not be. Any input containing 2-digit year values is ambiguous, because the century is unknown. Such values must be interpreted into 4-digit form because MySQL stores years internally using four digits. For @code{DATETIME}, @code{DATE}, @code{TIMESTAMP}, and @code{YEAR} types, MySQL interprets dates with ambiguous year values using the following rules: @itemize @bullet @item Year values in the range @code{00-69} are converted to @code{2000-2069}. @item Year values in the range @code{70-99} are converted to @code{1970-1999}. @end itemize Remember that these rules provide only reasonable guesses as to what your data mean. If the heuristics used by MySQL don't produce the correct values, you should provide unambiguous input containing 4-digit year values. @code{ORDER BY} will sort 2-digit @code{YEAR/DATE/DATETIME} types properly. Note also that some functions like @code{MIN()} and @code{MAX()} will convert a @code{TIMESTAMP/DATE} to a number. This means that a timestamp with a 2-digit year will not work properly with these functions. The fix in this case is to convert the @code{TIMESTAMP/DATE} to 4-digit year format or use something like @code{MIN(DATE_ADD(timestamp,INTERVAL 0 DAYS))}. @node DATETIME, TIME, Y2K issues, Date and time types @subsubsection The @code{DATETIME}, @code{DATE}, and @code{TIMESTAMP} Types @tindex DATETIME @tindex DATE @tindex TIMESTAMP The @code{DATETIME}, @code{DATE}, and @code{TIMESTAMP} types are related. This section describes their characteristics, how they are similar, and how they differ. The @code{DATETIME} type is used when you need values that contain both date and time information. MySQL retrieves and displays @code{DATETIME} values in @code{'YYYY-MM-DD HH:MM:SS'} format. The supported range is @code{'1000-01-01 00:00:00'} to @code{'9999-12-31 23:59:59'}. (``Supported'' means that although earlier values might work, there is no guarantee that they will.) The @code{DATE} type is used when you need only a date value, without a time part. MySQL retrieves and displays @code{DATE} values in @code{'YYYY-MM-DD'} format. The supported range is @code{'1000-01-01'} to @code{'9999-12-31'}. The @code{TIMESTAMP} column type provides a type that you can use to automatically mark @code{INSERT} or @code{UPDATE} operations with the current date and time. If you have multiple @code{TIMESTAMP} columns, only the first one is updated automatically. Automatic updating of the first @code{TIMESTAMP} column occurs under any of the following conditions: @itemize @bullet @item The column is not specified explicitly in an @code{INSERT} or @code{LOAD DATA INFILE} statement. @item The column is not specified explicitly in an @code{UPDATE} statement and some other column changes value. (Note that an @code{UPDATE} that sets a column to the value it already has will not cause the @code{TIMESTAMP} column to be updated, because if you set a column to its current value, MySQL ignores the update for efficiency.) @item You explicitly set the @code{TIMESTAMP} column to @code{NULL}. @end itemize @code{TIMESTAMP} columns other than the first may also be set to the current date and time. Just set the column to @code{NULL} or to @code{NOW()}. You can set any @code{TIMESTAMP} column to a value different than the current date and time by setting it explicitly to the desired value. This is true even for the first @code{TIMESTAMP} column. You can use this property if, for example, you want a @code{TIMESTAMP} to be set to the current date and time when you create a row, but not to be changed whenever the row is updated later: @itemize @bullet @item Let MySQL set the column when the row is created. This will initialize it to the current date and time. @item When you perform subsequent updates to other columns in the row, set the @code{TIMESTAMP} column explicitly to its current value. @end itemize On the other hand, you may find it just as easy to use a @code{DATETIME} column that you initialize to @code{NOW()} when the row is created and leave alone for subsequent updates. @code{TIMESTAMP} values may range from the beginning of 1970 to sometime in the year 2037, with a resolution of one second. Values are displayed as numbers. The format in which MySQL retrieves and displays @code{TIMESTAMP} values depends on the display size, as illustrated by the table below. The `full' @code{TIMESTAMP} format is 14 digits, but @code{TIMESTAMP} columns may be created with shorter display sizes: @multitable @columnfractions .3 .7 @item @strong{Column type} @tab @strong{Display format} @item @code{TIMESTAMP(14)} @tab @code{YYYYMMDDHHMMSS} @item @code{TIMESTAMP(12)} @tab @code{YYMMDDHHMMSS} @item @code{TIMESTAMP(10)} @tab @code{YYMMDDHHMM} @item @code{TIMESTAMP(8)} @tab @code{YYYYMMDD} @item @code{TIMESTAMP(6)} @tab @code{YYMMDD} @item @code{TIMESTAMP(4)} @tab @code{YYMM} @item @code{TIMESTAMP(2)} @tab @code{YY} @end multitable All @code{TIMESTAMP} columns have the same storage size, regardless of display size. The most common display sizes are 6, 8, 12, and 14. You can specify an arbitrary display size at table creation time, but values of 0 or greater than 14 are coerced to 14. Odd-valued sizes in the range from 1 to 13 are coerced to the next higher even number. You can specify @code{DATETIME}, @code{DATE}, and @code{TIMESTAMP} values using any of a common set of formats: @itemize @bullet @item As a string in either @code{'YYYY-MM-DD HH:MM:SS'} or @code{'YY-MM-DD HH:MM:SS'} format. A ``relaxed'' syntax is allowed---any punctuation character may be used as the delimiter between date parts or time parts. For example, @code{'98-12-31 11:30:45'}, @code{'98.12.31 11+30+45'}, @code{'98/12/31 11*30*45'}, and @code{'98@@12@@31 11^30^45'} are equivalent. @item As a string in either @code{'YYYY-MM-DD'} or @code{'YY-MM-DD'} format. A ``relaxed'' syntax is allowed here, too. For example, @code{'98-12-31'}, @code{'98.12.31'}, @code{'98/12/31'}, and @code{'98@@12@@31'} are equivalent. @item As a string with no delimiters in either @code{'YYYYMMDDHHMMSS'} or @code{'YYMMDDHHMMSS'} format, provided that the string makes sense as a date. For example, @code{'19970523091528'} and @code{'970523091528'} are interpreted as @code{'1997-05-23 09:15:28'}, but @code{'971122129015'} is illegal (it has a nonsensical minute part) and becomes @code{'0000-00-00 00:00:00'}. @item As a string with no delimiters in either @code{'YYYYMMDD'} or @code{'YYMMDD'} format, provided that the string makes sense as a date. For example, @code{'19970523'} and @code{'970523'} are interpreted as @code{'1997-05-23'}, but @code{'971332'} is illegal (it has nonsensical month and day parts) and becomes @code{'0000-00-00'}. @item As a number in either @code{YYYYMMDDHHMMSS} or @code{YYMMDDHHMMSS} format, provided that the number makes sense as a date. For example, @code{19830905132800} and @code{830905132800} are interpreted as @code{'1983-09-05 13:28:00'}. @item As a number in either @code{YYYYMMDD} or @code{YYMMDD} format, provided that the number makes sense as a date. For example, @code{19830905} and @code{830905} are interpreted as @code{'1983-09-05'}. @item As the result of a function that returns a value that is acceptable in a @code{DATETIME}, @code{DATE}, or @code{TIMESTAMP} context, such as @code{NOW()} or @code{CURRENT_DATE}. @end itemize Illegal @code{DATETIME}, @code{DATE}, or @code{TIMESTAMP} values are converted to the ``zero'' value of the appropriate type (@code{'0000-00-00 00:00:00'}, @code{'0000-00-00'}, or @code{00000000000000}). For values specified as strings that include date part delimiters, it is not necessary to specify two digits for month or day values that are less than @code{10}. @code{'1979-6-9'} is the same as @code{'1979-06-09'}. Similarly, for values specified as strings that include time part delimiters, it is not necessary to specify two digits for hour, month, or second values that are less than @code{10}. @code{'1979-10-30 1:2:3'} is the same as @code{'1979-10-30 01:02:03'}. Values specified as numbers should be 6, 8, 12, or 14 digits long. If the number is 8 or 14 digits long, it is assumed to be in @code{YYYYMMDD} or @code{YYYYMMDDHHMMSS} format and that the year is given by the first 4 digits. If the number is 6 or 12 digits long, it is assumed to be in @code{YYMMDD} or @code{YYMMDDHHMMSS} format and that the year is given by the first 2 digits. Numbers that are not one of these lengths are interpreted as though padded with leading zeros to the closest length. @cindex non-delimited strings @cindex strings, non-delimited Values specified as non-delimited strings are interpreted using their length as given. If the string is 8 or 14 characters long, the year is assumed to be given by the first 4 characters. Otherwise the year is assumed to be given by the first 2 characters. The string is interpreted from left to right to find year, month, day, hour, minute, and second values, for as many parts as are present in the string. This means you should not use strings that have fewer than 6 characters. For example, if you specify @code{'9903'}, thinking that will represent March, 1999, you will find that MySQL inserts a ``zero'' date into your table. This is because the year and month values are @code{99} and @code{03}, but the day part is missing (zero), so the value is not a legal date. @code{TIMESTAMP} columns store legal values using the full precision with which the value was specified, regardless of the display size. This has several implications: @itemize @bullet @item Always specify year, month, and day, even if your column types are @code{TIMESTAMP(4)} or @code{TIMESTAMP(2)}. Otherwise, the value will not be a legal date and @code{0} will be stored. @item If you use @code{ALTER TABLE} to widen a narrow @code{TIMESTAMP} column, information will be displayed that previously was ``hidden''. @item Similarly, narrowing a @code{TIMESTAMP} column does not cause information to be lost, except in the sense that less information is shown when the values are displayed. @item Although @code{TIMESTAMP} values are stored to full precision, the only function that operates directly on the underlying stored value is @code{UNIX_TIMESTAMP()}. Other functions operate on the formatted retrieved value. This means you cannot use functions such as @code{HOUR()} or @code{SECOND()} unless the relevant part of the @code{TIMESTAMP} value is included in the formatted value. For example, the @code{HH} part of a @code{TIMESTAMP} column is not displayed unless the display size is at least 10, so trying to use @code{HOUR()} on shorter @code{TIMESTAMP} values produces a meaningless result. @end itemize You can to some extent assign values of one date type to an object of a different date type. However, there may be some alteration of the value or loss of information: @itemize @bullet @item If you assign a @code{DATE} value to a @code{DATETIME} or @code{TIMESTAMP} object, the time part of the resulting value is set to @code{'00:00:00'}, because the @code{DATE} value contains no time information. @item If you assign a @code{DATETIME} or @code{TIMESTAMP} value to a @code{DATE} object, the time part of the resulting value is deleted, because the @code{DATE} type stores no time information. @item Remember that although @code{DATETIME}, @code{DATE}, and @code{TIMESTAMP} values all can be specified using the same set of formats, the types do not all have the same range of values. For example, @code{TIMESTAMP} values cannot be earlier than @code{1970} or later than @code{2037}. This means that a date such as @code{'1968-01-01'}, while legal as a @code{DATETIME} or @code{DATE} value, is not a valid @code{TIMESTAMP} value and will be converted to @code{0} if assigned to such an object. @end itemize @cindex problems, date values @cindex date values, problems Be aware of certain pitfalls when specifying date values: @itemize @bullet @item The relaxed format allowed for values specified as strings can be deceiving. For example, a value such as @code{'10:11:12'} might look like a time value because of the @samp{:} delimiter, but if used in a date context will be interpreted as the year @code{'2010-11-12'}. The value @code{'10:45:15'} will be converted to @code{'0000-00-00'} because @code{'45'} is not a legal month. @item Year values specified as two digits are ambiguous, because the century is unknown. MySQL interprets 2-digit year values using the following rules: @itemize @minus @item Year values in the range @code{00-69} are converted to @code{2000-2069}. @item Year values in the range @code{70-99} are converted to @code{1970-1999}. @end itemize @end itemize @node TIME, YEAR, DATETIME, Date and time types @subsubsection The @code{TIME} Type @tindex TIME MySQL retrieves and displays @code{TIME} values in @code{'HH:MM:SS'} format (or @code{'HHH:MM:SS'} format for large hours values). @code{TIME} values may range from @code{'-838:59:59'} to @code{'838:59:59'}. The reason the hours part may be so large is that the @code{TIME} type may be used not only to represent a time of day (which must be less than 24 hours), but also elapsed time or a time interval between two events (which may be much greater than 24 hours, or even negative). You can specify @code{TIME} values in a variety of formats: @itemize @bullet @item As a string in @code{'D HH:MM:SS.fraction'} format. (Note that MySQL doesn't yet store the fraction for the time column). One can also use one of the following ``relaxed'' syntax: @code{HH:MM:SS.fraction}, @code{HH:MM:SS}, @code{HH:MM}, @code{D HH:MM:SS}, @code{D HH:MM}, @code{D HH} or @code{SS}. Here @code{D} is days between 0-33. @item As a string with no delimiters in @code{'HHMMSS'} format, provided that it makes sense as a time. For example, @code{'101112'} is understood as @code{'10:11:12'}, but @code{'109712'} is illegal (it has a nonsensical minute part) and becomes @code{'00:00:00'}. @item As a number in @code{HHMMSS} format, provided that it makes sense as a time. For example, @code{101112} is understood as @code{'10:11:12'}. The following alternative formats are also understood: @code{SS}, @code{MMSS},@code{HHMMSS}, @code{HHMMSS.fraction}. Note that MySQL doesn't yet store the fraction part. @item As the result of a function that returns a value that is acceptable in a @code{TIME} context, such as @code{CURRENT_TIME}. @end itemize For @code{TIME} values specified as strings that include a time part delimiter, it is not necessary to specify two digits for hours, minutes, or seconds values that are less than @code{10}. @code{'8:3:2'} is the same as @code{'08:03:02'}. Be careful about assigning ``short'' @code{TIME} values to a @code{TIME} column. Without semicolon, MySQL interprets values using the assumption that the rightmost digits represent seconds. (MySQL interprets @code{TIME} values as elapsed time rather than as time of day.) For example, you might think of @code{'1112'} and @code{1112} as meaning @code{'11:12:00'} (12 minutes after 11 o'clock), but MySQL interprets them as @code{'00:11:12'} (11 minutes, 12 seconds). Similarly, @code{'12'} and @code{12} are interpreted as @code{'00:00:12'}. @code{TIME} values with semicolon, instead, are always treated as time of the day. That is @code{'11:12'} will mean @code{'11:12:00'}, not @code{'00:11:12'}. Values that lie outside the @code{TIME} range but are otherwise legal are clipped to the appropriate endpoint of the range. For example, @code{'-850:00:00'} and @code{'850:00:00'} are converted to @code{'-838:59:59'} and @code{'838:59:59'}. Illegal @code{TIME} values are converted to @code{'00:00:00'}. Note that because @code{'00:00:00'} is itself a legal @code{TIME} value, there is no way to tell, from a value of @code{'00:00:00'} stored in a table, whether the original value was specified as @code{'00:00:00'} or whether it was illegal. @node YEAR, , TIME, Date and time types @subsubsection The @code{YEAR} Type @tindex YEAR The @code{YEAR} type is a 1-byte type used for representing years. MySQL retrieves and displays @code{YEAR} values in @code{YYYY} format. The range is @code{1901} to @code{2155}. You can specify @code{YEAR} values in a variety of formats: @itemize @bullet @item As a four-digit string in the range @code{'1901'} to @code{'2155'}. @item As a four-digit number in the range @code{1901} to @code{2155}. @item As a two-digit string in the range @code{'00'} to @code{'99'}. Values in the ranges @code{'00'} to @code{'69'} and @code{'70'} to @code{'99'} are converted to @code{YEAR} values in the ranges @code{2000} to @code{2069} and @code{1970} to @code{1999}. @item As a two-digit number in the range @code{1} to @code{99}. Values in the ranges @code{1} to @code{69} and @code{70} to @code{99} are converted to @code{YEAR} values in the ranges @code{2001} to @code{2069} and @code{1970} to @code{1999}. Note that the range for two-digit numbers is slightly different than the range for two-digit strings, because you cannot specify zero directly as a number and have it be interpreted as @code{2000}. You @emph{must} specify it as a string @code{'0'} or @code{'00'} or it will be interpreted as @code{0000}. @item As the result of a function that returns a value that is acceptable in a @code{YEAR} context, such as @code{NOW()}. @end itemize Illegal @code{YEAR} values are converted to @code{0000}. @node String types, Choosing types, Date and time types, Column types @subsection String Types @cindex types, strings @cindex string types @tindex CHAR @tindex VARCHAR @menu * CHAR:: The @code{CHAR} and @code{VARCHAR} types * BLOB:: The @code{BLOB} and @code{TEXT} types * ENUM:: The @code{ENUM} type * SET:: The @code{SET} type @end menu The string types are @code{CHAR}, @code{VARCHAR}, @code{BLOB}, @code{TEXT}, @code{ENUM}, and @code{SET}. This section describes how these types work, their storage requirements, and how to use them in your queries. @node CHAR, BLOB, String types, String types @subsubsection The @code{CHAR} and @code{VARCHAR} Types The @code{CHAR} and @code{VARCHAR} types are similar, but differ in the way they are stored and retrieved. The length of a @code{CHAR} column is fixed to the length that you declare when you create the table. The length can be any value between 1 and 255. (As of MySQL Version 3.23, the length of @code{CHAR} may be 0 to 255.) When @code{CHAR} values are stored, they are right-padded with spaces to the specified length. When @code{CHAR} values are retrieved, trailing spaces are removed. Values in @code{VARCHAR} columns are variable-length strings. You can declare a @code{VARCHAR} column to be any length between 1 and 255, just as for @code{CHAR} columns. However, in contrast to @code{CHAR}, @code{VARCHAR} values are stored using only as many characters as are needed, plus one byte to record the length. Values are not padded; instead, trailing spaces are removed when values are stored. (This space removal differs from the ANSI SQL specification.) If you assign a value to a @code{CHAR} or @code{VARCHAR} column that exceeds the column's maximum length, the value is truncated to fit. The table below illustrates the differences between the two types of columns by showing the result of storing various string values into @code{CHAR(4)} and @code{VARCHAR(4)} columns: @c Need to use @(space) to make sure second column values retain spacing @c in output for table below. @multitable @columnfractions .2 .15 .2 .2 .25 @item @strong{Value} @tab @code{CHAR(4)} @tab @strong{Storage required} @tab @code{VARCHAR(4)} @tab @strong{Storage required} @item @code{''} @tab @code{'@ @ @ @ '} @tab 4 bytes @tab @code{''} @tab 1 byte @item @code{'ab'} @tab @code{'ab@ @ '} @tab 4 bytes @tab @code{'ab'} @tab 3 bytes @item @code{'abcd'} @tab @code{'abcd'} @tab 4 bytes @tab @code{'abcd'} @tab 5 bytes @item @code{'abcdefgh'} @tab @code{'abcd'} @tab 4 bytes @tab @code{'abcd'} @tab 5 bytes @end multitable The values retrieved from the @code{CHAR(4)} and @code{VARCHAR(4)} columns will be the same in each case, because trailing spaces are removed from @code{CHAR} columns upon retrieval. Values in @code{CHAR} and @code{VARCHAR} columns are sorted and compared in case-insensitive fashion, unless the @code{BINARY} attribute was specified when the table was created. The @code{BINARY} attribute means that column values are sorted and compared in case-sensitive fashion according to the ASCII order of the machine where the MySQL server is running. @code{BINARY} doesn't affect how the column is stored or retrieved. The @code{BINARY} attribute is sticky. This means that if a column marked @code{BINARY} is used in an expression, the whole expression is compared as a @code{BINARY} value. MySQL may silently change the type of a @code{CHAR} or @code{VARCHAR} column at table creation time. @xref{Silent column changes}. @node BLOB, ENUM, CHAR, String types @subsubsection The @code{BLOB} and @code{TEXT} Types @tindex BLOB @tindex TEXT A @code{BLOB} is a binary large object that can hold a variable amount of data. The four @code{BLOB} types @code{TINYBLOB}, @code{BLOB}, @code{MEDIUMBLOB}, and @code{LONGBLOB} differ only in the maximum length of the values they can hold. @xref{Storage requirements}. The four @code{TEXT} types @code{TINYTEXT}, @code{TEXT}, @code{MEDIUMTEXT}, and @code{LONGTEXT} correspond to the four @code{BLOB} types and have the same maximum lengths and storage requirements. The only difference between @code{BLOB} and @code{TEXT} types is that sorting and comparison is performed in case-sensitive fashion for @code{BLOB} values and case-insensitive fashion for @code{TEXT} values. In other words, a @code{TEXT} is a case-insensitive @code{BLOB}. If you assign a value to a @code{BLOB} or @code{TEXT} column that exceeds the column type's maximum length, the value is truncated to fit. In most respects, you can regard a @code{TEXT} column as a @code{VARCHAR} column that can be as big as you like. Similarly, you can regard a @code{BLOB} column as a @code{VARCHAR BINARY} column. The differences are: @itemize @bullet @item You can have indexes on @code{BLOB} and @code{TEXT} columns with MySQL Version 3.23.2 and newer. Older versions of MySQL did not support this. @item There is no trailing-space removal for @code{BLOB} and @code{TEXT} columns when values are stored, as there is for @code{VARCHAR} columns. @item @cindex default values, @code{BLOB} and @code{TEXT} columns @cindex @code{BLOB} columns, default values @cindex @code{TEXT} columns, default values @code{BLOB} and @code{TEXT} columns cannot have @code{DEFAULT} values. @end itemize @strong{MyODBC} defines @code{BLOB} values as @code{LONGVARBINARY} and @code{TEXT} values as @code{LONGVARCHAR}. Because @code{BLOB} and @code{TEXT} values may be extremely long, you may run up against some constraints when using them: @itemize @bullet @item If you want to use @code{GROUP BY} or @code{ORDER BY} on a @code{BLOB} or @code{TEXT} column, you must convert the column value into a fixed-length object. The standard way to do this is with the @code{SUBSTRING} function. For example: @example mysql> select comment from tbl_name,substring(comment,20) as substr ORDER BY substr; @end example If you don't do this, only the first @code{max_sort_length} bytes of the column are used when sorting. The default value of @code{max_sort_length} is 1024; this value can be changed using the @code{-O} option when starting the @code{mysqld} server. You can group on an expression involving @code{BLOB} or @code{TEXT} values by specifying the column position or by using an alias: @example mysql> select id,substring(blob_col,1,100) from tbl_name GROUP BY 2; mysql> select id,substring(blob_col,1,100) as b from tbl_name GROUP BY b; @end example @item The maximum size of a @code{BLOB} or @code{TEXT} object is determined by its type, but the largest value you can actually transmit between the client and server is determined by the amount of available memory and the size of the communications buffers. You can change the message buffer size, but you must do so on both the server and client ends. @xref{Server parameters}. @end itemize Note that each @code{BLOB} or @code{TEXT} value is represented internally by a separately allocated object. This is in contrast to all other column types, for which storage is allocated once per column when the table is opened. @node ENUM, SET, BLOB, String types @subsubsection The @code{ENUM} Type @tindex ENUM An @code{ENUM} is a string object whose value normally is chosen from a list of allowed values that are enumerated explicitly in the column specification at table creation time. The value may also be the empty string (@code{""}) or @code{NULL} under certain circumstances: @itemize @bullet @item If you insert an invalid value into an @code{ENUM} (that is, a string not present in the list of allowed values), the empty string is inserted instead as a special error value. @item If an @code{ENUM} is declared @code{NULL}, @code{NULL} is also a legal value for the column, and the default value is @code{NULL}. If an @code{ENUM} is declared @code{NOT NULL}, the default value is the first element of the list of allowed values. @end itemize Each enumeration value has an index: @itemize @bullet @item Values from the list of allowable elements in the column specification are numbered beginning with 1. @item The index value of the empty string error value is 0. This means that you can use the following @code{SELECT} statement to find rows into which invalid @code{ENUM} values were assigned: @example mysql> SELECT * FROM tbl_name WHERE enum_col=0; @end example @item The index of the @code{NULL} value is @code{NULL}. @end itemize For example, a column specified as @code{ENUM("one", "two", "three")} can have any of the values shown below. The index of each value is also shown: @multitable @columnfractions .2 .8 @item @strong{Value} @tab @strong{Index} @item @code{NULL} @tab @code{NULL} @item @code{""} @tab 0 @item @code{"one"} @tab 1 @item @code{"two"} @tab 2 @item @code{"three"} @tab 3 @end multitable An enumeration can have a maximum of 65535 elements. Lettercase is irrelevant when you assign values to an @code{ENUM} column. However, values retrieved from the column later have lettercase matching the values that were used to specify the allowable values at table creation time. If you retrieve an @code{ENUM} in a numeric context, the column value's index is returned. For example, you can retrieve numeric values from an @code{ENUM} column like this: @example mysql> SELECT enum_col+0 FROM tbl_name; @end example If you store a number into an @code{ENUM}, the number is treated as an index, and the value stored is the enumeration member with that index. (However, this will not work with @code{LOAD DATA}, which treats all input as strings.) @code{ENUM} values are sorted according to the order in which the enumeration members were listed in the column specification. (In other words, @code{ENUM} values are sorted according to their index numbers.) For example, @code{"a"} sorts before @code{"b"} for @code{ENUM("a", "b")}, but @code{"b"} sorts before @code{"a"} for @code{ENUM("b", "a")}. The empty string sorts before non-empty strings, and @code{NULL} values sort before all other enumeration values. If you want to get all possible values for an @code{ENUM} column, you should use: @code{SHOW COLUMNS FROM table_name LIKE enum_column_name} and parse the @code{ENUM} definition in the second column. @node SET, , ENUM, String types @subsubsection The @code{SET} Type @tindex SET A @code{SET} is a string object that can have zero or more values, each of which must be chosen from a list of allowed values specified when the table is created. @code{SET} column values that consist of multiple set members are specified with members separated by commas (@samp{,}). A consequence of this is that @code{SET} member values cannot themselves contain commas. For example, a column specified as @code{SET("one", "two") NOT NULL} can have any of these values: @example "" "one" "two" "one,two" @end example A @code{SET} can have a maximum of 64 different members. MySQL stores @code{SET} values numerically, with the low-order bit of the stored value corresponding to the first set member. If you retrieve a @code{SET} value in a numeric context, the value retrieved has bits set corresponding to the set members that make up the column value. For example, you can retrieve numeric values from a @code{SET} column like this: @example mysql> SELECT set_col+0 FROM tbl_name; @end example If a number is stored into a @code{SET} column, the bits that are set in the binary representation of the number determine the set members in the column value. Suppose a column is specified as @code{SET("a","b","c","d")}. Then the members have the following bit values: @multitable @columnfractions .2 .2 .6 @item @code{SET} @strong{member} @tab @strong{Decimal value} @tab @strong{Binary value} @item @code{a} @tab @code{1} @tab @code{0001} @item @code{b} @tab @code{2} @tab @code{0010} @item @code{c} @tab @code{4} @tab @code{0100} @item @code{d} @tab @code{8} @tab @code{1000} @end multitable If you assign a value of @code{9} to this column, that is @code{1001} in binary, so the first and fourth @code{SET} value members @code{"a"} and @code{"d"} are selected and the resulting value is @code{"a,d"}. For a value containing more than one @code{SET} element, it does not matter what order the elements are listed in when you insert the value. It also does not matter how many times a given element is listed in the value. When the value is retrieved later, each element in the value will appear once, with elements listed according to the order in which they were specified at table creation time. For example, if a column is specified as @code{SET("a","b","c","d")}, then @code{"a,d"}, @code{"d,a"}, and @code{"d,a,a,d,d"} will all appear as @code{"a,d"} when retrieved. @code{SET} values are sorted numerically. @code{NULL} values sort before non-@code{NULL} @code{SET} values. Normally, you perform a @code{SELECT} on a @code{SET} column using the @code{LIKE} operator or the @code{FIND_IN_SET()} function: @example mysql> SELECT * FROM tbl_name WHERE set_col LIKE '%value%'; mysql> SELECT * FROM tbl_name WHERE FIND_IN_SET('value',set_col)>0; @end example But the following will also work: @example mysql> SELECT * FROM tbl_name WHERE set_col = 'val1,val2'; mysql> SELECT * FROM tbl_name WHERE set_col & 1; @end example The first of these statements looks for an exact match. The second looks for values containing the first set member. If you want to get all possible values for a @code{SET} column, you should use: @code{SHOW COLUMNS FROM table_name LIKE set_column_name} and parse the @code{SET} definition in the second column. @node Choosing types, Other-vendor column types, String types, Column types @subsection Choosing the Right Type for a Column @cindex types, columns @cindex choosing types For the most efficient use of storage, try to use the most precise type in all cases. For example, if an integer column will be used for values in the range between @code{1} and @code{99999}, @code{MEDIUMINT UNSIGNED} is the best type. Accurate representation of monetary values is a common problem. In MySQL, you should use the @code{DECIMAL} type. This is stored as a string, so no loss of accuracy should occur. If accuracy is not too important, the @code{DOUBLE} type may also be good enough. For high precision, you can always convert to a fixed-point type stored in a @code{BIGINT}. This allows you to do all calculations with integers and convert results back to floating-point values only when necessary. @node Other-vendor column types, Storage requirements, Choosing types, Column types @subsection Using Column Types from Other Database Engines @cindex types, portability @cindex portability, types @cindex columns, other types To make it easier to use code written for SQL implementations from other vendors, MySQL maps column types as shown in the table below. These mappings make it easier to move table definitions from other database engines to MySQL: @multitable @columnfractions .4 .6 @item @strong{Other vendor type} @tab @strong{MySQL type} @item @code{BINARY(NUM)} @tab @code{CHAR(NUM) BINARY} @item @code{CHAR VARYING(NUM)} @tab @code{VARCHAR(NUM)} @item @code{FLOAT4} @tab @code{FLOAT} @item @code{FLOAT8} @tab @code{DOUBLE} @item @code{INT1} @tab @code{TINYINT} @item @code{INT2} @tab @code{SMALLINT} @item @code{INT3} @tab @code{MEDIUMINT} @item @code{INT4} @tab @code{INT} @item @code{INT8} @tab @code{BIGINT} @item @code{LONG VARBINARY} @tab @code{MEDIUMBLOB} @item @code{LONG VARCHAR} @tab @code{MEDIUMTEXT} @item @code{MIDDLEINT} @tab @code{MEDIUMINT} @item @code{VARBINARY(NUM)} @tab @code{VARCHAR(NUM) BINARY} @end multitable Column type mapping occurs at table creation time. If you create a table with types used by other vendors and then issue a @code{DESCRIBE tbl_name} statement, MySQL reports the table structure using the equivalent MySQL types. @node Storage requirements, , Other-vendor column types, Column types @subsection Column Type Storage Requirements @cindex storage requirements, column type @cindex columns, storage requirements The storage requirements for each of the column types supported by MySQL are listed below by category. @cindex numeric types @cindex types, numeric @subsubheading Storage requirements for numeric types @multitable @columnfractions .5 .5 @item @strong{Column type} @tab @strong{Storage required} @item @code{TINYINT} @tab 1 byte @item @code{SMALLINT} @tab 2 bytes @item @code{MEDIUMINT} @tab 3 bytes @item @code{INT} @tab 4 bytes @item @code{INTEGER} @tab 4 bytes @item @code{BIGINT} @tab 8 bytes @item @code{FLOAT(X)} @tab 4 if X <= 24 or 8 if 25 <= X <= 53 @item @code{FLOAT} @tab 4 bytes @item @code{DOUBLE} @tab 8 bytes @item @code{DOUBLE PRECISION} @tab 8 bytes @item @code{REAL} @tab 8 bytes @item @code{DECIMAL(M,D)} @tab @code{M+2} bytes if D > 0, @code{M+1} bytes if D = 0 (@code{D}+2, if @code{M < D}) @item @code{NUMERIC(M,D)} @tab @code{M+2} bytes if D > 0, @code{M+1} bytes if D = 0 (@code{D}+2, if @code{M < D}) @end multitable @cindex date types @cindex time types @cindex types, date @cindex types, time @subsubheading Storage requirements for date and time types @multitable @columnfractions .5 .5 @item @strong{Column type} @tab @strong{Storage required} @item @code{DATE} @tab 3 bytes @item @code{DATETIME} @tab 8 bytes @item @code{TIMESTAMP} @tab 4 bytes @item @code{TIME} @tab 3 bytes @item @code{YEAR} @tab 1 byte @end multitable @subsubheading Storage requirements for string types @multitable @columnfractions .5 .5 @item @strong{Column type} @tab @strong{Storage required} @item @code{CHAR(M)} @tab @code{M} bytes, @code{1 <= M <= 255} @item @code{VARCHAR(M)} @tab @code{L}+1 bytes, where @code{L <= M} and @code{1 <= M <= 255} @item @code{TINYBLOB}, @code{TINYTEXT} @tab @code{L}+1 bytes, where @code{L} < 2^8 @item @code{BLOB}, @code{TEXT} @tab @code{L}+2 bytes, where @code{L} < 2^16 @item @code{MEDIUMBLOB}, @code{MEDIUMTEXT} @tab @code{L}+3 bytes, where @code{L} < 2^24 @item @code{LONGBLOB}, @code{LONGTEXT} @tab @code{L}+4 bytes, where @code{L} < 2^32 @item @code{ENUM('value1','value2',...)} @tab 1 or 2 bytes, depending on the number of enumeration values (65535 values maximum) @item @code{SET('value1','value2',...)} @tab 1, 2, 3, 4 or 8 bytes, depending on the number of set members (64 members maximum) @end multitable @cindex @code{BLOB}, size @cindex @code{TEXT}, size @cindex @code{VARCHAR}, size @code{VARCHAR} and the @code{BLOB} and @code{TEXT} types are variable-length types, for which the storage requirements depend on the actual length of column values (represented by @code{L} in the preceding table), rather than on the type's maximum possible size. For example, a @code{VARCHAR(10)} column can hold a string with a maximum length of 10 characters. The actual storage required is the length of the string (@code{L}), plus 1 byte to record the length of the string. For the string @code{'abcd'}, @code{L} is 4 and the storage requirement is 5 bytes. The @code{BLOB} and @code{TEXT} types require 1, 2, 3, or 4 bytes to record the length of the column value, depending on the maximum possible length of the type. @xref{BLOB}. If a table includes any variable-length column types, the record format will also be variable-length. Note that when a table is created, MySQL may, under certain conditions, change a column from a variable-length type to a fixed-length type, or vice-versa. @xref{Silent column changes}. @cindex ENUM, size The size of an @code{ENUM} object is determined by the number of different enumeration values. One byte is used for enumerations with up to 255 possible values. Two bytes are used for enumerations with up to 65535 values. @xref{ENUM}. @cindex SET, size The size of a @code{SET} object is determined by the number of different set members. If the set size is @code{N}, the object occupies @code{(N+7)/8} bytes, rounded up to 1, 2, 3, 4, or 8 bytes. A @code{SET} can have a maximum of 64 members. @xref{SET}. @node Functions, Data Manipulation, Column types, Reference @section Functions for Use in @code{SELECT} and @code{WHERE} Clauses @cindex functions for @code{SELECT} and @code{WHERE} clauses A @code{select_expression} or @code{where_definition} in a SQL statement can consist of any expression using the functions described below. An expression that contains @code{NULL} always produces a @code{NULL} value unless otherwise indicated in the documentation for the operators and functions involved in the expression. @strong{NOTE:} There must be no whitespace between a function name and the parenthesis following it. This helps the MySQL parser distinguish between function calls and references to tables or columns that happen to have the same name as a function. Spaces around arguments are permitted, though. You can force MySQL to accept spaces after the function name by starting @code{mysqld} with @code{--ansi} or using the @code{CLIENT_IGNORE_SPACE} to @code{mysql_connect()}, but in this case all function names will become reserved words. @xref{ANSI mode}. @need 2000 For the sake of brevity, examples display the output from the @code{mysql} program in abbreviated form. So this: @example mysql> select MOD(29,9); 1 rows in set (0.00 sec) +-----------+ | mod(29,9) | +-----------+ | 2 | +-----------+ @end example is displayed like this: @example mysql> select MOD(29,9); -> 2 @end example @menu * Non-typed Operators:: Non-Type-Specific Operators and Functions * String functions:: String functions * Numeric Functions:: Numeric Functions * Date and time functions:: Date and time functions * Other Functions:: Other Functions @end menu @node Non-typed Operators, String functions, Functions, Functions @subsection Non-Type-Specific Operators and Functions @menu * Parenthesis:: Parenthesis * Comparison Operators:: Comparison Operators * Logical Operators:: Logical Operators * Control flow functions:: Control Flow Functions @end menu @node Parenthesis, Comparison Operators, Non-typed Operators, Non-typed Operators @subsubsection Parenthesis @findex () (parentheses) @findex parentheses ( and ) @cindex functions, grouping @cindex grouping, expressions @example ( ... ) @end example Use parenthesis to force the order of evaluation in an expression. For example: @example mysql> select 1+2*3; -> 7 mysql> select (1+2)*3; -> 9 @end example @node Comparison Operators, Logical Operators, Parenthesis, Non-typed Operators @subsubsection Comparison Operators @findex comparison operators @cindex casts @cindex type conversions Comparison operations result in a value of @code{1} (TRUE), @code{0} (FALSE), or @code{NULL}. These functions work for both numbers and strings. Strings are automatically converted to numbers and numbers to strings as needed (as in Perl). MySQL performs comparisons using the following rules: @itemize @bullet @item If one or both arguments are @code{NULL}, the result of the comparison is @code{NULL}, except for the @code{<=>} operator. @item If both arguments in a comparison operation are strings, they are compared as strings. @item If both arguments are integers, they are compared as integers. @item Hexadecimal values are treated as binary strings if not compared to a number. @item @cindex ODBC compatibility @cindex compatibility, with ODBC If one of the arguments is a @code{TIMESTAMP} or @code{DATETIME} column and the other argument is a constant, the constant is converted to a timestamp before the comparison is performed. This is done to be more ODBC-friendly. @item In all other cases, the arguments are compared as floating-point (real) numbers. @end itemize By default, string comparisons are done in case-independent fashion using the current character set (ISO-8859-1 Latin1 by default, which also works excellently for English). The examples below illustrate conversion of strings to numbers for comparison operations: @example mysql> SELECT 1 > '6x'; -> 0 mysql> SELECT 7 > '6x'; -> 1 mysql> SELECT 0 > 'x6'; -> 0 mysql> SELECT 0 = 'x6'; -> 1 @end example @table @code @findex = (equal) @findex equal (=) @item = Equal: @example mysql> select 1 = 0; -> 0 mysql> select '0' = 0; -> 1 mysql> select '0.0' = 0; -> 1 mysql> select '0.01' = 0; -> 0 mysql> select '.01' = 0.01; -> 1 @end example @findex <> (not equal) @findex not equal (<>) @findex != (not equal) @findex not equal (!=) @item <> @itemx != Not equal: @example mysql> select '.01' <> '0.01'; -> 1 mysql> select .01 <> '0.01'; -> 0 mysql> select 'zapp' <> 'zappp'; -> 1 @end example @findex <= (less than or equal) @findex less than or equal (<=) @item <= Less than or equal: @example mysql> select 0.1 <= 2; -> 1 @end example @findex < (less than) @findex less than (<) @item < Less than: @example mysql> select 2 < 2; -> 0 @end example @findex >= (greater than or equal) @findex greater than or equal (>=) @item >= Greater than or equal: @example mysql> select 2 >= 2; -> 1 @end example @findex > (greater than) @findex greater than (>) @item > Greater than: @example mysql> select 2 > 2; -> 0 @end example @cindex @code{NULL}, testing for null @findex <=> (Equal to) @item <=> Null safe equal: @example mysql> select 1 <=> 1, NULL <=> NULL, 1 <=> NULL; -> 1 1 0 @end example @cindex @code{NULL}, testing for null @findex IS NULL @findex IS NOT NULL @item IS NULL @itemx IS NOT NULL Test whether or not a value is or is not @code{NULL}: @example mysql> select 1 IS NULL, 0 IS NULL, NULL IS NULL; -> 0 0 1 mysql> select 1 IS NOT NULL, 0 IS NOT NULL, NULL IS NOT NULL; -> 1 1 0 @end example @findex BETWEEN ... AND @item expr BETWEEN min AND max If @code{expr} is greater than or equal to @code{min} and @code{expr} is less than or equal to @code{max}, @code{BETWEEN} returns @code{1}, otherwise it returns @code{0}. This is equivalent to the expression @code{(min <= expr AND expr <= max)} if all the arguments are of the same type. The first argument (@code{expr}) determines how the comparison is performed as follows: @itemize @bullet @item If @code{expr} is a @code{TIMESTAMP}, @code{DATE}, or @code{DATETIME} column, @code{MIN()} and @code{MAX()} are formatted to the same format if they are constants. @item If @code{expr} is a case-insensitive string expression, a case-insensitive string comparison is done. @item If @code{expr} is a case-sensitive string expression, a case-sensitive string comparison is done. @item If @code{expr} is an integer expression, an integer comparison is done. @item Otherwise, a floating-point (real) comparison is done. @end itemize @example mysql> select 1 BETWEEN 2 AND 3; -> 0 mysql> select 'b' BETWEEN 'a' AND 'c'; -> 1 mysql> select 2 BETWEEN 2 AND '3'; -> 1 mysql> select 2 BETWEEN 2 AND 'x-3'; -> 0 @end example @findex IN @item expr IN (value,...) Returns @code{1} if @code{expr} is any of the values in the @code{IN} list, else returns @code{0}. If all values are constants, then all values are evaluated according to the type of @code{expr} and sorted. The search for the item is then done using a binary search. This means @code{IN} is very quick if the @code{IN} value list consists entirely of constants. If @code{expr} is a case-sensitive string expression, the string comparison is performed in case-sensitive fashion: @example mysql> select 2 IN (0,3,5,'wefwf'); -> 0 mysql> select 'wefwf' IN (0,3,5,'wefwf'); -> 1 @end example @findex NOT IN @item expr NOT IN (value,...) Same as @code{NOT (expr IN (value,...))}. @findex ISNULL() @item ISNULL(expr) If @code{expr} is @code{NULL}, @code{ISNULL()} returns @code{1}, otherwise it returns @code{0}: @example mysql> select ISNULL(1+1); -> 0 mysql> select ISNULL(1/0); -> 1 @end example Note that a comparison of @code{NULL} values using @code{=} will always be false! @cindex @code{NULL}, testing for null @findex COALESCE() @item COALESCE(list) Returns first non-@code{NULL} element in list: @example mysql> select COALESCE(NULL,1); -> 1 mysql> select COALESCE(NULL,NULL,NULL); -> NULL @end example @findex INTERVAL() @item INTERVAL(N,N1,N2,N3,...) Returns @code{0} if @code{N} < @code{N1}, @code{1} if @code{N} < @code{N2} and so on. All arguments are treated as integers. It is required that @code{N1} < @code{N2} < @code{N3} < @code{...} < @code{Nn} for this function to work correctly. This is because a binary search is used (very fast): @example mysql> select INTERVAL(23, 1, 15, 17, 30, 44, 200); -> 3 mysql> select INTERVAL(10, 1, 10, 100, 1000); -> 2 mysql> select INTERVAL(22, 23, 30, 44, 200); -> 0 @end example @end table If you are comparing case sensitive string with any of the standard operators (@code{=}, @code{<>}..., but not @code{LIKE}) end space will be ignored. @example mysql> select "a" ="A "; -> 1 @end example @node Logical Operators, Control flow functions, Comparison Operators, Non-typed Operators @subsubsection Logical Operators @findex Logical functions @findex Functions, logical All logical functions return @code{1} (TRUE), @code{0} (FALSE) or @code{NULL} (unknown, which is in most cases the same as FALSE): @table @code @findex NOT, logical @findex ! (logical NOT) @item NOT @itemx ! Logical NOT. Returns @code{1} if the argument is @code{0}, otherwise returns @code{0}. Exception: @code{NOT NULL} returns @code{NULL}: @example mysql> select NOT 1; -> 0 mysql> select NOT NULL; -> NULL mysql> select ! (1+1); -> 0 mysql> select ! 1+1; -> 1 @end example The last example returns @code{1} because the expression evaluates the same way as @code{(!1)+1}. @findex OR, logical @findex || (logical OR) @item OR @itemx || Logical OR. Returns @code{1} if either argument is not @code{0} and not @code{NULL}: @example mysql> select 1 || 0; -> 1 mysql> select 0 || 0; -> 0 mysql> select 1 || NULL; -> 1 @end example @findex AND, logical @findex && (logical AND) @item AND @itemx && Logical AND. Returns @code{0} if either argument is @code{0} or @code{NULL}, otherwise returns @code{1}: @example mysql> select 1 && NULL; -> 0 mysql> select 1 && 0; -> 0 @end example @end table @node Control flow functions, , Logical Operators, Non-typed Operators @subsubsection Control Flow Functions @findex control flow functions @findex functions, control flow @table @code @cindex @code{NULL}, testing for null @findex IFNULL() @item IFNULL(expr1,expr2) If @code{expr1} is not @code{NULL}, @code{IFNULL()} returns @code{expr1}, else it returns @code{expr2}. @code{IFNULL()} returns a numeric or string value, depending on the context in which it is used: @example mysql> select IFNULL(1,0); -> 1 mysql> select IFNULL(NULL,10); -> 10 mysql> select IFNULL(1/0,10); -> 10 mysql> select IFNULL(1/0,'yes'); -> 'yes' @end example @findex NULLIF() @item NULLIF(expr1,expr2) If @code{expr1 = expr2} is true, return @code{NULL} else return @code{expr1}. This is the same as @code{CASE WHEN x = y THEN NULL ELSE x END}: @example mysql> select NULLIF(1,1); -> NULL mysql> select NULLIF(1,2); -> 1 @end example Note that @code{expr1} is evaluated twice in MySQL if the arguments are equal. @findex IF() @item IF(expr1,expr2,expr3) If @code{expr1} is TRUE (@code{expr1 <> 0} and @code{expr1 <> NULL}) then @code{IF()} returns @code{expr2}, else it returns @code{expr3}. @code{IF()} returns a numeric or string value, depending on the context in which it is used: @example mysql> select IF(1>2,2,3); -> 3 mysql> select IF(1<2,'yes','no'); -> 'yes' mysql> select IF(strcmp('test','test1'),'no','yes'); -> 'no' @end example @code{expr1} is evaluated as an integer value, which means that if you are testing floating-point or string values, you should do so using a comparison operation: @example mysql> select IF(0.1,1,0); -> 0 mysql> select IF(0.1<>0,1,0); -> 1 @end example In the first case above, @code{IF(0.1)} returns @code{0} because @code{0.1} is converted to an integer value, resulting in a test of @code{IF(0)}. This may not be what you expect. In the second case, the comparison tests the original floating-point value to see whether it is non-zero. The result of the comparison is used as an integer. The default return type of @code{IF()} (which may matter when it is stored into a temporary table) is calculated in MySQL Version 3.23 as follows: @multitable @columnfractions .55 .45 @item @strong{Expression} @tab @strong{Return value} @item expr2 or expr3 returns string @tab string @item expr2 or expr3 returns a floating-point value @tab floating-point @item expr2 or expr3 returns an integer @tab integer @end multitable If expr2 and expr3 are strings, then the result is case sensitive if both strings are case sensitive. (Starting from 3.23.51) @findex CASE @item CASE value WHEN [compare-value] THEN result [WHEN [compare-value] THEN result ...] [ELSE result] END @item CASE WHEN [condition] THEN result [WHEN [condition] THEN result ...] [ELSE result] END The first version returns the @code{result} where @code{value=compare-value}. The second version returns the result for the first condition, which is true. If there was no matching result value, then the result after @code{ELSE} is returned. If there is no @code{ELSE} part then @code{NULL} is returned: @example mysql> SELECT CASE 1 WHEN 1 THEN "one" WHEN 2 THEN "two" ELSE "more" END; -> "one" mysql> SELECT CASE WHEN 1>0 THEN "true" ELSE "false" END; -> "true" mysql> SELECT CASE BINARY "B" when "a" then 1 when "b" then 2 END; -> NULL @end example @end table The type of the return value (@code{INTEGER}, @code{DOUBLE} or @code{STRING}) is the same as the type of the first returned value (the expression after the first @code{THEN}). @node String functions, Numeric Functions, Non-typed Operators, Functions @subsection String Functions @findex string functions @findex functions, string String-valued functions return @code{NULL} if the length of the result would be greater than the @code{max_allowed_packet} server parameter. @xref{Server parameters}. For functions that operate on string positions, the first position is numbered 1. @table @code @findex ASCII() @item ASCII(str) Returns the ASCII code value of the leftmost character of the string @code{str}. Returns @code{0} if @code{str} is the empty string. Returns @code{NULL} if @code{str} is @code{NULL}: @example mysql> select ASCII('2'); -> 50 mysql> select ASCII(2); -> 50 mysql> select ASCII('dx'); -> 100 @end example See also the @code{ORD()} function. @findex ORD() @item ORD(str) If the leftmost character of the string str is a multi-byte character, returns the code of multi-byte character by returning the ASCII code value of the character in the format of: @code{((first byte ASCII code)*256+(second byte ASCII code))[*256+third byte ASCII code...]}. If the leftmost character is not a multi-byte character, returns the same value as the like @code{ASCII()} function does: @example mysql> select ORD('2'); -> 50 @end example @findex CONV() @item CONV(N,from_base,to_base) Converts numbers between different number bases. Returns a string representation of the number @code{N}, converted from base @code{from_base} to base @code{to_base}. Returns @code{NULL} if any argument is @code{NULL}. The argument @code{N} is interpreted as an integer, but may be specified as an integer or a string. The minimum base is @code{2} and the maximum base is @code{36}. If @code{to_base} is a negative number, @code{N} is regarded as a signed number. Otherwise, @code{N} is treated as unsigned. @code{CONV} works with 64-bit precision: @example mysql> select CONV("a",16,2); -> '1010' mysql> select CONV("6E",18,8); -> '172' mysql> select CONV(-17,10,-18); -> '-H' mysql> select CONV(10+"10"+'10'+0xa,10,10); -> '40' @end example @findex BIN() @item BIN(N) Returns a string representation of the binary value of @code{N}, where @code{N} is a longlong (@code{BIGINT}) number. This is equivalent to @code{CONV(N,10,2)}. Returns @code{NULL} if @code{N} is @code{NULL}: @example mysql> select BIN(12); -> '1100' @end example @findex OCT() @item OCT(N) Returns a string representation of the octal value of @code{N}, where @code{N} is a longlong number. This is equivalent to @code{CONV(N,10,8)}. Returns @code{NULL} if @code{N} is @code{NULL}: @example mysql> select OCT(12); -> '14' @end example @findex HEX() @item HEX(N) Returns a string representation of the hexadecimal value of @code{N}, where @code{N} is a longlong (@code{BIGINT}) number. This is equivalent to @code{CONV(N,10,16)}. Returns @code{NULL} if @code{N} is @code{NULL}: @example mysql> select HEX(255); -> 'FF' @end example @findex CHAR() @item CHAR(N,...) @code{CHAR()} interprets the arguments as integers and returns a string consisting of the characters given by the ASCII code values of those integers. @code{NULL} values are skipped: @example mysql> select CHAR(77,121,83,81,'76'); -> 'MySQL' mysql> select CHAR(77,77.3,'77.3'); -> 'MMM' @end example @findex CONCAT() @item CONCAT(str1,str2,...) Returns the string that results from concatenating the arguments. Returns @code{NULL} if any argument is @code{NULL}. May have more than 2 arguments. A numeric argument is converted to the equivalent string form: @example mysql> select CONCAT('My', 'S', 'QL'); -> 'MySQL' mysql> select CONCAT('My', NULL, 'QL'); -> NULL mysql> select CONCAT(14.3); -> '14.3' @end example @findex CONCAT_WS() @item CONCAT_WS(separator, str1, str2,...) @code{CONCAT_WS()} stands for CONCAT With Separator and is a special form of @code{CONCAT()}. The first argument is the separator for the rest of the arguments. The separator can be a string as well as the rest of the arguments. If the separator is @code{NULL}, the result will be @code{NULL}. The function will skip any @code{NULL}s and empty strings, after the separator argument. The separator will be added between the strings to be concatenated: @example mysql> select CONCAT_WS(",","First name","Second name","Last Name"); -> 'First name,Second name,Last Name' mysql> select CONCAT_WS(",","First name",NULL,"Last Name"); -> 'First name,Last Name' @end example @findex LENGTH() @findex OCTET_LENGTH() @findex CHAR_LENGTH() @findex CHARACTER_LENGTH() @item LENGTH(str) @itemx OCTET_LENGTH(str) @itemx CHAR_LENGTH(str) @itemx CHARACTER_LENGTH(str) Returns the length of the string @code{str}: @example mysql> select LENGTH('text'); -> 4 mysql> select OCTET_LENGTH('text'); -> 4 @end example Note that for @code{CHAR_LENGTH()}, multi-byte characters are only counted once. @findex LOCATE() @findex POSITION() @item LOCATE(substr,str) @itemx POSITION(substr IN str) Returns the position of the first occurrence of substring @code{substr} in string @code{str}. Returns @code{0} if @code{substr} is not in @code{str}: @example mysql> select LOCATE('bar', 'foobarbar'); -> 4 mysql> select LOCATE('xbar', 'foobar'); -> 0 @end example This function is multi-byte safe. @findex LOCATE() @item LOCATE(substr,str,pos) Returns the position of the first occurrence of substring @code{substr} in string @code{str}, starting at position @code{pos}. Returns @code{0} if @code{substr} is not in @code{str}: @example mysql> select LOCATE('bar', 'foobarbar',5); -> 7 @end example This function is multi-byte safe. @findex INSTR() @item INSTR(str,substr) Returns the position of the first occurrence of substring @code{substr} in string @code{str}. This is the same as the two-argument form of @code{LOCATE()}, except that the arguments are swapped: @example mysql> select INSTR('foobarbar', 'bar'); -> 4 mysql> select INSTR('xbar', 'foobar'); -> 0 @end example This function is multi-byte safe. @findex LPAD() @item LPAD(str,len,padstr) Returns the string @code{str}, left-padded with the string @code{padstr} until @code{str} is @code{len} characters long. If @code{str} is longer than @code{len'} then it will be shortened to @code{len} characters. @example mysql> select LPAD('hi',4,'??'); -> '??hi' @end example @findex RPAD() @item RPAD(str,len,padstr) Returns the string @code{str}, right-padded with the string @code{padstr} until @code{str} is @code{len} characters long. If @code{str} is longer than @code{len'} then it will be shortened to @code{len} characters. @example mysql> select RPAD('hi',5,'?'); -> 'hi???' @end example @findex LEFT() @item LEFT(str,len) Returns the leftmost @code{len} characters from the string @code{str}: @example mysql> select LEFT('foobarbar', 5); -> 'fooba' @end example This function is multi-byte safe. @findex RIGHT() @item RIGHT(str,len) Returns the rightmost @code{len} characters from the string @code{str}: @example mysql> select RIGHT('foobarbar', 4); -> 'rbar' @end example This function is multi-byte safe. @findex SUBSTRING() @findex MID() @item SUBSTRING(str,pos,len) @itemx SUBSTRING(str FROM pos FOR len) @itemx MID(str,pos,len) Returns a substring @code{len} characters long from string @code{str}, starting at position @code{pos}. The variant form that uses @code{FROM} is ANSI SQL92 syntax: @example mysql> select SUBSTRING('Quadratically',5,6); -> 'ratica' @end example This function is multi-byte safe. @findex SUBSTRING() @item SUBSTRING(str,pos) @item SUBSTRING(str FROM pos) Returns a substring from string @code{str} starting at position @code{pos}: @example mysql> select SUBSTRING('Quadratically',5); -> 'ratically' mysql> select SUBSTRING('foobarbar' FROM 4); -> 'barbar' @end example This function is multi-byte safe. @findex SUBSTRING_INDEX() @item SUBSTRING_INDEX(str,delim,count) Returns the substring from string @code{str} before @code{count} occurrences of the delimiter @code{delim}. If @code{count} is positive, everything to the left of the final delimiter (counting from the left) is returned. If @code{count} is negative, everything to the right of the final delimiter (counting from the right) is returned: @example mysql> select SUBSTRING_INDEX('www.mysql.com', '.', 2); -> 'www.mysql' mysql> select SUBSTRING_INDEX('www.mysql.com', '.', -2); -> 'mysql.com' @end example This function is multi-byte safe. @findex LTRIM() @item LTRIM(str) Returns the string @code{str} with leading space characters removed: @example mysql> select LTRIM(' barbar'); -> 'barbar' @end example @findex RTRIM() @item RTRIM(str) Returns the string @code{str} with trailing space characters removed: @example mysql> select RTRIM('barbar '); -> 'barbar' @end example This function is multi-byte safe. @findex TRIM() @item TRIM([[BOTH | LEADING | TRAILING] [remstr] FROM] str) Returns the string @code{str} with all @code{remstr} prefixes and/or suffixes removed. If none of the specifiers @code{BOTH}, @code{LEADING} or @code{TRAILING} are given, @code{BOTH} is assumed. If @code{remstr} is not specified, spaces are removed: @example mysql> select TRIM(' bar '); -> 'bar' mysql> select TRIM(LEADING 'x' FROM 'xxxbarxxx'); -> 'barxxx' mysql> select TRIM(BOTH 'x' FROM 'xxxbarxxx'); -> 'bar' mysql> select TRIM(TRAILING 'xyz' FROM 'barxxyz'); -> 'barx' @end example This function is multi-byte safe. @findex SOUNDEX() @item SOUNDEX(str) Returns a soundex string from @code{str}. Two strings that sound almost the same should have identical soundex strings. A standard soundex string is 4 characters long, but the @code{SOUNDEX()} function returns an arbitrarily long string. You can use @code{SUBSTRING()} on the result to get a standard soundex string. All non-alphanumeric characters are ignored in the given string. All international alpha characters outside the A-Z range are treated as vowels: @example mysql> select SOUNDEX('Hello'); -> 'H400' mysql> select SOUNDEX('Quadratically'); -> 'Q36324' @end example @findex SPACE() @item SPACE(N) Returns a string consisting of @code{N} space characters: @example mysql> select SPACE(6); -> ' ' @end example @findex REPLACE() @item REPLACE(str,from_str,to_str) Returns the string @code{str} with all all occurrences of the string @code{from_str} replaced by the string @code{to_str}: @example mysql> select REPLACE('www.mysql.com', 'w', 'Ww'); -> 'WwWwWw.mysql.com' @end example This function is multi-byte safe. @findex REPEAT() @item REPEAT(str,count) Returns a string consisting of the string @code{str} repeated @code{count} times. If @code{count <= 0}, returns an empty string. Returns @code{NULL} if @code{str} or @code{count} are @code{NULL}: @example mysql> select REPEAT('MySQL', 3); -> 'MySQLMySQLMySQL' @end example @findex REVERSE() @item REVERSE(str) Returns the string @code{str} with the order of the characters reversed: @example mysql> select REVERSE('abc'); -> 'cba' @end example This function is multi-byte safe. @findex INSERT() @item INSERT(str,pos,len,newstr) Returns the string @code{str}, with the substring beginning at position @code{pos} and @code{len} characters long replaced by the string @code{newstr}: @example mysql> select INSERT('Quadratic', 3, 4, 'What'); -> 'QuWhattic' @end example This function is multi-byte safe. @findex ELT() @item ELT(N,str1,str2,str3,...) Returns @code{str1} if @code{N} = @code{1}, @code{str2} if @code{N} = @code{2}, and so on. Returns @code{NULL} if @code{N} is less than @code{1} or greater than the number of arguments. @code{ELT()} is the complement of @code{FIELD()}: @example mysql> select ELT(1, 'ej', 'Heja', 'hej', 'foo'); -> 'ej' mysql> select ELT(4, 'ej', 'Heja', 'hej', 'foo'); -> 'foo' @end example @findex FIELD() @item FIELD(str,str1,str2,str3,...) Returns the index of @code{str} in the @code{str1}, @code{str2}, @code{str3}, @code{...} list. Returns @code{0} if @code{str} is not found. @code{FIELD()} is the complement of @code{ELT()}: @example mysql> select FIELD('ej', 'Hej', 'ej', 'Heja', 'hej', 'foo'); -> 2 mysql> select FIELD('fo', 'Hej', 'ej', 'Heja', 'hej', 'foo'); -> 0 @end example @findex FIND_IN_SET() @item FIND_IN_SET(str,strlist) Returns a value @code{1} to @code{N} if the string @code{str} is in the list @code{strlist} consisting of @code{N} substrings. A string list is a string composed of substrings separated by @samp{,} characters. If the first argument is a constant string and the second is a column of type @code{SET}, the @code{FIND_IN_SET()} function is optimized to use bit arithmetic! Returns @code{0} if @code{str} is not in @code{strlist} or if @code{strlist} is the empty string. Returns @code{NULL} if either argument is @code{NULL}. This function will not work properly if the first argument contains a @samp{,}: @example mysql> SELECT FIND_IN_SET('b','a,b,c,d'); -> 2 @end example @findex MAKE_SET() @item MAKE_SET(bits,str1,str2,...) Returns a set (a string containing substrings separated by @samp{,} characters) consisting of the strings that have the corresponding bit in @code{bits} set. @code{str1} corresponds to bit 0, @code{str2} to bit 1, etc. @code{NULL} strings in @code{str1}, @code{str2}, @code{...} are not appended to the result: @example mysql> SELECT MAKE_SET(1,'a','b','c'); -> 'a' mysql> SELECT MAKE_SET(1 | 4,'hello','nice','world'); -> 'hello,world' mysql> SELECT MAKE_SET(0,'a','b','c'); -> '' @end example @findex EXPORT_SET() @item EXPORT_SET(bits,on,off,[separator,[number_of_bits]]) Returns a string where for every bit set in 'bit', you get an 'on' string and for every reset bit you get an 'off' string. Each string is separated with 'separator' (default ',') and only 'number_of_bits' (default 64) of 'bits' is used: @example mysql> select EXPORT_SET(5,'Y','N',',',4) -> Y,N,Y,N @end example @findex LCASE() @findex LOWER() @item LCASE(str) @itemx LOWER(str) Returns the string @code{str} with all characters changed to lowercase according to the current character set mapping (the default is ISO-8859-1 Latin1): @example mysql> select LCASE('QUADRATICALLY'); -> 'quadratically' @end example This function is multi-byte safe. @findex UCASE() @findex UPPER() @item UCASE(str) @itemx UPPER(str) Returns the string @code{str} with all characters changed to uppercase according to the current character set mapping (the default is ISO-8859-1 Latin1): @example mysql> select UCASE('Hej'); -> 'HEJ' @end example This function is multi-byte safe. @findex FILE @findex LOAD_FILE() @item LOAD_FILE(file_name) Reads the file and returns the file contents as a string. The file must be on the server, you must specify the full pathname to the file, and you must have the @strong{file} privilege. The file must be readable by all and be smaller than @code{max_allowed_packet}. If the file doesn't exist or can't be read due to one of the above reasons, the function returns @code{NULL}: @example mysql> UPDATE table_name SET blob_column=LOAD_FILE("/tmp/picture") WHERE id=1; @end example @end table If you are not using MySQL Version 3.23, you have to do the reading of the file inside your application and create an @code{INSERT} statement to update the database with the file information. One way to do this, if you are using the MySQL++ library, can be found at @uref{http://www.mysql.com/documentation/mysql++/mysql++-examples.html}. MySQL automatically converts numbers to strings as necessary, and vice-versa: @example mysql> SELECT 1+"1"; -> 2 mysql> SELECT CONCAT(2,' test'); -> '2 test' @end example If you want to convert a number to a string explicitly, pass it as the argument to @code{CONCAT()}. If a string function is given a binary string as an argument, the resulting string is also a binary string. A number converted to a string is treated as a binary string. This only affects comparisons. @menu * String comparison functions:: String Comparison Functions * Case Sensitivity Operators:: Case Sensitivity @end menu @node String comparison functions, Case Sensitivity Operators, String functions, String functions @subsubsection String Comparison Functions @findex string comparison functions @findex functions, string comparison @cindex case sensitivity, in string comparisons @cindex string comparisons, case sensitivity Normally, if any expression in a string comparison is case sensitive, the comparison is performed in case-sensitive fashion. @table @code @findex LIKE @item expr LIKE pat [ESCAPE 'escape-char'] Pattern matching using SQL simple regular expression comparison. Returns @code{1} (TRUE) or @code{0} (FALSE). With @code{LIKE} you can use the following two wild-card characters in the pattern: @multitable @columnfractions .1 .9 @item @code{%} @tab Matches any number of characters, even zero characters @item @code{_} @tab Matches exactly one character @end multitable @example mysql> select 'David!' LIKE 'David_'; -> 1 mysql> select 'David!' LIKE '%D%v%'; -> 1 @end example To test for literal instances of a wild-card character, precede the character with the escape character. If you don't specify the @code{ESCAPE} character, @samp{\} is assumed: @multitable @columnfractions .1 .9 @item @code{\%} @tab Matches one @code{%} character @item @code{\_} @tab Matches one @code{_} character @end multitable @example mysql> select 'David!' LIKE 'David\_'; -> 0 mysql> select 'David_' LIKE 'David\_'; -> 1 @end example To specify a different escape character, use the @code{ESCAPE} clause: @example mysql> select 'David_' LIKE 'David|_' ESCAPE '|'; -> 1 @end example The following two statements illustrate that string comparisons are case insensitive unless one of the operands is a binary string: @example mysql> select 'abc' LIKE 'ABC'; -> 1 mysql> SELECT 'abc' LIKE BINARY 'ABC'; -> 0 @end example @code{LIKE} is allowed on numeric expressions! (This is a MySQL extension to the ANSI SQL @code{LIKE}.) @example mysql> select 10 LIKE '1%'; -> 1 @end example Note: Because MySQL uses the C escape syntax in strings (for example, @samp{\n}), you must double any @samp{\} that you use in your @code{LIKE} strings. For example, to search for @samp{\n}, specify it as @samp{\\n}. To search for @samp{\}, specify it as @samp{\\\\} (the backslashes are stripped once by the parser and another time when the pattern match is done, leaving a single backslash to be matched). @findex NOT LIKE @item expr NOT LIKE pat [ESCAPE 'escape-char'] Same as @code{NOT (expr LIKE pat [ESCAPE 'escape-char'])}. @cindex mSQL compatibility @cindex compatibility, with mSQL @findex REGEXP @findex RLIKE @item expr REGEXP pat @itemx expr RLIKE pat Performs a pattern match of a string expression @code{expr} against a pattern @code{pat}. The pattern can be an extended regular expression. @xref{Regexp}. Returns @code{1} if @code{expr} matches @code{pat}, otherwise returns @code{0}. @code{RLIKE} is a synonym for @code{REGEXP}, provided for @code{mSQL} compatibility. Note: Because MySQL uses the C escape syntax in strings (for example, @samp{\n}), you must double any @samp{\} that you use in your @code{REGEXP} strings. As of MySQL Version 3.23.4, @code{REGEXP} is case insensitive for normal (not binary) strings: @example mysql> select 'Monty!' REGEXP 'm%y%%'; -> 0 mysql> select 'Monty!' REGEXP '.*'; -> 1 mysql> select 'new*\n*line' REGEXP 'new\\*.\\*line'; -> 1 mysql> select "a" REGEXP "A", "a" REGEXP BINARY "A"; -> 1 0 mysql> select "a" REGEXP "^[a-d]"; -> 1 @end example @item @code{REGEXP} and @code{RLIKE} use the current character set (ISO-8859-1 Latin1 by default) when deciding the type of a character. @findex NOT REGEXP @item expr NOT REGEXP pat @itemx expr NOT RLIKE pat Same as @code{NOT (expr REGEXP pat)}. @findex STRCMP() @item STRCMP(expr1,expr2) @code{STRCMP()} returns @code{0} if the strings are the same, @code{-1} if the first argument is smaller than the second according to the current sort order, and @code{1} otherwise: @example mysql> select STRCMP('text', 'text2'); -> -1 mysql> select STRCMP('text2', 'text'); -> 1 mysql> select STRCMP('text', 'text'); -> 0 @end example @findex MATCH ... AGAINST() @item MATCH (col1,col2,...) AGAINST (expr) @code{MATCH ... AGAINST()} is used for full-text search and returns relevance - similarity measure between the text in columns @code{(col1,col2,...)} and the query @code{expr}. Relevance is a positive floating-point number. Zero relevance means no similarity. For @code{MATCH ... AGAINST()} to work, a @strong{FULLTEXT} index must be created first. @xref{CREATE TABLE, , @code{CREATE TABLE}}. @code{MATCH ... AGAINST()} is available in MySQL Version 3.23.23 or later. For details and usage examples @pxref{Fulltext Search}. @end table @node Case Sensitivity Operators, , String comparison functions, String functions @subsubsection Case Sensitivity @findex casts @cindex cast operators @cindex operators, cast @table @code @findex BINARY @item @code{BINARY} The @code{BINARY} operator casts the string following it to a binary string. This is an easy way to force a column comparison to be case sensitive even if the column isn't defined as @code{BINARY} or @code{BLOB}: @example mysql> select "a" = "A"; -> 1 mysql> select BINARY "a" = "A"; -> 0 @end example @code{BINARY} was introduced in MySQL Version 3.23.0. Note that in some context MySQL will not be able to use the index efficiently when you cast an indexed column to @code{BINARY}. @end table If you want to compare a blob case-insensitively you can always convert the blob to upper case before doing the comparison: @example SELECT 'A' LIKE UPPER(blob_col) FROM table_name; @end example We plan to soon introduce casting between different character sets to make string comparison even more flexible. @node Numeric Functions, Date and time functions, String functions, Functions @subsection Numeric Functions @menu * Arithmetic functions:: Arithmetic Operations * Mathematical functions:: Mathematical Functions @end menu @node Arithmetic functions, Mathematical functions, Numeric Functions, Numeric Functions @subsubsection Arithmetic Operations The usual arithmetic operators are available. Note that in the case of @samp{-}, @samp{+}, and @samp{*}, the result is calculated with @code{BIGINT} (64-bit) precision if both arguments are integers! @cindex operations, arithmetic @cindex arithmetic expressions @table @code @findex + (addition) @findex addition (+) @item + Addition: @example mysql> select 3+5; -> 8 @end example @findex - (subtraction) @findex subtraction (-) @item - Subtraction: @example mysql> select 3-5; -> -2 @end example @findex * (multiplication) @findex multiplication (*) @item * Multiplication: @example mysql> select 3*5; -> 15 mysql> select 18014398509481984*18014398509481984.0; -> 324518553658426726783156020576256.0 mysql> select 18014398509481984*18014398509481984; -> 0 @end example The result of the last expression is incorrect because the result of the integer multiplication exceeds the 64-bit range of @code{BIGINT} calculations. @findex / (division) @findex division (/) @item / Division: @example mysql> select 3/5; -> 0.60 @end example Division by zero produces a @code{NULL} result: @example mysql> select 102/(1-1); -> NULL @end example A division will be calculated with @code{BIGINT} arithmetic only if performed in a context where its result is converted to an integer! @end table @node Mathematical functions, , Arithmetic functions, Numeric Functions @subsubsection Mathematical Functions All mathematical functions return @code{NULL} in case of an error. @findex mathematical functions @findex functions, mathematical @table @code @findex - (unary minus) @findex minus, unary (-) @findex unary minus (-) @item - Unary minus. Changes the sign of the argument: @example mysql> select - 2; -> -2 @end example Note that if this operator is used with a @code{BIGINT}, the return value is a @code{BIGINT}! This means that you should avoid using @code{-} on integers that may have the value of @code{-2^63}! @findex ABS() @item ABS(X) Returns the absolute value of @code{X}: @example mysql> select ABS(2); -> 2 mysql> select ABS(-32); -> 32 @end example This function is safe to use with @code{BIGINT} values. @findex SIGN() @item SIGN(X) Returns the sign of the argument as @code{-1}, @code{0}, or @code{1}, depending on whether @code{X} is negative, zero, or positive: @example mysql> select SIGN(-32); -> -1 mysql> select SIGN(0); -> 0 mysql> select SIGN(234); -> 1 @end example @findex MOD() @findex % (modulo) @findex modulo (%) @item MOD(N,M) @itemx % Modulo (like the @code{%} operator in C). Returns the remainder of @code{N} divided by @code{M}: @example mysql> select MOD(234, 10); -> 4 mysql> select 253 % 7; -> 1 mysql> select MOD(29,9); -> 2 @end example This function is safe to use with @code{BIGINT} values. @findex FLOOR() @item FLOOR(X) Returns the largest integer value not greater than @code{X}: @example mysql> select FLOOR(1.23); -> 1 mysql> select FLOOR(-1.23); -> -2 @end example Note that the return value is converted to a @code{BIGINT}! @findex CEILING() @item CEILING(X) Returns the smallest integer value not less than @code{X}: @example mysql> select CEILING(1.23); -> 2 mysql> select CEILING(-1.23); -> -1 @end example Note that the return value is converted to a @code{BIGINT}! @findex ROUND() @item ROUND(X) Returns the argument @code{X}, rounded to the nearest integer: @example mysql> select ROUND(-1.23); -> -1 mysql> select ROUND(-1.58); -> -2 mysql> select ROUND(1.58); -> 2 @end example Note that the behavior of @code{ROUND()} when the argument is half way between two integers depends on the C library implementation. Some round to the nearest even number, always up, always down, or always towards zero. If you need one kind of rounding, you should use a well-defined function like @code{TRUNCATE()} or @code{FLOOR()} instead. @findex ROUND() @item ROUND(X,D) Returns the argument @code{X}, rounded to a number with @code{D} decimals. If @code{D} is @code{0}, the result will have no decimal point or fractional part: @example mysql> select ROUND(1.298, 1); -> 1.3 mysql> select ROUND(1.298, 0); -> 1 @end example @findex EXP() @item EXP(X) Returns the value of @code{e} (the base of natural logarithms) raised to the power of @code{X}: @example mysql> select EXP(2); -> 7.389056 mysql> select EXP(-2); -> 0.135335 @end example @findex LOG() @item LOG(X) Returns the natural logarithm of @code{X}: @example mysql> select LOG(2); -> 0.693147 mysql> select LOG(-2); -> NULL @end example If you want the log of a number @code{X} to some arbitary base @code{B}, use the formula @code{LOG(X)/LOG(B)}. @findex LOG10() @item LOG10(X) Returns the base-10 logarithm of @code{X}: @example mysql> select LOG10(2); -> 0.301030 mysql> select LOG10(100); -> 2.000000 mysql> select LOG10(-100); -> NULL @end example @findex POW() @findex POWER() @item POW(X,Y) @itemx POWER(X,Y) Returns the value of @code{X} raised to the power of @code{Y}: @example mysql> select POW(2,2); -> 4.000000 mysql> select POW(2,-2); -> 0.250000 @end example @findex SQRT() @item SQRT(X) Returns the non-negative square root of @code{X}: @example mysql> select SQRT(4); -> 2.000000 mysql> select SQRT(20); -> 4.472136 @end example @findex PI() @item PI() Returns the value of PI. The default shown number of decimals is 5, but MySQL internally uses the full double precession for PI. @example mysql> select PI(); -> 3.141593 mysql> SELECT PI()+0.000000000000000000; -> 3.141592653589793116 @end example @findex COS() @item COS(X) Returns the cosine of @code{X}, where @code{X} is given in radians: @example mysql> select COS(PI()); -> -1.000000 @end example @findex SIN() @item SIN(X) Returns the sine of @code{X}, where @code{X} is given in radians: @example mysql> select SIN(PI()); -> 0.000000 @end example @findex TAN() @item TAN(X) Returns the tangent of @code{X}, where @code{X} is given in radians: @example mysql> select TAN(PI()+1); -> 1.557408 @end example @findex ACOS() @item ACOS(X) Returns the arc cosine of @code{X}, that is, the value whose cosine is @code{X}. Returns @code{NULL} if @code{X} is not in the range @code{-1} to @code{1}: @example mysql> select ACOS(1); -> 0.000000 mysql> select ACOS(1.0001); -> NULL mysql> select ACOS(0); -> 1.570796 @end example @findex ASIN() @item ASIN(X) Returns the arc sine of @code{X}, that is, the value whose sine is @code{X}. Returns @code{NULL} if @code{X} is not in the range @code{-1} to @code{1}: @example mysql> select ASIN(0.2); -> 0.201358 mysql> select ASIN('foo'); -> 0.000000 @end example @findex ATAN() @item ATAN(X) Returns the arc tangent of @code{X}, that is, the value whose tangent is @code{X}: @example mysql> select ATAN(2); -> 1.107149 mysql> select ATAN(-2); -> -1.107149 @end example @findex ATAN2() @item ATAN(Y,X) @itemx ATAN2(Y,X) Returns the arc tangent of the two variables @code{X} and @code{Y}. It is similar to calculating the arc tangent of @code{Y / X}, except that the signs of both arguments are used to determine the quadrant of the result: @example mysql> select ATAN(-2,2); -> -0.785398 mysql> select ATAN2(PI(),0); -> 1.570796 @end example @findex COT() @item COT(X) Returns the cotangent of @code{X}: @example mysql> select COT(12); -> -1.57267341 mysql> select COT(0); -> NULL @end example @findex RAND() @item RAND() @itemx RAND(N) Returns a random floating-point value in the range @code{0} to @code{1.0}. If an integer argument @code{N} is specified, it is used as the seed value: @example mysql> select RAND(); -> 0.5925 mysql> select RAND(20); -> 0.1811 mysql> select RAND(20); -> 0.1811 mysql> select RAND(); -> 0.2079 mysql> select RAND(); -> 0.7888 @end example You can't use a column with @code{RAND()} values in an @code{ORDER BY} clause, because @code{ORDER BY} would evaluate the column multiple times. In MySQL Version 3.23, you can, however, do: @code{SELECT * FROM table_name ORDER BY RAND()} This is useful to get a random sample of a set @code{SELECT * FROM table1,table2 WHERE a=b AND c<d ORDER BY RAND() LIMIT 1000}. Note that a @code{RAND()} in a @code{WHERE} clause will be re-evaluated every time the @code{WHERE} is executed. @findex LEAST() @item LEAST(X,Y,...) With two or more arguments, returns the smallest (minimum-valued) argument. The arguments are compared using the following rules: @itemize @bullet @item If the return value is used in an @code{INTEGER} context, or all arguments are integer-valued, they are compared as integers. @item If the return value is used in a @code{REAL} context, or all arguments are real-valued, they are compared as reals. @item If any argument is a case-sensitive string, the arguments are compared as case-sensitive strings. @item In other cases, the arguments are compared as case-insensitive strings: @end itemize @example mysql> select LEAST(2,0); -> 0 mysql> select LEAST(34.0,3.0,5.0,767.0); -> 3.0 mysql> select LEAST("B","A","C"); -> "A" @end example In MySQL versions prior to Version 3.22.5, you can use @code{MIN()} instead of @code{LEAST}. @findex GREATEST() @item GREATEST(X,Y,...) Returns the largest (maximum-valued) argument. The arguments are compared using the same rules as for @code{LEAST}: @example mysql> select GREATEST(2,0); -> 2 mysql> select GREATEST(34.0,3.0,5.0,767.0); -> 767.0 mysql> select GREATEST("B","A","C"); -> "C" @end example In MySQL versions prior to Version 3.22.5, you can use @code{MAX()} instead of @code{GREATEST}. @findex DEGREES() @item DEGREES(X) Returns the argument @code{X}, converted from radians to degrees: @example mysql> select DEGREES(PI()); -> 180.000000 @end example @findex RADIANS() @item RADIANS(X) Returns the argument @code{X}, converted from degrees to radians: @example mysql> select RADIANS(90); -> 1.570796 @end example @findex TRUNCATE() @item TRUNCATE(X,D) Returns the number @code{X}, truncated to @code{D} decimals. If @code{D} is @code{0}, the result will have no decimal point or fractional part: @example mysql> select TRUNCATE(1.223,1); -> 1.2 mysql> select TRUNCATE(1.999,1); -> 1.9 mysql> select TRUNCATE(1.999,0); -> 1 mysql> select TRUNCATE(-1.999,1); -> -1.9 @end example Starting from MySQL 3.23.51 all numbers are rounded towards zero. If @code{D} is negative, then the whole part of the number is zeroed out: @example mysql> select truncate(122,-2); -> 100 @end example Note that as decimal numbers are normally not stored as exact numbers in computers, but as double values, you may be fooled by the following result: @cindex rounding errors @example mysql> select TRUNCATE(10.28*100,0); -> 1027 @end example The above happens because 10.28 is actually stored as something like 10.2799999999999999. @end table @node Date and time functions, Other Functions, Numeric Functions, Functions @subsection Date and Time Functions @findex date and time functions @findex functions, date and time See @ref{Date and time types} for a description of the range of values each type has and the valid formats in which date and time values may be specified. Here is an example that uses date functions. The query below selects all records with a @code{date_col} value from within the last 30 days: @example mysql> SELECT something FROM table WHERE TO_DAYS(NOW()) - TO_DAYS(date_col) <= 30; @end example @table @code @findex DAYOFWEEK() @item DAYOFWEEK(date) Returns the weekday index for @code{date} (@code{1} = Sunday, @code{2} = Monday, ... @code{7} = Saturday). These index values correspond to the ODBC standard: @example mysql> select DAYOFWEEK('1998-02-03'); -> 3 @end example @findex WEEKDAY() @item WEEKDAY(date) Returns the weekday index for @code{date} (@code{0} = Monday, @code{1} = Tuesday, ... @code{6} = Sunday): @example mysql> select WEEKDAY('1997-10-04 22:23:00'); -> 5 mysql> select WEEKDAY('1997-11-05'); -> 2 @end example @findex DAYOFMONTH() @item DAYOFMONTH(date) Returns the day of the month for @code{date}, in the range @code{1} to @code{31}: @example mysql> select DAYOFMONTH('1998-02-03'); -> 3 @end example @findex DAYOFYEAR() @item DAYOFYEAR(date) Returns the day of the year for @code{date}, in the range @code{1} to @code{366}: @example mysql> select DAYOFYEAR('1998-02-03'); -> 34 @end example @findex MONTH() @item MONTH(date) Returns the month for @code{date}, in the range @code{1} to @code{12}: @example mysql> select MONTH('1998-02-03'); -> 2 @end example @findex DAYNAME() @item DAYNAME(date) Returns the name of the weekday for @code{date}: @example mysql> select DAYNAME("1998-02-05"); -> 'Thursday' @end example @findex MONTHNAME() @item MONTHNAME(date) Returns the name of the month for @code{date}: @example mysql> select MONTHNAME("1998-02-05"); -> 'February' @end example @findex QUARTER() @item QUARTER(date) Returns the quarter of the year for @code{date}, in the range @code{1} to @code{4}: @example mysql> select QUARTER('98-04-01'); -> 2 @end example @findex WEEK() @item WEEK(date) @itemx WEEK(date,first) With a single argument, returns the week for @code{date}, in the range @code{0} to @code{53} (yes, there may be the beginnings of a week 53), for locations where Sunday is the first day of the week. The two-argument form of @code{WEEK()} allows you to specify whether the week starts on Sunday or Monday. The week starts on Sunday if the second argument is @code{0}, on Monday if the second argument is @code{1}: @example mysql> select WEEK('1998-02-20'); -> 7 mysql> select WEEK('1998-02-20',0); -> 7 mysql> select WEEK('1998-02-20',1); -> 8 mysql> select WEEK('1998-12-31',1); -> 53 @end example @findex YEAR() @item YEAR(date) Returns the year for @code{date}, in the range @code{1000} to @code{9999}: @example mysql> select YEAR('98-02-03'); -> 1998 @end example @item YEARWEEK(date) @itemx YEARWEEK(date,first) Returns year and week for a date. The second arguments works exactly like the second argument to @code{WEEK()}. Note that the year may be different from the year in the date argument for the first and the last week of the year: @example mysql> select YEARWEEK('1987-01-01'); -> 198653 @end example @findex HOUR() @item HOUR(time) Returns the hour for @code{time}, in the range @code{0} to @code{23}: @example mysql> select HOUR('10:05:03'); -> 10 @end example @findex MINUTE() @item MINUTE(time) Returns the minute for @code{time}, in the range @code{0} to @code{59}: @example mysql> select MINUTE('98-02-03 10:05:03'); -> 5 @end example @findex SECOND() @item SECOND(time) Returns the second for @code{time}, in the range @code{0} to @code{59}: @example mysql> select SECOND('10:05:03'); -> 3 @end example @findex PERIOD_ADD() @item PERIOD_ADD(P,N) Adds @code{N} months to period @code{P} (in the format @code{YYMM} or @code{YYYYMM}). Returns a value in the format @code{YYYYMM}. Note that the period argument @code{P} is @emph{not} a date value: @example mysql> select PERIOD_ADD(9801,2); -> 199803 @end example @findex PERIOD_DIFF() @item PERIOD_DIFF(P1,P2) Returns the number of months between periods @code{P1} and @code{P2}. @code{P1} and @code{P2} should be in the format @code{YYMM} or @code{YYYYMM}. Note that the period arguments @code{P1} and @code{P2} are @emph{not} date values: @example mysql> select PERIOD_DIFF(9802,199703); -> 11 @end example @findex DATE_ADD() @findex DATE_SUB() @findex ADDDATE() @findex SUBDATE() @findex EXTRACT() @item DATE_ADD(date,INTERVAL expr type) @itemx DATE_SUB(date,INTERVAL expr type) @itemx ADDDATE(date,INTERVAL expr type) @itemx SUBDATE(date,INTERVAL expr type) These functions perform date arithmetic. They are new for MySQL Version 3.22. @code{ADDDATE()} and @code{SUBDATE()} are synonyms for @code{DATE_ADD()} and @code{DATE_SUB()}. In MySQL Version 3.23, you can use @code{+} and @code{-} instead of @code{DATE_ADD()} and @code{DATE_SUB()} if the expression on the right side is a date or datetime column. (See example) @code{date} is a @code{DATETIME} or @code{DATE} value specifying the starting date. @code{expr} is an expression specifying the interval value to be added or subtracted from the starting date. @code{expr} is a string; it may start with a @samp{-} for negative intervals. @code{type} is a keyword indicating how the expression should be interpreted. The related function @code{EXTRACT(type FROM date)} returns the 'type' interval from the date. The following table shows how the @code{type} and @code{expr} arguments are related: @multitable @columnfractions .5 .5 @item @code{type} @strong{value} @tab @strong{Expected} @code{expr} @strong{format} @item @code{SECOND} @tab @code{SECONDS} @item @code{MINUTE} @tab @code{MINUTES} @item @code{HOUR} @tab @code{HOURS} @item @code{DAY} @tab @code{DAYS} @item @code{MONTH} @tab @code{MONTHS} @item @code{YEAR} @tab @code{YEARS} @item @code{MINUTE_SECOND} @tab @code{"MINUTES:SECONDS"} @item @code{HOUR_MINUTE} @tab @code{"HOURS:MINUTES"} @item @code{DAY_HOUR} @tab @code{"DAYS HOURS"} @item @code{YEAR_MONTH} @tab @code{"YEARS-MONTHS"} @item @code{HOUR_SECOND} @tab @code{"HOURS:MINUTES:SECONDS"} @item @code{DAY_MINUTE} @tab @code{"DAYS HOURS:MINUTES"} @item @code{DAY_SECOND} @tab @code{"DAYS HOURS:MINUTES:SECONDS"} @end multitable MySQL allows any punctuation delimiter in the @code{expr} format. Those shown in the table are the suggested delimiters. If the @code{date} argument is a @code{DATE} value and your calculations involve only @code{YEAR}, @code{MONTH}, and @code{DAY} parts (that is, no time parts), the result is a @code{DATE} value. Otherwise the result is a @code{DATETIME} value: @example mysql> SELECT "1997-12-31 23:59:59" + INTERVAL 1 SECOND; -> 1998-01-01 00:00:00 mysql> SELECT INTERVAL 1 DAY + "1997-12-31"; -> 1998-01-01 mysql> SELECT "1998-01-01" - INTERVAL 1 SECOND; -> 1997-12-31 23:59:59 mysql> SELECT DATE_ADD("1997-12-31 23:59:59", INTERVAL 1 SECOND); -> 1998-01-01 00:00:00 mysql> SELECT DATE_ADD("1997-12-31 23:59:59", INTERVAL 1 DAY); -> 1998-01-01 23:59:59 mysql> SELECT DATE_ADD("1997-12-31 23:59:59", INTERVAL "1:1" MINUTE_SECOND); -> 1998-01-01 00:01:00 mysql> SELECT DATE_SUB("1998-01-01 00:00:00", INTERVAL "1 1:1:1" DAY_SECOND); -> 1997-12-30 22:58:59 mysql> SELECT DATE_ADD("1998-01-01 00:00:00", INTERVAL "-1 10" DAY_HOUR); -> 1997-12-30 14:00:00 mysql> SELECT DATE_SUB("1998-01-02", INTERVAL 31 DAY); -> 1997-12-02 @end example If you specify an interval value that is too short (does not include all the interval parts that would be expected from the @code{type} keyword), MySQL assumes you have left out the leftmost parts of the interval value. For example, if you specify a @code{type} of @code{DAY_SECOND}, the value of @code{expr} is expected to have days, hours, minutes, and seconds parts. If you specify a value like @code{"1:10"}, MySQL assumes that the days and hours parts are missing and the value represents minutes and seconds. In other words, @code{"1:10" DAY_SECOND} is interpreted in such a way that it is equivalent to @code{"1:10" MINUTE_SECOND}. This is analogous to the way that MySQL interprets @code{TIME} values as representing elapsed time rather than as time of day. Note that if you add or subtract a date value against something that contains a time part, the date value will be automatically converted to a datetime value: @example mysql> select date_add("1999-01-01", interval 1 day); -> 1999-01-02 mysql> select date_add("1999-01-01", interval 1 hour); -> 1999-01-01 01:00:00 @end example If you use really incorrect dates, the result is @code{NULL}. If you add @code{MONTH}, @code{YEAR_MONTH}, or @code{YEAR} and the resulting date has a day that is larger than the maximum day for the new month, the day is adjusted to the maximum days in the new month: @example mysql> select DATE_ADD('1998-01-30', Interval 1 month); -> 1998-02-28 @end example Note from the preceding example that the word @code{INTERVAL} and the @code{type} keyword are not case sensitive. @findex EXTRACT() @item EXTRACT(type FROM date) The @code{EXTRACT()} function uses the same kinds of interval type specifiers as @code{DATE_ADD()} or @code{DATE_SUB()}, but extracts parts from the date rather than performing date arithmetic. @example mysql> SELECT EXTRACT(YEAR FROM "1999-07-02"); -> 1999 mysql> SELECT EXTRACT(YEAR_MONTH FROM "1999-07-02 01:02:03"); -> 199907 mysql> SELECT EXTRACT(DAY_MINUTE FROM "1999-07-02 01:02:03"); -> 20102 @end example @findex TO_DAYS() @item TO_DAYS(date) Given a date @code{date}, returns a daynumber (the number of days since year 0): @example mysql> select TO_DAYS(950501); -> 728779 mysql> select TO_DAYS('1997-10-07'); -> 729669 @end example @code{TO_DAYS()} is not intended for use with values that precede the advent of the Gregorian calendar (1582), because it doesn't take into account the days that were lost when the calendar was changed. @findex FROM_DAYS() @item FROM_DAYS(N) Given a daynumber @code{N}, returns a @code{DATE} value: @example mysql> select FROM_DAYS(729669); -> '1997-10-07' @end example @code{FROM_DAYS()} is not intended for use with values that precede the advent of the Gregorian calendar (1582), because it doesn't take into account the days that were lost when the calendar was changed. @findex DATE_FORMAT() @item DATE_FORMAT(date,format) Formats the @code{date} value according to the @code{format} string. The following specifiers may be used in the @code{format} string: @multitable @columnfractions .1 .6 @item @code{%M} @tab Month name (@code{January}..@code{December}) @item @code{%W} @tab Weekday name (@code{Sunday}..@code{Saturday}) @item @code{%D} @tab Day of the month with English suffix (@code{1st}, @code{2nd}, @code{3rd}, etc.) @item @code{%Y} @tab Year, numeric, 4 digits @item @code{%y} @tab Year, numeric, 2 digits @item @code{%X} @tab Year for the week where Sunday is the first day of the week, numeric, 4 digits, used with '%V' @item @code{%x} @tab Year for the week, where Monday is the first day of the week, numeric, 4 digits, used with '%v' @item @code{%a} @tab Abbreviated weekday name (@code{Sun}..@code{Sat}) @item @code{%d} @tab Day of the month, numeric (@code{00}..@code{31}) @item @code{%e} @tab Day of the month, numeric (@code{0}..@code{31}) @item @code{%m} @tab Month, numeric (@code{01}..@code{12}) @item @code{%c} @tab Month, numeric (@code{1}..@code{12}) @item @code{%b} @tab Abbreviated month name (@code{Jan}..@code{Dec}) @item @code{%j} @tab Day of year (@code{001}..@code{366}) @item @code{%H} @tab Hour (@code{00}..@code{23}) @item @code{%k} @tab Hour (@code{0}..@code{23}) @item @code{%h} @tab Hour (@code{01}..@code{12}) @item @code{%I} @tab Hour (@code{01}..@code{12}) @item @code{%l} @tab Hour (@code{1}..@code{12}) @item @code{%i} @tab Minutes, numeric (@code{00}..@code{59}) @item @code{%r} @tab Time, 12-hour (@code{hh:mm:ss [AP]M}) @item @code{%T} @tab Time, 24-hour (@code{hh:mm:ss}) @item @code{%S} @tab Seconds (@code{00}..@code{59}) @item @code{%s} @tab Seconds (@code{00}..@code{59}) @item @code{%p} @tab @code{AM} or @code{PM} @item @code{%w} @tab Day of the week (@code{0}=Sunday..@code{6}=Saturday) @item @code{%U} @tab Week (@code{0}..@code{53}), where Sunday is the first day of the week @item @code{%u} @tab Week (@code{0}..@code{53}), where Monday is the first day of the week @item @code{%V} @tab Week (@code{1}..@code{53}), where Sunday is the first day of the week. Used with '%X' @item @code{%v} @tab Week (@code{1}..@code{53}), where Monday is the first day of the week. Used with '%x' @item @code{%%} @tab A literal @samp{%}. @end multitable All other characters are just copied to the result without interpretation: @example mysql> select DATE_FORMAT('1997-10-04 22:23:00', '%W %M %Y'); -> 'Saturday October 1997' mysql> select DATE_FORMAT('1997-10-04 22:23:00', '%H:%i:%s'); -> '22:23:00' mysql> select DATE_FORMAT('1997-10-04 22:23:00', '%D %y %a %d %m %b %j'); -> '4th 97 Sat 04 10 Oct 277' mysql> select DATE_FORMAT('1997-10-04 22:23:00', '%H %k %I %r %T %S %w'); -> '22 22 10 10:23:00 PM 22:23:00 00 6' mysql> select DATE_FORMAT('1999-01-01', '%X %V'); -> '1998 52' @end example As of MySQL Version 3.23, the @samp{%} character is required before format specifier characters. In earlier versions of MySQL, @samp{%} was optional. @findex TIME_FORMAT() @item TIME_FORMAT(time,format) This is used like the @code{DATE_FORMAT()} function above, but the @code{format} string may contain only those format specifiers that handle hours, minutes, and seconds. Other specifiers produce a @code{NULL} value or @code{0}. @findex CURDATE() @findex CURRENT_DATE @item CURDATE() @itemx CURRENT_DATE Returns today's date as a value in @code{'YYYY-MM-DD'} or @code{YYYYMMDD} format, depending on whether the function is used in a string or numeric context: @example mysql> select CURDATE(); -> '1997-12-15' mysql> select CURDATE() + 0; -> 19971215 @end example @findex CURTIME() @findex CURRENT_TIME @item CURTIME() @itemx CURRENT_TIME Returns the current time as a value in @code{'HH:MM:SS'} or @code{HHMMSS} format, depending on whether the function is used in a string or numeric context: @example mysql> select CURTIME(); -> '23:50:26' mysql> select CURTIME() + 0; -> 235026 @end example @findex NOW() @findex SYSDATE() @findex CURRENT_TIMESTAMP @item NOW() @itemx SYSDATE() @itemx CURRENT_TIMESTAMP Returns the current date and time as a value in @code{'YYYY-MM-DD HH:MM:SS'} or @code{YYYYMMDDHHMMSS} format, depending on whether the function is used in a string or numeric context: @example mysql> select NOW(); -> '1997-12-15 23:50:26' mysql> select NOW() + 0; -> 19971215235026 @end example @findex UNIX_TIMESTAMP() @item UNIX_TIMESTAMP() @itemx UNIX_TIMESTAMP(date) If called with no argument, returns a Unix timestamp (seconds since @code{'1970-01-01 00:00:00'} GMT). If @code{UNIX_TIMESTAMP()} is called with a @code{date} argument, it returns the value of the argument as seconds since @code{'1970-01-01 00:00:00'} GMT. @code{date} may be a @code{DATE} string, a @code{DATETIME} string, a @code{TIMESTAMP}, or a number in the format @code{YYMMDD} or @code{YYYYMMDD} in local time: @example mysql> select UNIX_TIMESTAMP(); -> 882226357 mysql> select UNIX_TIMESTAMP('1997-10-04 22:23:00'); -> 875996580 @end example When @code{UNIX_TIMESTAMP} is used on a @code{TIMESTAMP} column, the function will receive the value directly, with no implicit ``string-to-unix-timestamp'' conversion. If you give @code{UNIX_TIMESTAMP()} a wrong or out-of-range date, it will return 0. @findex FROM_UNIXTIME() @item FROM_UNIXTIME(unix_timestamp) Returns a representation of the @code{unix_timestamp} argument as a value in @code{'YYYY-MM-DD HH:MM:SS'} or @code{YYYYMMDDHHMMSS} format, depending on whether the function is used in a string or numeric context: @example mysql> select FROM_UNIXTIME(875996580); -> '1997-10-04 22:23:00' mysql> select FROM_UNIXTIME(875996580) + 0; -> 19971004222300 @end example @findex FROM_UNIXTIME() @item FROM_UNIXTIME(unix_timestamp,format) Returns a string representation of the Unix timestamp, formatted according to the @code{format} string. @code{format} may contain the same specifiers as those listed in the entry for the @code{DATE_FORMAT()} function: @example mysql> select FROM_UNIXTIME(UNIX_TIMESTAMP(), '%Y %D %M %h:%i:%s %x'); -> '1997 23rd December 03:43:30 x' @end example @findex SEC_TO_TIME() @item SEC_TO_TIME(seconds) Returns the @code{seconds} argument, converted to hours, minutes, and seconds, as a value in @code{'HH:MM:SS'} or @code{HHMMSS} format, depending on whether the function is used in a string or numeric context: @example mysql> select SEC_TO_TIME(2378); -> '00:39:38' mysql> select SEC_TO_TIME(2378) + 0; -> 3938 @end example @findex TIME_TO_SEC() @item TIME_TO_SEC(time) Returns the @code{time} argument, converted to seconds: @example mysql> select TIME_TO_SEC('22:23:00'); -> 80580 mysql> select TIME_TO_SEC('00:39:38'); -> 2378 @end example @end table @node Other Functions, , Date and time functions, Functions @subsection Other Functions @menu * Bit functions:: Bit Functions * Miscellaneous functions:: Miscellaneous Functions @end menu @node Bit functions, Miscellaneous functions, Other Functions, Other Functions @subsubsection Bit Functions @findex arithmetic functions @findex bit functions @findex functions, arithmetic @findex functions, bit MySQL uses @code{BIGINT} (64-bit) arithmetic for bit operations, so these operators have a maximum range of 64 bits. @table @code @findex | (bitwise OR) @findex OR, bitwise @item | Bitwise OR: @example mysql> select 29 | 15; -> 31 @end example @findex & (bitwise AND) @findex AND, bitwise @item & Bitwise AND: @example mysql> select 29 & 15; -> 13 @end example @findex << (left shift) @item << Shifts a longlong (@code{BIGINT}) number to the left: @example mysql> select 1 << 2; -> 4 @end example @findex >> (right shift) @item >> Shifts a longlong (@code{BIGINT}) number to the right: @example mysql> select 4 >> 2; -> 1 @end example @findex ~ @item ~ Invert all bits: @example mysql> select 5 & ~1; -> 4 @end example @findex BIT_COUNT() @item BIT_COUNT(N) Returns the number of bits that are set in the argument @code{N}: @example mysql> select BIT_COUNT(29); -> 4 @end example @end table @node Miscellaneous functions, , Bit functions, Other Functions @subsubsection Miscellaneous Functions @findex miscellaneous functions @findex functions, miscellaneous @table @code @findex DATABASE() @item DATABASE() Returns the current database name: @example mysql> select DATABASE(); -> 'test' @end example If there is no current database, @code{DATABASE()} returns the empty string. @findex USER() @findex SYSTEM_USER() @findex SESSION_USER() @item USER() @itemx SYSTEM_USER() @itemx SESSION_USER() Returns the current MySQL user name: @example mysql> select USER(); -> 'davida@@localhost' @end example In MySQL Version 3.22.11 or later, this includes the client hostname as well as the user name. You can extract just the user name part like this (which works whether or not the value includes a hostname part): @example mysql> select substring_index(USER(),"@@",1); -> 'davida' @end example @findex PASSWORD() @item PASSWORD(str) Calculates a password string from the plaintext password @code{str}. This is the function that is used for encrypting MySQL passwords for storage in the @code{Password} column of the @code{user} grant table: @example mysql> select PASSWORD('badpwd'); -> '7f84554057dd964b' @end example @cindex password encryption, reversibility of @code{PASSWORD()} encryption is non-reversible. @code{PASSWORD()} does not perform password encryption in the same way that Unix passwords are encrypted. You should not assume that if your Unix password and your MySQL password are the same, @code{PASSWORD()} will result in the same encrypted value as is stored in the Unix password file. See @code{ENCRYPT()}. @findex ENCRYPT() @item ENCRYPT(str[,salt]) Encrypt @code{str} using the Unix @code{crypt()} system call. The @code{salt} argument should be a string with two characters. (As of MySQL Version 3.22.16, @code{salt} may be longer than two characters.): @example mysql> select ENCRYPT("hello"); -> 'VxuFAJXVARROc' @end example If @code{crypt()} is not available on your system, @code{ENCRYPT()} always returns @code{NULL}. @code{ENCRYPT()} ignores all but the first 8 characters of @code{str}, at least on some systems. This will be determined by the behavior of the underlying @code{crypt()} system call. @findex ENCODE() @item ENCODE(str,pass_str) Encrypt @code{str} using @code{pass_str} as the password. To decrypt the result, use @code{DECODE()}. The results is a binary string of the same length as @code{string}. If you want to save it in a column, use a @code{BLOB} column type. @findex DECODE() @item DECODE(crypt_str,pass_str) Descrypts the encrypted string @code{crypt_str} using @code{pass_str} as the password. @code{crypt_str} should be a string returned from @code{ENCODE()}. @findex MD5() @item MD5(string) Calculates a MD5 checksum for the string. Value is returned as a 32 long hex number that may, for example, be used as a hash key: @example mysql> select MD5("testing"); -> 'ae2b1fca515949e5d54fb22b8ed95575' @end example This is an "RSA Data Security, Inc. MD5 Message-Digest Algorithm". @findex LAST_INSERT_ID([expr]) @item LAST_INSERT_ID([expr]) Returns the last automatically generated value that was inserted into an @code{AUTO_INCREMENT} column. @xref{mysql_insert_id,, @code{mysql_insert_id()}}. @example mysql> select LAST_INSERT_ID(); -> 195 @end example The last ID that was generated is maintained in the server on a per-connection basis. It will not be changed by another client. It will not even be changed if you update another @code{AUTO_INCREMENT} column with a non-magic value (that is, a value that is not @code{NULL} and not @code{0}). If you insert many rows at the same time with an insert statement, @code{LAST_INSERT_ID()} returns the value for the first inserted row. The reason for this is so that you it makes it possible to easily reproduce the same @code{INSERT} statement against some other server. @cindex sequence emulation If @code{expr} is given as an argument to @code{LAST_INSERT_ID()}, then the value of the argument is returned by the function, is set as the next value to be returned by @code{LAST_INSERT_ID()} and used as the next auto_increment value. This can be used to simulate sequences: First create the table: @example mysql> create table sequence (id int not null); mysql> insert into sequence values (0); @end example Then the table can be used to generate sequence numbers like this: @example mysql> update sequence set id=LAST_INSERT_ID(id+1); @end example You can generate sequences without calling @code{LAST_INSERT_ID()}, but the utility of using the function this way is that the ID value is maintained in the server as the last automatically generated value. You can retrieve the new ID as you would read any normal @code{AUTO_INCREMENT} value in MySQL. For example, @code{LAST_INSERT_ID()} (without an argument) will return the new ID. The C API function @code{mysql_insert_id()} can also be used to get the value. Note that as @code{mysql_insert_id()} is only updated after @code{INSERT} and @code{UPDATE} statements, you can't use this function to retrieve the value for @code{LAST_INSERT_ID(expr)} after executing other SQL statements like @code{SELECT} or @code{SET}. @findex FORMAT() @item FORMAT(X,D) Formats the number @code{X} to a format like @code{'#,###,###.##'}, rounded to @code{D} decimals. If @code{D} is @code{0}, the result will have no decimal point or fractional part: @example mysql> select FORMAT(12332.123456, 4); -> '12,332.1235' mysql> select FORMAT(12332.1,4); -> '12,332.1000' mysql> select FORMAT(12332.2,0); -> '12,332' @end example @findex VERSION() @item VERSION() Returns a string indicating the MySQL server version: @example mysql> select VERSION(); -> '3.23.13-log' @end example Note that if your version ends with @code{-log} this means that logging is enabled. @findex CONNECTION_ID() @item CONNECTION_ID() Returns the connection id (@code{thread_id}) for the connection. Every connection has its own unique id: @example mysql> select CONNECTION_ID(); -> 1 @end example @cindex timeout @findex GET_LOCK() @item GET_LOCK(str,timeout) Tries to obtain a lock with a name given by the string @code{str}, with a timeout of @code{timeout} seconds. Returns @code{1} if the lock was obtained successfully, @code{0} if the attempt timed out, or @code{NULL} if an error occurred (such as running out of memory or the thread was killed with @code{mysqladmin kill}). A lock is released when you execute @code{RELEASE_LOCK()}, execute a new @code{GET_LOCK()}, or the thread terminates. This function can be used to implement application locks or to simulate record locks. It blocks requests by other clients for locks with the same name; clients that agree on a given lock string name can use the string to perform cooperative advisory locking: @example mysql> select GET_LOCK("lock1",10); -> 1 mysql> select GET_LOCK("lock2",10); -> 1 mysql> select RELEASE_LOCK("lock2"); -> 1 mysql> select RELEASE_LOCK("lock1"); -> NULL @end example Note that the second @code{RELEASE_LOCK()} call returns @code{NULL} because the lock @code{"lock1"} was automatically released by the second @code{GET_LOCK()} call. @findex RELEASE_LOCK() @item RELEASE_LOCK(str) Releases the lock named by the string @code{str} that was obtained with @code{GET_LOCK()}. Returns @code{1} if the lock was released, @code{0} if the lock wasn't locked by this thread (in which case the lock is not released), and @code{NULL} if the named lock didn't exist. The lock will not exist if it was never obtained by a call to @code{GET_LOCK()} or if it already has been released. The @code{DO} statement is convinient to use with @code{RELEASE_LOCK()}. @xref{DO}. @findex BENCHMARK() @item BENCHMARK(count,expr) The @code{BENCHMARK()} function executes the expression @code{expr} repeatedly @code{count} times. It may be used to time how fast MySQL processes the expression. The result value is always @code{0}. The intended use is in the @code{mysql} client, which reports query execution times: @example mysql> select BENCHMARK(1000000,encode("hello","goodbye")); +----------------------------------------------+ | BENCHMARK(1000000,encode("hello","goodbye")) | +----------------------------------------------+ | 0 | +----------------------------------------------+ 1 row in set (4.74 sec) @end example The time reported is elapsed time on the client end, not CPU time on the server end. It may be advisable to execute @code{BENCHMARK()} several times, and interpret the result with regard to how heavily loaded the server machine is. @findex INET_NTOA() @item INET_NTOA(expr) Returns the network address (4 or 8 byte) for the numeric expression: @example mysql> select INET_NTOA(3520061480); -> "209.207.224.40" @end example @findex INET_ATON() @item INET_ATON(expr) Returns an integer that represents the numeric value for a network address. Addresses may be 4 or 8 byte addresses: @example mysql> select INET_ATON("209.207.224.40"); -> 3520061480 @end example The generated number is always in network byte order; For example the above number is calculated as @code{209*256^3 + 207*256^2 + 224*256 +40}. @findex MASTER_POS_WAIT() @item MASTER_POS_WAIT(log_name, log_pos) Blocks until the slave reaches the specified position in the master log during replication. If master information is not initialized, returns NULL. If the slave is not running, will block and wait until it is started and goes to or past the specified position. If the slave is already past the specified position, returns immediately. The return value is the number of log events it had to wait to get to the specified position, or NULL in case of error. Useful for control of master-slave synchronization, but was originally written to facilitate replication testing. @end table @node Data Manipulation, Data Definition, Functions, Reference @section Data Manipulation: @code{SELECT}, @code{INSERT}, @code{UPDATE}, @code{DELETE} @menu * SELECT:: @code{SELECT} Syntax * INSERT:: @code{INSERT} Syntax * INSERT DELAYED:: @code{INSERT DELAYED} syntax * UPDATE:: @code{UPDATE} Syntax * DELETE:: @code{DELETE} Syntax * TRUNCATE:: @code{TRUNCATE} Syntax * REPLACE:: @code{REPLACE} Syntax * LOAD DATA:: @code{LOAD DATA INFILE} Syntax * DO:: @code{DO} Syntax @end menu @node SELECT, INSERT, Data Manipulation, Data Manipulation @subsection @code{SELECT} Syntax @findex SELECT @c help SELECT @example SELECT [STRAIGHT_JOIN] [SQL_SMALL_RESULT] [SQL_BIG_RESULT] [SQL_BUFFER_RESULT] [HIGH_PRIORITY] [DISTINCT | DISTINCTROW | ALL] select_expression,... [INTO @{OUTFILE | DUMPFILE@} 'file_name' export_options] [FROM table_references [WHERE where_definition] [GROUP BY @{unsigned_integer | col_name | formula@} [ASC | DESC], ...] [HAVING where_definition] [ORDER BY @{unsigned_integer | col_name | formula@} [ASC | DESC] ,...] [LIMIT [offset,] rows] [PROCEDURE procedure_name] [FOR UPDATE | LOCK IN SHARE MODE]] @end example @c help end @code{SELECT} is used to retrieve rows selected from one or more tables. @code{select_expression} indicates the columns you want to retrieve. @code{SELECT} may also be used to retrieve rows computed without reference to any table. For example: @example mysql> SELECT 1 + 1; -> 2 @end example All keywords used must be given in exactly the order shown above. For example, a @code{HAVING} clause must come after any @code{GROUP BY} clause and before any @code{ORDER BY} clause. @itemize @bullet @item @cindex aliases, on expressions @cindex expression aliases A @code{SELECT} expression may be given an alias using @code{AS}. The alias is used as the expression's column name and can be used with @code{ORDER BY} or @code{HAVING} clauses. For example: @example mysql> select concat(last_name,', ',first_name) AS full_name from mytable ORDER BY full_name; @end example @item The @code{FROM table_references} clause indicates the tables from which to retrieve rows. If you name more than one table, you are performing a join. For information on join syntax, see @ref{JOIN, , @code{JOIN}}. @item You can refer to a column as @code{col_name}, @code{tbl_name.col_name}, or @code{db_name.tbl_name.col_name}. You need not specify a @code{tbl_name} or @code{db_name.tbl_name} prefix for a column reference in a @code{SELECT} statement unless the reference would be ambiguous. See @ref{Legal names}, for examples of ambiguity that require the more explicit column reference forms. @item @cindex aliases, for tables @cindex table aliases A table reference may be aliased using @code{tbl_name [AS] alias_name}: @example mysql> select t1.name, t2.salary from employee AS t1, info AS t2 where t1.name = t2.name; mysql> select t1.name, t2.salary from employee t1, info t2 where t1.name = t2.name; @end example @item Columns selected for output may be referred to in @code{ORDER BY} and @code{GROUP BY} clauses using column names, column aliases, or column positions. Column positions begin with 1: @example mysql> select college, region, seed from tournament ORDER BY region, seed; mysql> select college, region AS r, seed AS s from tournament ORDER BY r, s; mysql> select college, region, seed from tournament ORDER BY 2, 3; @end example To sort in reverse order, add the @code{DESC} (descending) keyword to the name of the column in the @code{ORDER BY} clause that you are sorting by. The default is ascending order; this may be specified explicitly using the @code{ASC} keyword. @item You can in the @code{WHERE} clause use any of the functions that MySQL support. @xref{Functions}. @item The @code{HAVING} clause can refer to any column or alias named in the @code{select_expression}. It is applied last, just before items are sent to the client, with no optimization. Don't use @code{HAVING} for items that should be in the @code{WHERE} clause. For example, do not write this: @example mysql> select col_name from tbl_name HAVING col_name > 0; @end example Write this instead: @example mysql> select col_name from tbl_name WHERE col_name > 0; @end example In MySQL Version 3.22.5 or later, you can also write queries like this: @example mysql> select user,max(salary) from users group by user HAVING max(salary)>10; @end example In older MySQL versions, you can write this instead: @example mysql> select user,max(salary) AS sum from users group by user HAVING sum>10; @end example @item @code{SQL_SMALL_RESULT}, @code{SQL_BIG_RESULT}, @code{SQL_BUFFER_RESULT}, @code{STRAIGHT_JOIN}, and @code{HIGH_PRIORITY} are MySQL extensions to ANSI SQL92. @item @code{HIGH_PRIORITY} will give the @code{SELECT} higher priority than a statement that updates a table. You should only use this for queries that are very fast and must be done at once. A @code{SELECT HIGH_PRIORITY} query will run if the table is locked for read even if there is an update statement that is waiting for the table to be free. @item @code{SQL_BIG_RESULT} can be used with @code{GROUP BY} or @code{DISTINCT} to tell the optimizer that the result set will have many rows. In this case, MySQL will directly use disk-based temporary tables if needed. MySQL will also, in this case, prefer sorting to doing a temporary table with a key on the @code{GROUP BY} elements. @item @code{SQL_BUFFER_RESULT} will force the result to be put into a temporary table. This will help MySQL free the table locks early and will help in cases where it takes a long time to send the result set to the client. @item @code{SQL_SMALL_RESULT}, a MySQL-specific option, can be used with @code{GROUP BY} or @code{DISTINCT} to tell the optimizer that the result set will be small. In this case, MySQL will use fast temporary tables to store the resulting table instead of using sorting. In MySQL Version 3.23 this shouldn't normally be needed. @item @cindex @code{GROUP BY}, extensions to ANSI SQL If you use @code{GROUP BY}, the output rows will be sorted according to the @code{GROUP BY} as if you would have had an @code{ORDER BY} over all the fields in the @code{GROUP BY}. MySQL has extended the @code{GROUP BY} so that you can also specify @code{ASC} and @code{DESC} to @code{GROUP BY}: @example SELECT a,COUNT(b) FROM test_table GROUP BY a DESC @end example @item MySQL has extended the use of @code{GROUP BY} to allow you to select fields which are not mentioned in the @code{GROUP BY} clause. If you are not getting the results you expect from your query, please read the @code{GROUP BY} description. @xref{Group by functions}. @item @code{STRAIGHT_JOIN} forces the optimizer to join the tables in the order in which they are listed in the @code{FROM} clause. You can use this to speed up a query if the optimizer joins the tables in non-optimal order. @xref{EXPLAIN, , @code{EXPLAIN}}. @item The @code{LIMIT} clause can be used to constrain the number of rows returned by the @code{SELECT} statement. @code{LIMIT} takes one or two numeric arguments. If two arguments are given, the first specifies the offset of the first row to return, the second specifies the maximum number of rows to return. The offset of the initial row is 0 (not 1): @example mysql> select * from table LIMIT 5,10; # Retrieve rows 6-15 @end example If one argument is given, it indicates the maximum number of rows to return: @example mysql> select * from table LIMIT 5; # Retrieve first 5 rows @end example In other words, @code{LIMIT n} is equivalent to @code{LIMIT 0,n}. @item @tindex /etc/passwd The @code{SELECT ... INTO OUTFILE 'file_name'} form of @code{SELECT} writes the selected rows to a file. The file is created on the server host and cannot already exist (among other things, this prevents database tables and files such as @file{/etc/passwd} from being destroyed). You must have the @strong{file} privilege on the server host to use this form of @code{SELECT}. @code{SELECT ... INTO OUTFILE} is mainly intended to let you very quickly dump a table on the server machine. If you want to create the resulting file on some other host than the server host you can't use @code{SELECT ... INTO OUTFILE}. In this case you should instead use some client program like @code{mysqldump --tab} or @code{mysql -e "SELECT ..." > outfile} to generate the file. @code{SELECT ... INTO OUTFILE} is the complement of @code{LOAD DATA INFILE}; the syntax for the @code{export_options} part of the statement consists of the same @code{FIELDS} and @code{LINES} clauses that are used with the @code{LOAD DATA INFILE} statement. @xref{LOAD DATA, , @code{LOAD DATA}}. In the resulting text file, only the following characters are escaped by the @code{ESCAPED BY} character: @itemize @bullet @item The @code{ESCAPED BY} character @item The first character in @code{FIELDS TERMINATED BY} @item The first character in @code{LINES TERMINATED BY} @end itemize Additionally, @code{ASCII 0} is converted to @code{ESCAPED BY} followed by 0 (@code{ASCII 48}). The reason for the above is that you MUST escape any @code{FIELDS TERMINATED BY}, @code{ESCAPED BY}, or @code{LINES TERMINATED BY} characters to reliably be able to read the file back. @code{ASCII 0} is escaped to make it easier to view with some pagers. As the resulting file doesn't have to conform to the SQL syntax, nothing else need be escaped. Here follows an example of getting a file in the format used by many old programs. @example SELECT a,b,a+b INTO OUTFILE "/tmp/result.text" FIELDS TERMINATED BY ',' OPTIONALLY ENCLOSED BY '"' LINES TERMINATED BY "\n" FROM test_table; @end example @item @findex DUMPFILE If you use @code{INTO DUMPFILE} instead of @code{INTO OUTFILE}, MySQL will only write one row into the file, without any column or line terminations and without any escaping. This is useful if you want to store a blob in a file. @item Note that any file created by @code{INTO OUTFILE} and @code{INTO DUMPFILE} is going to be readable for all users! The reason is that the MySQL server can't create a file that is owned by anyone else than the user it's running as (you should never run @code{mysqld} as root), the file has to be word readable so that you can retrieve the rows. @item If you are using @code{FOR UPDATE} on a table handler with page/row locks, the examined rows will be write locked. @end itemize @menu * JOIN:: @code{JOIN} Syntax @end menu @node JOIN, , SELECT, SELECT @subsubsection @code{JOIN} Syntax @findex JOIN @findex INNER JOIN @findex CROSS JOIN @findex LEFT JOIN @findex LEFT OUTER JOIN @findex NATURAL LEFT JOIN @findex NATURAL LEFT OUTER JOIN @findex RIGHT JOIN @findex RIGHT OUTER JOIN @findex NATURAL RIGHT JOIN @findex NATURAL RIGHT OUTER JOIN @findex STRAIGHT_JOIN MySQL supports the following @code{JOIN} syntaxes for use in @code{SELECT} statements: @example table_reference, table_reference table_reference [CROSS] JOIN table_reference table_reference INNER JOIN table_reference join_condition table_reference STRAIGHT_JOIN table_reference table_reference LEFT [OUTER] JOIN table_reference join_condition table_reference LEFT [OUTER] JOIN table_reference table_reference NATURAL [LEFT [OUTER]] JOIN table_reference @{ oj table_reference LEFT OUTER JOIN table_reference ON conditional_expr @} table_reference RIGHT [OUTER] JOIN table_reference join_condition table_reference RIGHT [OUTER] JOIN table_reference table_reference NATURAL [RIGHT [OUTER]] JOIN table_reference @end example Where @code{table_reference} is defined as: @example table_name [[AS] alias] [USE INDEX (key_list)] [IGNORE INDEX (key_list)] @end example and @code{join_condition} is defined as: @example ON conditional_expr | USING (column_list) @end example You should never have any conditions in the @code{ON} part that are used to restrict which rows you have in the result set. If you want to restrict which rows should be in the result, you have to do this in the @code{WHERE} clause. Note that in versions before Version 3.23.17, the @code{INNER JOIN} didn't take a @code{join_condition}! @cindex ODBC compatibility @cindex compatibility, with ODBC The last @code{LEFT OUTER JOIN} syntax shown above exists only for compatibility with ODBC: @itemize @bullet @item A table reference may be aliased using @code{tbl_name AS alias_name} or @code{tbl_name alias_name}: @example mysql> select t1.name, t2.salary from employee AS t1, info AS t2 where t1.name = t2.name; @end example @item The @code{ON} conditional is any conditional of the form that may be used in a @code{WHERE} clause. @item If there is no matching record for the right table in the @code{ON} or @code{USING} part in a @code{LEFT JOIN}, a row with all columns set to @code{NULL} is used for the right table. You can use this fact to find records in a table that have no counterpart in another table: @example mysql> select table1.* from table1 LEFT JOIN table2 ON table1.id=table2.id where table2.id is NULL; @end example This example finds all rows in @code{table1} with an @code{id} value that is not present in @code{table2} (that is, all rows in @code{table1} with no corresponding row in @code{table2}). This assumes that @code{table2.id} is declared @code{NOT NULL}, of course. @xref{LEFT JOIN optimization}. @item The @code{USING} @code{(column_list)} clause names a list of columns that must exist in both tables. A @code{USING} clause such as: @example A LEFT JOIN B USING (C1,C2,C3,...) @end example is defined to be semantically identical to an @code{ON} expression like this: @example A.C1=B.C1 AND A.C2=B.C2 AND A.C3=B.C3,... @end example @item The @code{NATURAL [LEFT] JOIN} of two tables is defined to be semantically equivalent to an @code{INNER JOIN} or a @code{LEFT JOIN} with a @code{USING} clause that names all columns that exist in both tables. @item @code{RIGHT JOIN} works analogously as @code{LEFT JOIN}. To keep code portable across databases, it's recommended to use @code{LEFT JOIN} instead of @code{RIGHT JOIN}. @item @code{STRAIGHT_JOIN} is identical to @code{JOIN}, except that the left table is always read before the right table. This can be used for those (few) cases where the join optimizer puts the tables in the wrong order. @item As of MySQL Version 3.23.12, you can give hints about which index MySQL should use when retrieving information from a table. This is useful if @code{EXPLAIN} shows that MySQL is using the wrong index. By specifying @code{USE INDEX (key_list)}, you can tell MySQL to use only one of the specified indexes to find rows in the table. The alternative syntax @code{IGNORE INDEX (key_list)} can be used to tell MySQL to not use some particular index. @end itemize Some examples: @example mysql> select * from table1,table2 where table1.id=table2.id; mysql> select * from table1 LEFT JOIN table2 ON table1.id=table2.id; mysql> select * from table1 LEFT JOIN table2 USING (id); mysql> select * from table1 LEFT JOIN table2 ON table1.id=table2.id LEFT JOIN table3 ON table2.id=table3.id; mysql> select * from table1 USE INDEX (key1,key2) WHERE key1=1 and key2=2 AND key3=3; mysql> select * from table1 IGNORE INDEX (key3) WHERE key1=1 and key2=2 AND key3=3; @end example @xref{LEFT JOIN optimization, , @code{LEFT JOIN} optimization}. @node INSERT, INSERT DELAYED, SELECT, Data Manipulation @subsection @code{INSERT} Syntax @findex INSERT @example INSERT [LOW_PRIORITY | DELAYED] [IGNORE] [INTO] tbl_name [(col_name,...)] VALUES (expression,...),(...),... or INSERT [LOW_PRIORITY | DELAYED] [IGNORE] [INTO] tbl_name [(col_name,...)] SELECT ... or INSERT [LOW_PRIORITY | DELAYED] [IGNORE] [INTO] tbl_name SET col_name=expression, col_name=expression, ... @end example @code{INSERT} inserts new rows into an existing table. The @code{INSERT ... VALUES} form of the statement inserts rows based on explicitly specified values. The @code{INSERT ... SELECT} form inserts rows selected from another table or tables. The @code{INSERT ... VALUES} form with multiple value lists is supported in MySQL Version 3.22.5 or later. The @code{col_name=expression} syntax is supported in MySQL Version 3.22.10 or later. @code{tbl_name} is the table into which rows should be inserted. The column name list or the @code{SET} clause indicates which columns the statement specifies values for: @itemize @bullet @item If you specify no column list for @code{INSERT ... VALUES} or @code{INSERT ... SELECT}, values for all columns must be provided in the @code{VALUES()} list or by the @code{SELECT}. If you don't know the order of the columns in the table, use @code{DESCRIBE tbl_name} to find out. @item Any column not explicitly given a value is set to its default value. For example, if you specify a column list that doesn't name all the columns in the table, unnamed columns are set to their default values. Default value assignment is described in @ref{CREATE TABLE, , @code{CREATE TABLE}}. @item An @code{expression} may refer to any column that was set earlier in a value list. For example, you can say this: @example mysql> INSERT INTO tbl_name (col1,col2) VALUES(15,col1*2); @end example But not this: @example mysql> INSERT INTO tbl_name (col1,col2) VALUES(col2*2,15); @end example @item If you specify the keyword @code{LOW_PRIORITY}, execution of the @code{INSERT} is delayed until no other clients are reading from the table. In this case the client has to wait until the insert statement is completed, which may take a long time if the table is in heavy use. This is in contrast to @code{INSERT DELAYED}, which lets the client continue at once. @xref{INSERT DELAYED}. Note that @code{LOW_PRIORITY} should normally not be used with @code{MyISAM} tables as this disables concurrent inserts. @xref{MyISAM}. @item If you specify the keyword @code{IGNORE} in an @code{INSERT} with many value rows, any rows that duplicate an existing @code{PRIMARY} or @code{UNIQUE} key in the table are ignored and are not inserted. If you do not specify @code{IGNORE}, the insert is aborted if there is any row that duplicates an existing key value. You can determine with the C API function @code{mysql_info()} how many rows were inserted into the table. @item If MySQL was configured using the @code{DONT_USE_DEFAULT_FIELDS} option, @code{INSERT} statements generate an error unless you explicitly specify values for all columns that require a non-@code{NULL} value. @xref{configure options, , @code{configure} options}. @item You can find the value used for an @code{AUTO_INCREMENT} column with the @code{mysql_insert_id} function. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @end itemize @findex mysql_info() If you use @code{INSERT ... SELECT} or an @code{INSERT ... VALUES} statement with multiple value lists, you can use the C API function @code{mysql_info()} to get information about the query. The format of the information string is shown below: @example Records: 100 Duplicates: 0 Warnings: 0 @end example @code{Duplicates} indicates the number of rows that couldn't be inserted because they would duplicate some existing unique index value. @code{Warnings} indicates the number of attempts to insert column values that were problematic in some way. Warnings can occur under any of the following conditions: @itemize @bullet @item Inserting @code{NULL} into a column that has been declared @code{NOT NULL}. The column is set to its default value. @item Setting a numeric column to a value that lies outside the column's range. The value is clipped to the appropriate endpoint of the range. @item Setting a numeric column to a value such as @code{'10.34 a'}. The trailing garbage is stripped and the remaining numeric part is inserted. If the value doesn't make sense as a number at all, the column is set to @code{0}. @item Inserting a string into a @code{CHAR}, @code{VARCHAR}, @code{TEXT}, or @code{BLOB} column that exceeds the column's maximum length. The value is truncated to the column's maximum length. @item Inserting a value into a date or time column that is illegal for the column type. The column is set to the appropriate zero value for the type. @end itemize @findex REPLACE ... SELECT @findex INSERT ... SELECT @menu * INSERT SELECT:: @code{INSERT ... SELECT} Syntax @end menu @node INSERT SELECT, , INSERT, INSERT @subsubsection @code{INSERT ... SELECT} Syntax @example INSERT [LOW_PRIORITY] [IGNORE] [INTO] tbl_name [(column list)] SELECT ... @end example With @code{INSERT ... SELECT} statement you can quickly insert many rows into a table from one or many tables. @example INSERT INTO tblTemp2 (fldID) SELECT tblTemp1.fldOrder_ID FROM tblTemp1 WHERE tblTemp1.fldOrder_ID > 100; @end example The following conditions hold for an @code{INSERT ... SELECT} statement: @itemize @minus @item The target table of the @code{INSERT} statement cannot appear in the @code{FROM} clause of the @code{SELECT} part of the query because it's forbidden in ANSI SQL to @code{SELECT} from the same table into which you are inserting. (The problem is that the @code{SELECT} possibly would find records that were inserted earlier during the same run. When using sub-select clauses, the situation could easily be very confusing!) @item @code{AUTO_INCREMENT} columns work as usual. @item You can use the C API function @code{mysql_info()} to get information about the query. @xref{INSERT}. @item To ensure that the update log/binary log can be used to re-create the original tables, MySQL will not allow concurrent inserts during @code{INSERT .... SELECT}. @end itemize You can of course also use @code{REPLACE} instead of @code{INSERT} to overwrite old rows. @node INSERT DELAYED, UPDATE, INSERT, Data Manipulation @subsection @code{INSERT DELAYED} syntax @findex INSERT DELAYED @findex DELAYED @cindex INSERT DELAYED @example INSERT DELAYED ... @end example The @code{DELAYED} option for the @code{INSERT} statement is a MySQL-specific option that is very useful if you have clients that can't wait for the @code{INSERT} to complete. This is a common problem when you use MySQL for logging and you also periodically run @code{SELECT} and @code{UPDATE} statements that take a long time to complete. @code{DELAYED} was introduced in MySQL Version 3.22.15. It is a MySQL extension to ANSI SQL92. @code{INSERT DELAYED} only works with @code{ISAM} and @code{MyISAM} tables. Note that as @code{MyISAM} tables supports concurrent @code{SELECT} and @code{INSERT}, if there is no free blocks in the middle of the data file, you very seldom need to use @code{INSERT DELAYED} with @code{MyISAM}. @xref{MyISAM}. When you use @code{INSERT DELAYED}, the client will get an OK at once and the row will be inserted when the table is not in use by any other thread. Another major benefit of using @code{INSERT DELAYED} is that inserts from many clients are bundled together and written in one block. This is much faster than doing many separate inserts. Note that currently the queued rows are only stored in memory until they are inserted into the table. This means that if you kill @code{mysqld} the hard way (@code{kill -9}) or if @code{mysqld} dies unexpectedly, any queued rows that weren't written to disk are lost! The following describes in detail what happens when you use the @code{DELAYED} option to @code{INSERT} or @code{REPLACE}. In this description, the ``thread'' is the thread that received an @code{INSERT DELAYED} command and ``handler'' is the thread that handles all @code{INSERT DELAYED} statements for a particular table. @itemize @bullet @item When a thread executes a @code{DELAYED} statement for a table, a handler thread is created to process all @code{DELAYED} statements for the table, if no such handler already exists. @item The thread checks whether or not the handler has acquired a @code{DELAYED} lock already; if not, it tells the handler thread to do so. The @code{DELAYED} lock can be obtained even if other threads have a @code{READ} or @code{WRITE} lock on the table. However, the handler will wait for all @code{ALTER TABLE} locks or @code{FLUSH TABLES} to ensure that the table structure is up to date. @item The thread executes the @code{INSERT} statement, but instead of writing the row to the table, it puts a copy of the final row into a queue that is managed by the handler thread. Any syntax errors are noticed by the thread and reported to the client program. @item The client can't report the number of duplicates or the @code{AUTO_INCREMENT} value for the resulting row; it can't obtain them from the server, because the @code{INSERT} returns before the insert operation has been completed. If you use the C API, the @code{mysql_info()} function doesn't return anything meaningful, for the same reason. @item The update log is updated by the handler thread when the row is inserted into the table. In case of multiple-row inserts, the update log is updated when the first row is inserted. @item After every @code{delayed_insert_limit} rows are written, the handler checks whether or not any @code{SELECT} statements are still pending. If so, it allows these to execute before continuing. @cindex delayed_insert_limit @cindex timeout @item When the handler has no more rows in its queue, the table is unlocked. If no new @code{INSERT DELAYED} commands are received within @code{delayed_insert_timeout} seconds, the handler terminates. @item If more than @code{delayed_queue_size} rows are pending already in a specific handler queue, the thread requesting @code{INSERT DELAYED} waits until there is room in the queue. This is done to ensure that the @code{mysqld} server doesn't use all memory for the delayed memory queue. @item The handler thread will show up in the MySQL process list with @code{delayed_insert} in the @code{Command} column. It will be killed if you execute a @code{FLUSH TABLES} command or kill it with @code{KILL thread_id}. However, it will first store all queued rows into the table before exiting. During this time it will not accept any new @code{INSERT} commands from another thread. If you execute an @code{INSERT DELAYED} command after this, a new handler thread will be created. @item Note that the above means that @code{INSERT DELAYED} commands have higher priority than normal @code{INSERT} commands if there is an @code{INSERT DELAYED} handler already running! Other update commands will have to wait until the @code{INSERT DELAYED} queue is empty, someone kills the handler thread (with @code{KILL thread_id}), or someone executes @code{FLUSH TABLES}. @item The following status variables provide information about @code{INSERT DELAYED} commands: @multitable @columnfractions .35 .65 @item @strong{Variable} @tab @strong{Meaning} @item @code{Delayed_insert_threads} @tab Number of handler threads @item @code{Delayed_writes} @tab Number of rows written with @code{INSERT DELAYED} @item @code{Not_flushed_delayed_rows} @tab Number of rows waiting to be written @end multitable You can view these variables by issuing a @code{SHOW STATUS} statement or by executing a @code{mysqladmin extended-status} command. @end itemize Note that @code{INSERT DELAYED} is slower than a normal INSERT if the table is not in use. There is also the additional overhead for the server to handle a separate thread for each table on which you use @code{INSERT DELAYED}. This means that you should only use @code{INSERT DELAYED} when you are really sure you need it! @node UPDATE, DELETE, INSERT DELAYED, Data Manipulation @subsection @code{UPDATE} Syntax @findex UPDATE @example UPDATE [LOW_PRIORITY] [IGNORE] tbl_name SET col_name1=expr1, [col_name2=expr2, ...] [WHERE where_definition] [LIMIT #] @end example @code{UPDATE} updates columns in existing table rows with new values. The @code{SET} clause indicates which columns to modify and the values they should be given. The @code{WHERE} clause, if given, specifies which rows should be updated. Otherwise all rows are updated. If the @code{ORDER BY} clause is specified, the rows will be updated in the order that is specified. If you specify the keyword @code{LOW_PRIORITY}, execution of the @code{UPDATE} is delayed until no other clients are reading from the table. If you specify the keyword @code{IGNORE}, the update statement will not abort even if we get duplicate key errors during the update. Rows that would cause conflicts will not be updated. If you access a column from @code{tbl_name} in an expression, @code{UPDATE} uses the current value of the column. For example, the following statement sets the @code{age} column to one more than its current value: @example mysql> UPDATE persondata SET age=age+1; @end example @code{UPDATE} assignments are evaluated from left to right. For example, the following statement doubles the @code{age} column, then increments it: @example mysql> UPDATE persondata SET age=age*2, age=age+1; @end example If you set a column to the value it currently has, MySQL notices this and doesn't update it. @findex mysql_info() @code{UPDATE} returns the number of rows that were actually changed. In MySQL Version 3.22 or later, the C API function @code{mysql_info()} returns the number of rows that were matched and updated and the number of warnings that occurred during the @code{UPDATE}. In MySQL Version 3.23, you can use @code{LIMIT #} to ensure that only a given number of rows are changed. @node DELETE, TRUNCATE, UPDATE, Data Manipulation @subsection @code{DELETE} Syntax @findex DELETE @example DELETE [LOW_PRIORITY] FROM tbl_name [WHERE where_definition] [LIMIT rows] @end example @code{DELETE} deletes rows from @code{tbl_name} that satisfy the condition given by @code{where_definition}, and returns the number of records deleted. If you issue a @code{DELETE} with no @code{WHERE} clause, all rows are deleted. If you do this in @code{AUTOCOMMIT} mode, this works as @code{TRUNCATE}. @xref{TRUNCATE}. One problem with this is that @code{DELETE} will return zero as the number of affected records, but this will be fixed in 4.0. If you really want to know how many records are deleted when you are deleting all rows, and are willing to suffer a speed penalty, you can use a @code{DELETE} statement of this form: @example mysql> DELETE FROM tbl_name WHERE 1>0; @end example Note that this is MUCH slower than @code{DELETE FROM tbl_name} with no @code{WHERE} clause, because it deletes rows one at a time. If you specify the keyword @code{LOW_PRIORITY}, execution of the @code{DELETE} is delayed until no other clients are reading from the table. Deleted records are maintained in a linked list and subsequent @code{INSERT} operations reuse old record positions. To reclaim unused space and reduce file sizes, use the @code{OPTIMIZE TABLE} statement or the @code{myisamchk} utility to reorganize tables. @code{OPTIMIZE TABLE} is easier, but @code{myisamchk} is faster. See @ref{OPTIMIZE TABLE, , @code{OPTIMIZE TABLE}} and @ref{Optimization}. The MySQL-specific @code{LIMIT rows} option to @code{DELETE} tells the server the maximum number of rows to be deleted before control is returned to the client. This can be used to ensure that a specific @code{DELETE} command doesn't take too much time. You can simply repeat the @code{DELETE} command until the number of affected rows is less than the @code{LIMIT} value. @node TRUNCATE, REPLACE, DELETE, Data Manipulation @subsection @code{TRUNCATE} Syntax @findex TRUNCATE @example TRUNCATE TABLE table_name @end example Is in 3.23 and the same thing as @code{DELETE FROM table_name}. @xref{DELETE}. The differences are: @itemize @bullet @item Implemented as a drop and re-create of the table, which makes this much faster when deleting many rows. @item Not transaction-safe; @code{TRUNCATE TABLE} will automatically end the current transaction as if @code{COMMIT} would have been called. @item Doesn't return the number of deleted rows. @item As long as the table definition file @file{table_name.frm} is valid, the table can be re-created this way, even if the data or index files have become corrupted. @end itemize @code{TRUNCATE} is an Oracle SQL extension. @node REPLACE, LOAD DATA, TRUNCATE, Data Manipulation @subsection @code{REPLACE} Syntax @findex REPLACE @example REPLACE [LOW_PRIORITY | DELAYED] [INTO] tbl_name [(col_name,...)] VALUES (expression,...),(...),... or REPLACE [LOW_PRIORITY | DELAYED] [INTO] tbl_name [(col_name,...)] SELECT ... or REPLACE [LOW_PRIORITY | DELAYED] [INTO] tbl_name SET col_name=expression, col_name=expression,... @end example @code{REPLACE} works exactly like @code{INSERT}, except that if an old record in the table has the same value as a new record on a unique index, the old record is deleted before the new record is inserted. @xref{INSERT, , @code{INSERT}}. In other words, you can't access the values of the old row from a @code{REPLACE} statement. In some old MySQL version it looked like you could do this, but that was a bug that has been corrected. When one uses a @code{REPLACE} command, @code{mysql_affected_rows()} will return 2 if the new row replaced and old row. This is because in this case one row was inserted and then the duplicate was deleted. The above makes it easy to check if @code{REPLACE} added or replaced a row. @node LOAD DATA, DO, REPLACE, Data Manipulation @subsection @code{LOAD DATA INFILE} Syntax @findex LOAD DATA INFILE @example LOAD DATA [LOW_PRIORITY | CONCURRENT] [LOCAL] INFILE 'file_name.txt' [REPLACE | IGNORE] INTO TABLE tbl_name [FIELDS [TERMINATED BY '\t'] [[OPTIONALLY] ENCLOSED BY ''] [ESCAPED BY '\\' ] ] [LINES TERMINATED BY '\n'] [IGNORE number LINES] [(col_name,...)] @end example The @code{LOAD DATA INFILE} statement reads rows from a text file into a table at a very high speed. If the @code{LOCAL} keyword is specified, the file is read from the client host. If @code{LOCAL} is not specified, the file must be located on the server. (@code{LOCAL} is available in MySQL Version 3.22.6 or later.) For security reasons, when reading text files located on the server, the files must either reside in the database directory or be readable by all. Also, to use @code{LOAD DATA INFILE} on server files, you must have the @strong{file} privilege on the server host. @xref{Privileges provided}. If you specify the keyword @code{LOW_PRIORITY}, execution of the @code{LOAD DATA} statement is delayed until no other clients are reading from the table. If you specify the keyword @code{CONCURRENT} with a @code{MyISAM} table, then other threads can retrieve data from the table while @code{LOAD DATA} is executing. Using this option will of course affect the performance of @code{LOAD DATA} a bit even if no other thread is using the table at the same time. Using @code{LOCAL} will be a bit slower than letting the server access the files directly, because the contents of the file must travel from the client host to the server host. On the other hand, you do not need the @strong{file} privilege to load local files. @c old version If you are using MySQL before Version 3.23.24 you can't read from a FIFO with @code{LOAD DATA INFILE}. If you need to read from a FIFO (for example the output from gunzip), use @code{LOAD DATA LOCAL INFILE} instead. @cindex @code{mysqlimport} You can also load data files by using the @code{mysqlimport} utility; it operates by sending a @code{LOAD DATA INFILE} command to the server. The @code{--local} option causes @code{mysqlimport} to read data files from the client host. You can specify the @code{--compress} option to get better performance over slow networks if the client and server support the compressed protocol. When locating files on the server host, the server uses the following rules: @itemize @bullet @item If an absolute pathname is given, the server uses the pathname as is. @item If a relative pathname with one or more leading components is given, the server searches for the file relative to the server's data directory. @item If a filename with no leading components is given, the server looks for the file in the database directory of the current database. @end itemize Note that these rules mean a file given as @file{./myfile.txt} is read from the server's data directory, whereas a file given as @file{myfile.txt} is read from the database directory of the current database. For example, the following @code{LOAD DATA} statement reads the file @file{data.txt} from the database directory for @code{db1} because @code{db1} is the current database, even though the statement explicitly loads the file into a table in the @code{db2} database: @example mysql> USE db1; mysql> LOAD DATA INFILE "data.txt" INTO TABLE db2.my_table; @end example The @code{REPLACE} and @code{IGNORE} keywords control handling of input records that duplicate existing records on unique key values. If you specify @code{REPLACE}, new rows replace existing rows that have the same unique key value. If you specify @code{IGNORE}, input rows that duplicate an existing row on a unique key value are skipped. If you don't specify either option, an error occurs when a duplicate key value is found, and the rest of the text file is ignored. If you load data from a local file using the @code{LOCAL} keyword, the server has no way to stop transmission of the file in the middle of the operation, so the default bahavior is the same as if @code{IGNORE} is specified. If you use @code{LOAD DATA INFILE} on an empty @code{MyISAM} table, all non-unique indexes are created in a separate batch (like in @code{REPAIR}). This normally makes @code{LOAD DATA INFILE} much faster when you have many indexes. @code{LOAD DATA INFILE} is the complement of @code{SELECT ... INTO OUTFILE}. @xref{SELECT, , @code{SELECT}}. To write data from a database to a file, use @code{SELECT ... INTO OUTFILE}. To read the file back into the database, use @code{LOAD DATA INFILE}. The syntax of the @code{FIELDS} and @code{LINES} clauses is the same for both commands. Both clauses are optional, but @code{FIELDS} must precede @code{LINES} if both are specified. If you specify a @code{FIELDS} clause, each of its subclauses (@code{TERMINATED BY}, @code{[OPTIONALLY] ENCLOSED BY}, and @code{ESCAPED BY}) is also optional, except that you must specify at least one of them. If you don't specify a @code{FIELDS} clause, the defaults are the same as if you had written this: @example FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' @end example If you don't specify a @code{LINES} clause, the default is the same as if you had written this: @example LINES TERMINATED BY '\n' @end example In other words, the defaults cause @code{LOAD DATA INFILE} to act as follows when reading input: @itemize @bullet @item Look for line boundaries at newlines. @item Break lines into fields at tabs. @item Do not expect fields to be enclosed within any quoting characters. @item Interpret occurrences of tab, newline, or @samp{\} preceded by @samp{\} as literal characters that are part of field values. @end itemize Conversely, the defaults cause @code{SELECT ... INTO OUTFILE} to act as follows when writing output: @itemize @bullet @item Write tabs between fields. @item Do not enclose fields within any quoting characters. @item Use @samp{\} to escape instances of tab, newline or @samp{\} that occur within field values. @item Write newlines at the ends of lines. @end itemize Note that to write @code{FIELDS ESCAPED BY '\\'}, you must specify two backslashes for the value to be read as a single backslash. The @code{IGNORE number LINES} option can be used to ignore a header of column names at the start of the file: @example mysql> LOAD DATA INFILE "/tmp/file_name" into table test IGNORE 1 LINES; @end example When you use @code{SELECT ... INTO OUTFILE} in tandem with @code{LOAD DATA INFILE} to write data from a database into a file and then read the file back into the database later, the field and line handling options for both commands must match. Otherwise, @code{LOAD DATA INFILE} will not interpret the contents of the file properly. Suppose you use @code{SELECT ... INTO OUTFILE} to write a file with fields delimited by commas: @example mysql> SELECT * INTO OUTFILE 'data.txt' FIELDS TERMINATED BY ',' FROM ...; @end example To read the comma-delimited file back in, the correct statement would be: @example mysql> LOAD DATA INFILE 'data.txt' INTO TABLE table2 FIELDS TERMINATED BY ','; @end example If instead you tried to read in the file with the statement shown below, it wouldn't work because it instructs @code{LOAD DATA INFILE} to look for tabs between fields: @example mysql> LOAD DATA INFILE 'data.txt' INTO TABLE table2 FIELDS TERMINATED BY '\t'; @end example The likely result is that each input line would be interpreted as a single field. @code{LOAD DATA INFILE} can be used to read files obtained from external sources, too. For example, a file in dBASE format will have fields separated by commas and enclosed in double quotes. If lines in the file are terminated by newlines, the command shown below illustrates the field and line handling options you would use to load the file: @example mysql> LOAD DATA INFILE 'data.txt' INTO TABLE tbl_name FIELDS TERMINATED BY ',' ENCLOSED BY '"' LINES TERMINATED BY '\n'; @end example Any of the field or line handling options may specify an empty string (@code{''}). If not empty, the @code{FIELDS [OPTIONALLY] ENCLOSED BY} and @code{FIELDS ESCAPED BY} values must be a single character. The @code{FIELDS TERMINATED BY} and @code{LINES TERMINATED BY} values may be more than one character. For example, to write lines that are terminated by carriage return-linefeed pairs, or to read a file containing such lines, specify a @code{LINES TERMINATED BY '\r\n'} clause. For example, to read a file of jokes, that are separated with a line of @code{%%}, into a SQL table you can do: @example create table jokes (a int not null auto_increment primary key, joke text not null); load data infile "/tmp/jokes.txt" into table jokes fields terminated by "" lines terminated by "\n%%\n" (joke); @end example @code{FIELDS [OPTIONALLY] ENCLOSED BY} controls quoting of fields. For output (@code{SELECT ... INTO OUTFILE}), if you omit the word @code{OPTIONALLY}, all fields are enclosed by the @code{ENCLOSED BY} character. An example of such output (using a comma as the field delimiter) is shown below: @example "1","a string","100.20" "2","a string containing a , comma","102.20" "3","a string containing a \" quote","102.20" "4","a string containing a \", quote and comma","102.20" @end example If you specify @code{OPTIONALLY}, the @code{ENCLOSED BY} character is used only to enclose @code{CHAR} and @code{VARCHAR} fields: @example 1,"a string",100.20 2,"a string containing a , comma",102.20 3,"a string containing a \" quote",102.20 4,"a string containing a \", quote and comma",102.20 @end example Note that occurrences of the @code{ENCLOSED BY} character within a field value are escaped by prefixing them with the @code{ESCAPED BY} character. Also note that if you specify an empty @code{ESCAPED BY} value, it is possible to generate output that cannot be read properly by @code{LOAD DATA INFILE}. For example, the output just shown above would appear as shown below if the escape character is empty. Observe that the second field in the fourth line contains a comma following the quote, which (erroneously) appears to terminate the field: @example 1,"a string",100.20 2,"a string containing a , comma",102.20 3,"a string containing a " quote",102.20 4,"a string containing a ", quote and comma",102.20 @end example For input, the @code{ENCLOSED BY} character, if present, is stripped from the ends of field values. (This is true whether or not @code{OPTIONALLY} is specified; @code{OPTIONALLY} has no effect on input interpretation.) Occurrences of the @code{ENCLOSED BY} character preceded by the @code{ESCAPED BY} character are interpreted as part of the current field value. In addition, duplicated @code{ENCLOSED BY} characters occurring within fields are interpreted as single @code{ENCLOSED BY} characters if the field itself starts with that character. For example, if @code{ENCLOSED BY '"'} is specified, quotes are handled as shown below: @example "The ""BIG"" boss" -> The "BIG" boss The "BIG" boss -> The "BIG" boss The ""BIG"" boss -> The ""BIG"" boss @end example @code{FIELDS ESCAPED BY} controls how to write or read special characters. If the @code{FIELDS ESCAPED BY} character is not empty, it is used to prefix the following characters on output: @itemize @bullet @item The @code{FIELDS ESCAPED BY} character @item The @code{FIELDS [OPTIONALLY] ENCLOSED BY} character @item The first character of the @code{FIELDS TERMINATED BY} and @code{LINES TERMINATED BY} values @item ASCII @code{0} (what is actually written following the escape character is ASCII @code{'0'}, not a zero-valued byte) @end itemize If the @code{FIELDS ESCAPED BY} character is empty, no characters are escaped. It is probably not a good idea to specify an empty escape character, particularly if field values in your data contain any of the characters in the list just given. For input, if the @code{FIELDS ESCAPED BY} character is not empty, occurrences of that character are stripped and the following character is taken literally as part of a field value. The exceptions are an escaped @samp{0} or @samp{N} (for example, @code{\0} or @code{\N} if the escape character is @samp{\}). These sequences are interpreted as ASCII @code{0} (a zero-valued byte) and @code{NULL}. See below for the rules on @code{NULL} handling. For more information about @samp{\}-escape syntax, see @ref{Literals}. In certain cases, field and line handling options interact: @itemize @bullet @item If @code{LINES TERMINATED BY} is an empty string and @code{FIELDS TERMINATED BY} is non-empty, lines are also terminated with @code{FIELDS TERMINATED BY}. @item If the @code{FIELDS TERMINATED BY} and @code{FIELDS ENCLOSED BY} values are both empty (@code{''}), a fixed-row (non-delimited) format is used. With fixed-row format, no delimiters are used between fields. Instead, column values are written and read using the ``display'' widths of the columns. For example, if a column is declared as @code{INT(7)}, values for the column are written using 7-character fields. On input, values for the column are obtained by reading 7 characters. Fixed-row format also affects handling of @code{NULL} values; see below. Note that fixed-size format will not work if you are using a multi-byte character set. @end itemize Handling of @code{NULL} values varies, depending on the @code{FIELDS} and @code{LINES} options you use: @itemize @bullet @item For the default @code{FIELDS} and @code{LINES} values, @code{NULL} is written as @code{\N} for output and @code{\N} is read as @code{NULL} for input (assuming the @code{ESCAPED BY} character is @samp{\}). @item If @code{FIELDS ENCLOSED BY} is not empty, a field containing the literal word @code{NULL} as its value is read as a @code{NULL} value (this differs from the word @code{NULL} enclosed within @code{FIELDS ENCLOSED BY} characters, which is read as the string @code{'NULL'}). @item If @code{FIELDS ESCAPED BY} is empty, @code{NULL} is written as the word @code{NULL}. @item With fixed-row format (which happens when @code{FIELDS TERMINATED BY} and @code{FIELDS ENCLOSED BY} are both empty), @code{NULL} is written as an empty string. Note that this causes both @code{NULL} values and empty strings in the table to be indistinguishable when written to the file because they are both written as empty strings. If you need to be able to tell the two apart when reading the file back in, you should not use fixed-row format. @end itemize Some cases are not supported by @code{LOAD DATA INFILE}: @itemize @bullet @item Fixed-size rows (@code{FIELDS TERMINATED BY} and @code{FIELDS ENCLOSED BY} both empty) and @code{BLOB} or @code{TEXT} columns. @item If you specify one separator that is the same as or a prefix of another, @code{LOAD DATA INFILE} won't be able to interpret the input properly. For example, the following @code{FIELDS} clause would cause problems: @example FIELDS TERMINATED BY '"' ENCLOSED BY '"' @end example @item If @code{FIELDS ESCAPED BY} is empty, a field value that contains an occurrence of @code{FIELDS ENCLOSED BY} or @code{LINES TERMINATED BY} followed by the @code{FIELDS TERMINATED BY} value will cause @code{LOAD DATA INFILE} to stop reading a field or line too early. This happens because @code{LOAD DATA INFILE} cannot properly determine where the field or line value ends. @end itemize The following example loads all columns of the @code{persondata} table: @example mysql> LOAD DATA INFILE 'persondata.txt' INTO TABLE persondata; @end example No field list is specified, so @code{LOAD DATA INFILE} expects input rows to contain a field for each table column. The default @code{FIELDS} and @code{LINES} values are used. If you wish to load only some of a table's columns, specify a field list: @example mysql> LOAD DATA INFILE 'persondata.txt' INTO TABLE persondata (col1,col2,...); @end example You must also specify a field list if the order of the fields in the input file differs from the order of the columns in the table. Otherwise, MySQL cannot tell how to match up input fields with table columns. If a row has too few fields, the columns for which no input field is present are set to default values. Default value assignment is described in @ref{CREATE TABLE, , @code{CREATE TABLE}}. An empty field value is interpreted differently than if the field value is missing: @itemize @bullet @item For string types, the column is set to the empty string. @item For numeric types, the column is set to @code{0}. @item For date and time types, the column is set to the appropriate ``zero'' value for the type. @xref{Date and time types}. @end itemize Note that these are the same values that result if you assign an empty string explicitly to a string, numeric, or date or time type explicitly in an @code{INSERT} or @code{UPDATE} statement. @code{TIMESTAMP} columns are only set to the current date and time if there is a @code{NULL} value for the column, or (for the first @code{TIMESTAMP} column only) if the @code{TIMESTAMP} column is left out from the field list when a field list is specified. If an input row has too many fields, the extra fields are ignored and the number of warnings is incremented. @code{LOAD DATA INFILE} regards all input as strings, so you can't use numeric values for @code{ENUM} or @code{SET} columns the way you can with @code{INSERT} statements. All @code{ENUM} and @code{SET} values must be specified as strings! @findex mysql_info() If you are using the C API, you can get information about the query by calling the API function @code{mysql_info()} when the @code{LOAD DATA INFILE} query finishes. The format of the information string is shown below: @example Records: 1 Deleted: 0 Skipped: 0 Warnings: 0 @end example Warnings occur under the same circumstances as when values are inserted via the @code{INSERT} statement (@pxref{INSERT, , @code{INSERT}}), except that @code{LOAD DATA INFILE} also generates warnings when there are too few or too many fields in the input row. The warnings are not stored anywhere; the number of warnings can only be used as an indication if everything went well. If you get warnings and want to know exactly why you got them, one way to do this is to use @code{SELECT ... INTO OUTFILE} into another file and compare this to your original input file. If you need @code{LOAD DATA} to read from a pipe, you can use the following trick: @example mkfifo /mysql/db/x/x chmod 666 /mysql/db/x/x cat < /dev/tcp/10.1.1.12/4711 > /nt/mysql/db/x/x mysql -e "LOAD DATA INFILE 'x' INTO TABLE x" x @end example If you are using a version of MySQL older than 3.23.25 you can only do the above with @code{LOAD DATA LOCAL INFILE}. For more information about the efficiency of @code{INSERT} versus @code{LOAD DATA INFILE} and speeding up @code{LOAD DATA INFILE}, @xref{Insert speed}. @node DO, , LOAD DATA, Data Manipulation @subsection @code{DO} Syntax @findex DO @example DO expression, [expression, ...] @end example Execute the expression but don't return any results. This is a shorthand of @code{SELECT expression, expression}, but has the advantage that it's slightly faster when you don't care about the result. This is mainly useful with functions that has side effects, like @code{RELEASE_LOCK}. @node Data Definition, Basic User Commands, Data Manipulation, Reference @section Data Definition: @code{CREATE}, @code{DROP}, @code{ALTER} @menu * CREATE DATABASE:: @code{CREATE DATABASE} Syntax * DROP DATABASE:: @code{DROP DATABASE} Syntax * CREATE TABLE:: @code{CREATE TABLE} Syntax * ALTER TABLE:: @code{ALTER TABLE} Syntax * RENAME TABLE:: @code{RENAME TABLE} Syntax * DROP TABLE:: @code{DROP TABLE} Syntax * CREATE INDEX:: @code{CREATE INDEX} Syntax * DROP INDEX:: @code{DROP INDEX} Syntax @end menu @node CREATE DATABASE, DROP DATABASE, Data Definition, Data Definition @subsection @code{CREATE DATABASE} Syntax @findex CREATE DATABASE @example CREATE DATABASE [IF NOT EXISTS] db_name @end example @code{CREATE DATABASE} creates a database with the given name. Rules for allowable database names are given in @ref{Legal names}. An error occurs if the database already exists and you didn't specify @code{IF NOT EXISTS}. Databases in MySQL are implemented as directories containing files that correspond to tables in the database. Because there are no tables in a database when it is initially created, the @code{CREATE DATABASE} statement only creates a directory under the MySQL data directory. @cindex @code{mysqladmin} You can also create databases with @code{mysqladmin}. @xref{Client-Side Scripts}. @node DROP DATABASE, CREATE TABLE, CREATE DATABASE, Data Definition @subsection @code{DROP DATABASE} Syntax @findex DROP DATABASE @example DROP DATABASE [IF EXISTS] db_name @end example @code{DROP DATABASE} drops all tables in the database and deletes the database. If you do a @code{DROP DATABASE} on a symbolic linked database, both the link and the original database is deleted. @strong{Be VERY careful with this command!} @code{DROP DATABASE} returns the number of files that were removed from the database directory. Normally, this is three times the number of tables, because normally each table corresponds to a @file{.MYD} file, a @file{.MYI} file, and a @file{.frm} file. The @code{DROP DATABASE} command removes from the given database directory all files with the following extensions: @multitable @columnfractions .25 .25 .25 .25 @item .BAK @tab .DAT @tab .HSH @tab .ISD @item .ISM @tab .ISM @tab .MRG @tab .MYD @item .MYI @tab .db @tab .frm @end multitable All subdirectories that consists of 2 digits (@code{RAID} directories) are also removed. In MySQL Version 3.22 or later, you can use the keywords @code{IF EXISTS} to prevent an error from occurring if the database doesn't exist. @cindex @code{mysqladmin} You can also drop databases with @code{mysqladmin}. @xref{Client-Side Scripts}. @menu * CREATE TABLE:: @code{CREATE TABLE} Syntax * ALTER TABLE:: @code{ALTER TABLE} Syntax * RENAME TABLE:: @code{RENAME TABLE} Syntax * DROP TABLE:: @code{DROP TABLE} Syntax * CREATE INDEX:: @code{CREATE INDEX} Syntax * DROP INDEX:: @code{DROP INDEX} Syntax @end menu @node CREATE TABLE, ALTER TABLE, DROP DATABASE, Data Definition @subsection @code{CREATE TABLE} Syntax @findex CREATE TABLE @menu * Silent column changes:: Silent column changes @end menu @example CREATE [TEMPORARY] TABLE [IF NOT EXISTS] tbl_name [(create_definition,...)] [table_options] [select_statement] create_definition: col_name type [NOT NULL | NULL] [DEFAULT default_value] [AUTO_INCREMENT] [PRIMARY KEY] [reference_definition] or PRIMARY KEY (index_col_name,...) or KEY [index_name] (index_col_name,...) or INDEX [index_name] (index_col_name,...) or UNIQUE [INDEX] [index_name] (index_col_name,...) or FULLTEXT [INDEX] [index_name] (index_col_name,...) or [CONSTRAINT symbol] FOREIGN KEY index_name (index_col_name,...) [reference_definition] or CHECK (expr) type: TINYINT[(length)] [UNSIGNED] [ZEROFILL] or SMALLINT[(length)] [UNSIGNED] [ZEROFILL] or MEDIUMINT[(length)] [UNSIGNED] [ZEROFILL] or INT[(length)] [UNSIGNED] [ZEROFILL] or INTEGER[(length)] [UNSIGNED] [ZEROFILL] or BIGINT[(length)] [UNSIGNED] [ZEROFILL] or REAL[(length,decimals)] [UNSIGNED] [ZEROFILL] or DOUBLE[(length,decimals)] [UNSIGNED] [ZEROFILL] or FLOAT[(length,decimals)] [UNSIGNED] [ZEROFILL] or DECIMAL(length,decimals) [UNSIGNED] [ZEROFILL] or NUMERIC(length,decimals) [UNSIGNED] [ZEROFILL] or CHAR(length) [BINARY] or VARCHAR(length) [BINARY] or DATE or TIME or TIMESTAMP or DATETIME or TINYBLOB or BLOB or MEDIUMBLOB or LONGBLOB or TINYTEXT or TEXT or MEDIUMTEXT or LONGTEXT or ENUM(value1,value2,value3,...) or SET(value1,value2,value3,...) index_col_name: col_name [(length)] reference_definition: REFERENCES tbl_name [(index_col_name,...)] [MATCH FULL | MATCH PARTIAL] [ON DELETE reference_option] [ON UPDATE reference_option] reference_option: RESTRICT | CASCADE | SET NULL | NO ACTION | SET DEFAULT table_options: TYPE = @{BDB | HEAP | ISAM | InnoDB | MERGE | MYISAM @} or AUTO_INCREMENT = # or AVG_ROW_LENGTH = # or CHECKSUM = @{0 | 1@} or COMMENT = "string" or MAX_ROWS = # or MIN_ROWS = # or PACK_KEYS = @{0 | 1@} or PASSWORD = "string" or DELAY_KEY_WRITE = @{0 | 1@} or ROW_FORMAT= @{ default | dynamic | fixed | compressed @} or RAID_TYPE= @{1 | STRIPED | RAID0 @} RAID_CHUNKS=# RAID_CHUNKSIZE=# or UNION = (table_name,[table_name...]) or DATA DIRECTORY="directory" or INDEX DIRECTORY="directory" select_statement: [IGNORE | REPLACE] SELECT ... (Some legal select statement) @end example @code{CREATE TABLE} creates a table with the given name in the current database. Rules for allowable table names are given in @ref{Legal names}. An error occurs if there is no current database or if the table already exists. In MySQL Version 3.22 or later, the table name can be specified as @code{db_name.tbl_name}. This works whether or not there is a current database. In MySQL Version 3.23, you can use the @code{TEMPORARY} keyword when you create a table. A temporary table will automatically be deleted if a connection dies and the name is per connection. This means that two different connections can both use the same temporary table name without conflicting with each other or with an existing table of the same name. (The existing table is hidden until the temporary table is deleted). In MySQL Version 3.23 or later, you can use the keywords @code{IF NOT EXISTS} so that an error does not occur if the table already exists. Note that there is no verification that the table structures are identical. Each table @code{tbl_name} is represented by some files in the database directory. In the case of MyISAM-type tables you will get: @multitable @columnfractions .2 .8 @item @strong{File} @tab @strong{Purpose} @item @code{tbl_name.frm} @tab Table definition (form) file @item @code{tbl_name.MYD} @tab Data file @item @code{tbl_name.MYI} @tab Index file @end multitable For more information on the properties of the various column types, see @ref{Column types}: @itemize @bullet @item If neither @code{NULL} nor @code{NOT NULL} is specified, the column is treated as though @code{NULL} had been specified. @item An integer column may have the additional attribute @code{AUTO_INCREMENT}. When you insert a value of @code{NULL} (recommended) or @code{0} into an @code{AUTO_INCREMENT} column, the column is set to @code{value+1}, where @code{value} is the largest value for the column currently in the table. @code{AUTO_INCREMENT} sequences begin with @code{1}. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. If you delete the row containing the maximum value for an @code{AUTO_INCREMENT} column, the value will be reused with an @code{ISAM}, or @code{BDB} table but not with a @code{MyISAM} or @code{InnoDB} table. If you delete all rows in the table with @code{DELETE FROM table_name} (without a @code{WHERE}) in @code{AUTOCOMMIT} mode, the sequence starts over for all table types. @strong{NOTE:} There can be only one @code{AUTO_INCREMENT} column per table, and it must be indexed. MySQL Version 3.23 will also only work properly if the auto_increment column only has positive values. Inserting a negative number is regarded as inserting a very large positive number. This is done to avoid precision problems when numbers 'wrap' over from positive to negative and also to ensure that one doesn't accidentally get an auto_increment column that contains 0. @cindex ODBC compatibility @cindex compatibility, with ODBC To make MySQL compatible with some ODBC applications, you can find the last inserted row with the following query: @example SELECT * FROM tbl_name WHERE auto_col IS NULL @end example @item @code{NULL} values are handled differently for @code{TIMESTAMP} columns than for other column types. You cannot store a literal @code{NULL} in a @code{TIMESTAMP} column; setting the column to @code{NULL} sets it to the current date and time. Because @code{TIMESTAMP} columns behave this way, the @code{NULL} and @code{NOT NULL} attributes do not apply in the normal way and are ignored if you specify them. On the other hand, to make it easier for MySQL clients to use @code{TIMESTAMP} columns, the server reports that such columns may be assigned @code{NULL} values (which is true), even though @code{TIMESTAMP} never actually will contain a @code{NULL} value. You can see this when you use @code{DESCRIBE tbl_name} to get a description of your table. Note that setting a @code{TIMESTAMP} column to @code{0} is not the same as setting it to @code{NULL}, because @code{0} is a valid @code{TIMESTAMP} value. @item If no @code{DEFAULT} value is specified for a column, MySQL automatically assigns one. If the column may take @code{NULL} as a value, the default value is @code{NULL}. If the column is declared as @code{NOT NULL}, the default value depends on the column type: @itemize @minus @item For numeric types other than those declared with the @code{AUTO_INCREMENT} attribute, the default is @code{0}. For an @code{AUTO_INCREMENT} column, the default value is the next value in the sequence. @item For date and time types other than @code{TIMESTAMP}, the default is the appropriate zero value for the type. For the first @code{TIMESTAMP} column in a table, the default value is the current date and time. @xref{Date and time types}. @item For string types other than @code{ENUM}, the default value is the empty string. For @code{ENUM}, the default is the first enumeration value. @end itemize Default values must be constants. This means, for example, that you cannot set the default for a date column to be the value of a function such as @code{NOW()} or @code{CURRENT_DATE}. @item @code{KEY} is a synonym for @code{INDEX}. @item In MySQL, a @code{UNIQUE} key can have only distinct values. An error occurs if you try to add a new row with a key that matches an existing row. @item @tindex PRIMARY KEY A @code{PRIMARY KEY} is a unique @code{KEY} with the extra constraint that all key columns must be defined as @code{NOT NULL}. In MySQL the key is named @code{PRIMARY}. A table can have only one @code{PRIMARY KEY}. If you don't have a @code{PRIMARY KEY} and some applications ask for the @code{PRIMARY KEY} in your tables, MySQL will return the first @code{UNIQUE} key, which doesn't have any @code{NULL} columns, as the @code{PRIMARY KEY}. @item A @code{PRIMARY KEY} can be a multiple-column index. However, you cannot create a multiple-column index using the @code{PRIMARY KEY} key attibute in a column specification. Doing so will mark only that single column as primary. You must use the @code{PRIMARY KEY(index_col_name, ...)} syntax. @item If the @code{PRIMARY} or @code{UNIQUE} key consists of only one column and this is of type integer, you can also refer to it as @code{_rowid} (new in Version 3.23.11). @item If you don't assign a name to an index, the index will be assigned the same name as the first @code{index_col_name}, with an optional suffix (@code{_2}, @code{_3}, @code{...}) to make it unique. You can see index names for a table using @code{SHOW INDEX FROM tbl_name}. @xref{SHOW, , @code{SHOW}}. @item @cindex @code{NULL} values, and indexes @cindex indexes, and @code{NULL} values Only the @code{MyISAM} table type supports indexes on columns that can have @code{NULL} values. In other cases you must declare such columns @code{NOT NULL} or an error results. @item With @code{col_name(length)} syntax, you can specify an index that uses only a part of a @code{CHAR} or @code{VARCHAR} column. This can make the index file much smaller. @xref{Indexes}. @item @cindex @code{BLOB} columns, indexing @cindex indexes, and @code{BLOB} columns @cindex @code{TEXT} columns, indexing @cindex indexes, and @code{TEXT} columns Only the @code{MyISAM} table type supports indexing on @code{BLOB} and @code{TEXT} columns. When putting an index on a @code{BLOB} or @code{TEXT} column you MUST always specify the length of the index: @example CREATE TABLE test (blob_col BLOB, index(blob_col(10))); @end example @item When you use @code{ORDER BY} or @code{GROUP BY} with a @code{TEXT} or @code{BLOB} column, only the first @code{max_sort_length} bytes are used. @xref{BLOB, , @code{BLOB}}. @item In MySQL Version 3.23.23 or later, you can also create special @strong{FULLTEXT} indexes. They are used for full-text search. Only the @code{MyISAM} table type supports @code{FULLTEXT} indexes. They can be created only from @code{VARCHAR} and @code{TEXT} columns. Indexing always happens over the entire column, partial indexing is not supported. See @ref{Fulltext Search} for details of operation. @item The @code{FOREIGN KEY}, @code{CHECK}, and @code{REFERENCES} clauses don't actually do anything. The syntax for them is provided only for compatibility, to make it easier to port code from other SQL servers and to run applications that create tables with references. @xref{Missing functions}. @item Each @code{NULL} column takes one bit extra, rounded up to the nearest byte. @item The maximum record length in bytes can be calculated as follows: @example row length = 1 + (sum of column lengths) + (number of NULL columns + 7)/8 + (number of variable-length columns) @end example @item The @code{table_options} and @code{SELECT} options are only implemented in MySQL Version 3.23 and above. The different table types are: @multitable @columnfractions .20 .80 @item BDB or Berkeley_db @tab Transaction-safe tables with page locking. @xref{BDB}. @item HEAP @tab The data for this table is only stored in memory. @xref{HEAP}. @item ISAM @tab The original table handler. @xref{ISAM}. @item InnoDB @tab Transaction-safe tables with row locking. @xref{InnoDB}. @item MERGE @tab A collection of MyISAM tables used as one table. @xref{MERGE}. @item MyISAM @tab The new binary portable table handler that is replacing ISAM. @xref{MyISAM}. @end multitable @xref{Table types}. If a table type is specified, and that particular type is not available, MySQL will choose the closest table type to the one that you have specified. For example, if @code{TYPE=BDB} is specified, and that distribution of MySQL does not support @code{BDB} tables, the table will be created as @code{MyISAM} instead. The other table options are used to optimize the behavior of the table. In most cases, you don't have to specify any of them. The options work for all table types, if not otherwise indicated: @multitable @columnfractions .20 .80 @item @code{AUTO_INCREMENT} @tab The next auto_increment value you want to set for your table (MyISAM). @item @code{AVG_ROW_LENGTH} @tab An approximation of the average row length for your table. You only need to set this for large tables with variable size records. @item @code{CHECKSUM} @tab Set this to 1 if you want MySQL to maintain a checksum for all rows (makes the table a little slower to update but makes it easier to find corrupted tables) (MyISAM). @item @code{COMMENT} @tab A 60-character comment for your table. @item @code{MAX_ROWS} @tab Max number of rows you plan to store in the table. @item @code{MIN_ROWS} @tab Minimum number of rows you plan to store in the table. @item @code{PACK_KEYS} @tab Set this to 1 if you want to have a smaller index. This usually makes updates slower and reads faster (MyISAM, ISAM). @item @code{PASSWORD} @tab Encrypt the @code{.frm} file with a password. This option doesn't do anything in the standard MySQL version. @item @code{DELAY_KEY_WRITE} @tab Set this to 1 if want to delay key table updates until the table is closed (MyISAM). @item @code{ROW_FORMAT} @tab Defines how the rows should be stored. Currently you can only use the DYNAMIC and STATIC options for MyISAM tables. @end multitable When you use a @code{MyISAM} table, MySQL uses the product of @code{max_rows * avg_row_length} to decide how big the resulting table will be. If you don't specify any of the above options, the maximum size for a table will be 4G (or 2G if your operating systems only supports 2G tables). The reason for this is just to keep down the pointer sizes to make the index smaller and faster if you don't really need big files. If you don't use @code{PACK_KEYS}, the default is to only pack strings, not numbers. If you use @code{PACK_KEYS=1}, numbers will be packed as well. When packing binary number keys, MySQL will use prefix compression. This means that you will only get a big benefit of this if you have many numbers that are the same. Prefix compression means that every key needs one extra byte to indicate how many bytes of the previous key are the same for the next key (note that the pointer to the row is stored in high-byte-first-order directly after the key, to improve compression.) This means that if you have many equal keys on two rows in a row, all following 'same' keys will usually only take 2 bytes (including the pointer to the row). Compare this to the ordinary case where the following keys will take storage_size_for_key + pointer_size (usually 4). On the other hand, if all keys are totally different, you will lose 1 byte per key, if the key isn't a key that can have @code{NULL} values (In this case the packed key length will be stored in the same byte that is used to mark if a key is @code{NULL}.) @item If you specify a @code{SELECT} after the @code{CREATE} statement, MySQL will create new fields for all elements in the @code{SELECT}. For example: @example mysql> CREATE TABLE test (a int not null auto_increment, primary key (a), key(b)) TYPE=MyISAM SELECT b,c from test2; @end example This will create a @code{MyISAM} table with three columns, a, b, and c. Notice that the columns from the @code{SELECT} statement are appended to the right side of the table, not overlapped onto it. Take the following example: @example mysql> select * from foo; +---+ | n | +---+ | 1 | +---+ mysql> create table bar (m int) select n from foo; Query OK, 1 row affected (0.02 sec) Records: 1 Duplicates: 0 Warnings: 0 mysql> select * from bar; +------+---+ | m | n | +------+---+ | NULL | 1 | +------+---+ 1 row in set (0.00 sec) @end example For each row in table @code{foo}, a row is inserted in @code{bar} with the values from @code{foo} and default values for the new columns. @code{CREATE TABLE ... SELECT} will not automatically create any indexes for you. This is done intentionally to make the command as flexible as possible. If you want to have indexes in the created table, you should specify these before the @code{SELECT} statement: @example mysql> create table bar (unique (n)) select n from foo; @end example If any errors occur while copying the data to the table, it will automatically be deleted. To ensure that the update log/binary log can be used to re-create the original tables, MySQL will not allow concurrent inserts during @code{CREATE TABLE .... SELECT}. @item The @code{RAID_TYPE} option will help you to break the 2G/4G limit for the MyISAM data file (not the index file) on operating systems that don't support big files. You can get also more speed from the I/O bottleneck by putting @code{RAID} directories on different physical disks. @code{RAID_TYPE} will work on any OS, as long as you have configured MySQL with @code{--with-raid}. For now the only allowed @code{RAID_TYPE} is @code{STRIPED} (@code{1} and @code{RAID0} are aliases for this). If you specify @code{RAID_TYPE=STRIPED} for a @code{MyISAM} table, @code{MyISAM} will create @code{RAID_CHUNKS} subdirectories named 00, 01, 02 in the database directory. In each of these directories @code{MyISAM} will create a @code{table_name.MYD}. When writing data to the data file, the @code{RAID} handler will map the first @code{RAID_CHUNKSIZE} *1024 bytes to the first file, the next @code{RAID_CHUNKSIZE} *1024 bytes to the next file and so on. @item @code{UNION} is used when you want to use a collection of identical tables as one. This only works with MERGE tables. @xref{MERGE}. For the moment you need to have @code{SELECT}, @code{UPDATE}, and @code{DELETE} privileges on the tables you map to a @code{MERGE} table. All mapped tables must be in the same database as the @code{MERGE} table. @item In the created table the @code{PRIMARY} key will be placed first, followed by all @code{UNIQUE} keys and then the normal keys. This helps the MySQL optimizer to prioritize which key to use and also more quickly detect duplicated @code{UNIQUE} keys. @item By using @code{DATA DIRECTORY="directory"} or @code{INDEX DIRECTORY="directory"} you can specify where the table handler should put it's table and index files. This only works for @code{MyISAM} tables in @code{MySQL} 4.0, when you are not using the @code{--skip-symlink} option. @xref{Symbolic links to tables}. @end itemize @node Silent column changes, , CREATE TABLE, CREATE TABLE @subsubsection Silent Column Specification Changes @cindex silent column changes In some cases, MySQL silently changes a column specification from that given in a @code{CREATE TABLE} statement. (This may also occur with @code{ALTER TABLE}.): @itemize @bullet @item @code{VARCHAR} columns with a length less than four are changed to @code{CHAR}. @item If any column in a table has a variable length, the entire row is variable-length as a result. Therefore, if a table contains any variable-length columns (@code{VARCHAR}, @code{TEXT}, or @code{BLOB}), all @code{CHAR} columns longer than three characters are changed to @code{VARCHAR} columns. This doesn't affect how you use the columns in any way; in MySQL, @code{VARCHAR} is just a different way to store characters. MySQL performs this conversion because it saves space and makes table operations faster. @xref{Table types}. @item @code{TIMESTAMP} display sizes must be even and in the range from 2 to 14. If you specify a display size of 0 or greater than 14, the size is coerced to 14. Odd-valued sizes in the range from 1 to 13 are coerced to the next higher even number. @item You cannot store a literal @code{NULL} in a @code{TIMESTAMP} column; setting it to @code{NULL} sets it to the current date and time. Because @code{TIMESTAMP} columns behave this way, the @code{NULL} and @code{NOT NULL} attributes do not apply in the normal way and are ignored if you specify them. @code{DESCRIBE tbl_name} always reports that a @code{TIMESTAMP} column may be assigned @code{NULL} values. @item MySQL maps certain column types used by other SQL database vendors to MySQL types. @xref{Other-vendor column types}. @end itemize If you want to see whether or not MySQL used a column type other than the one you specified, issue a @code{DESCRIBE tbl_name} statement after creating or altering your table. @cindex @code{myisampack} Certain other column type changes may occur if you compress a table using @code{myisampack}. @xref{Compressed format}. @node ALTER TABLE, RENAME TABLE, CREATE TABLE, Data Definition @subsection @code{ALTER TABLE} Syntax @findex ALTER TABLE @example ALTER [IGNORE] TABLE tbl_name alter_spec [, alter_spec ...] alter_specification: ADD [COLUMN] create_definition [FIRST | AFTER column_name ] or ADD [COLUMN] (create_definition, create_definition,...) or ADD INDEX [index_name] (index_col_name,...) or ADD PRIMARY KEY (index_col_name,...) or ADD UNIQUE [index_name] (index_col_name,...) or ADD FULLTEXT [index_name] (index_col_name,...) or ADD [CONSTRAINT symbol] FOREIGN KEY index_name (index_col_name,...) [reference_definition] or ALTER [COLUMN] col_name @{SET DEFAULT literal | DROP DEFAULT@} or CHANGE [COLUMN] old_col_name create_definition or MODIFY [COLUMN] create_definition or DROP [COLUMN] col_name or DROP PRIMARY KEY or DROP INDEX index_name or RENAME [TO] new_tbl_name or ORDER BY col or table_options @end example @code{ALTER TABLE} allows you to change the structure of an existing table. For example, you can add or delete columns, create or destroy indexes, change the type of existing columns, or rename columns or the table itself. You can also change the comment for the table and type of the table. @xref{CREATE TABLE, , @code{CREATE TABLE}}. If you use @code{ALTER TABLE} to change a column specification but @code{DESCRIBE tbl_name} indicates that your column was not changed, it is possible that MySQL ignored your modification for one of the reasons described in @ref{Silent column changes}. For example, if you try to change a @code{VARCHAR} column to @code{CHAR}, MySQL will still use @code{VARCHAR} if the table contains other variable-length columns. @code{ALTER TABLE} works by making a temporary copy of the original table. The alteration is performed on the copy, then the original table is deleted and the new one is renamed. This is done in such a way that all updates are automatically redirected to the new table without any failed updates. While @code{ALTER TABLE} is executing, the original table is readable by other clients. Updates and writes to the table are stalled until the new table is ready. Note that if you use any other option to @code{ALTER TABLE} than @code{RENAME}, MySQL will always create a temporary table, even if the data wouldn't strictly need to be copied (like when you change the name of a column). We plan to fix this in the future, but as one doesn't normally do @code{ALTER TABLE} that often this isn't that high on our TODO. @itemize @bullet @item To use @code{ALTER TABLE}, you need @strong{ALTER}, @strong{INSERT}, and @strong{CREATE} privileges on the table. @item @code{IGNORE} is a MySQL extension to ANSI SQL92. It controls how @code{ALTER TABLE} works if there are duplicates on unique keys in the new table. If @code{IGNORE} isn't specified, the copy is aborted and rolled back. If @code{IGNORE} is specified, then for rows with duplicates on a unique key, only the first row is used; the others are deleted. @item You can issue multiple @code{ADD}, @code{ALTER}, @code{DROP}, and @code{CHANGE} clauses in a single @code{ALTER TABLE} statement. This is a MySQL extension to ANSI SQL92, which allows only one of each clause per @code{ALTER TABLE} statement. @item @code{CHANGE col_name}, @code{DROP col_name}, and @code{DROP INDEX} are MySQL extensions to ANSI SQL92. @item @code{MODIFY} is an Oracle extension to @code{ALTER TABLE}. @item The optional word @code{COLUMN} is a pure noise word and can be omitted. @item If you use @code{ALTER TABLE tbl_name RENAME TO new_name} without any other options, MySQL simply renames the files that correspond to the table @code{tbl_name}. There is no need to create the temporary table. @xref{RENAME TABLE,, @code{RENAME TABLE}}. @item @code{create_definition} clauses use the same syntax for @code{ADD} and @code{CHANGE} as for @code{CREATE TABLE}. Note that this syntax includes the column name, not just the column type. @xref{CREATE TABLE, , @code{CREATE TABLE}}. @item You can rename a column using a @code{CHANGE old_col_name create_definition} clause. To do so, specify the old and new column names and the type that the column currently has. For example, to rename an @code{INTEGER} column from @code{a} to @code{b}, you can do this: @example mysql> ALTER TABLE t1 CHANGE a b INTEGER; @end example If you want to change a column's type but not the name, @code{CHANGE} syntax still requires two column names even if they are the same. For example: @example mysql> ALTER TABLE t1 CHANGE b b BIGINT NOT NULL; @end example However, as of MySQL Version 3.22.16a, you can also use @code{MODIFY} to change a column's type without renaming it: @example mysql> ALTER TABLE t1 MODIFY b BIGINT NOT NULL; @end example @item If you use @code{CHANGE} or @code{MODIFY} to shorten a column for which an index exists on part of the column (for instance, if you have an index on the first 10 characters of a @code{VARCHAR} column), you cannot make the column shorter than the number of characters that are indexed. @item When you change a column type using @code{CHANGE} or @code{MODIFY}, MySQL tries to convert data to the new type as well as possible. @item In MySQL Version 3.22 or later, you can use @code{FIRST} or @code{ADD ... AFTER col_name} to add a column at a specific position within a table row. The default is to add the column last. @findex ALTER COLUMN @item @code{ALTER COLUMN} specifies a new default value for a column or removes the old default value. If the old default is removed and the column can be @code{NULL}, the new default is @code{NULL}. If the column cannot be @code{NULL}, MySQL assigns a default value, as described in @ref{CREATE TABLE, , @code{CREATE TABLE}}. @findex DROP INDEX @item @code{DROP INDEX} removes an index. This is a MySQL extension to ANSI SQL92. @xref{DROP INDEX}. @item If columns are dropped from a table, the columns are also removed from any index of which they are a part. If all columns that make up an index are dropped, the index is dropped as well. @item If a table contains only one column, the column cannot be dropped. If what you intend is to remove the table, use @code{DROP TABLE} instead. @findex DROP PRIMARY KEY @item @code{DROP PRIMARY KEY} drops the primary index. If no such index exists, it drops the first @code{UNIQUE} index in the table. (MySQL marks the first @code{UNIQUE} key as the @code{PRIMARY KEY} if no @code{PRIMARY KEY} was specified explicitly.) @findex UNIQUE @findex PRIMARY KEY If you add a @code{UNIQUE INDEX} or @code{PRIMARY KEY} to a table, this is stored before any not @code{UNIQUE} index so that MySQL can detect duplicate keys as early as possible. @findex ORDER BY @item @code{ORDER BY} allows you to create the new table with the rows in a specific order. Note that the table will not remain in this order after inserts and deletes. In some cases, it may make sorting easier for MySQL if the table is in order by the column that you wish to order it by later. This option is mainly useful when you know that you are mostly going to query the rows in a certain order; By using this option after big changes to the table, you may be able to get higher performance. @findex ALTER TABLE @item If you use @code{ALTER TABLE} on a @code{MyISAM} table, all non-unique indexes are created in a separate batch (like in @code{REPAIR}). This should make @code{ALTER TABLE} much faster when you have many indexes. @item @findex mysql_info() With the C API function @code{mysql_info()}, you can find out how many records were copied, and (when @code{IGNORE} is used) how many records were deleted due to duplication of unique key values. @item @cindex foreign keys @cindex references The @code{FOREIGN KEY}, @code{CHECK}, and @code{REFERENCES} clauses don't actually do anything. The syntax for them is provided only for compatibility, to make it easier to port code from other SQL servers and to run applications that create tables with references. @xref{Missing functions}. @end itemize Here is an example that shows some of the uses of @code{ALTER TABLE}. We begin with a table @code{t1} that is created as shown below: @example mysql> CREATE TABLE t1 (a INTEGER,b CHAR(10)); @end example To rename the table from @code{t1} to @code{t2}: @example mysql> ALTER TABLE t1 RENAME t2; @end example To change column @code{a} from @code{INTEGER} to @code{TINYINT NOT NULL} (leaving the name the same), and to change column @code{b} from @code{CHAR(10)} to @code{CHAR(20)} as well as renaming it from @code{b} to @code{c}: @example mysql> ALTER TABLE t2 MODIFY a TINYINT NOT NULL, CHANGE b c CHAR(20); @end example To add a new @code{TIMESTAMP} column named @code{d}: @example mysql> ALTER TABLE t2 ADD d TIMESTAMP; @end example To add an index on column @code{d}, and make column @code{a} the primary key: @example mysql> ALTER TABLE t2 ADD INDEX (d), ADD PRIMARY KEY (a); @end example To remove column @code{c}: @example mysql> ALTER TABLE t2 DROP COLUMN c; @end example To add a new @code{AUTO_INCREMENT} integer column named @code{c}: @example mysql> ALTER TABLE t2 ADD c INT UNSIGNED NOT NULL AUTO_INCREMENT, ADD INDEX (c); @end example Note that we indexed @code{c}, because @code{AUTO_INCREMENT} columns must be indexed, and also that we declare @code{c} as @code{NOT NULL}, because indexed columns cannot be @code{NULL}. When you add an @code{AUTO_INCREMENT} column, column values are filled in with sequence numbers for you automatically. You can set the first sequence number by executing @code{SET INSERT_ID=#} before @code{ALTER TABLE} or using the @code{AUTO_INCREMENT = #} table option. @xref{SET OPTION}. With MyISAM tables, if you don't change the @code{AUTO_INCREMENT} column, the sequence number will not be affected. If you drop an @code{AUTO_INCREMENT} column and then add another @code{AUTO_INCREMENT} column, the numbers will start from 1 again. @xref{ALTER TABLE problems}. @node RENAME TABLE, DROP TABLE, ALTER TABLE, Data Definition @subsection @code{RENAME TABLE} Syntax @findex RENAME TABLE @example RENAME TABLE tbl_name TO new_table_name[, tbl_name2 TO new_table_name2,...] @end example The rename is done atomically, which means that no other thread can access any of the tables while the rename is running. This makes it possible to replace a table with an empty one: @example CREATE TABLE new_table (...); RENAME TABLE old_table TO backup_table, new_table TO old_table; @end example The rename is done from left to right, which means that if you want to swap two tables names, you have to: @example RENAME TABLE old_table TO backup_table, new_table TO old_table, backup_table TO new_table; @end example As long as two databases are on the same disk you can also rename from one database to another: @example RENAME TABLE current_database.table_name TO other_database.table_name; @end example When you execute @code{RENAME}, you can't have any locked tables or active transactions. You must also have the @code{ALTER} and @code{DROP} privilege on the original table and @code{CREATE} and @code{INSERT} privilege on the new table. If MySQL encounters any errors in a multiple table rename, it will do a reverse rename for all renamed tables to get everything back to the original state. @node DROP TABLE, CREATE INDEX, RENAME TABLE, Data Definition @subsection @code{DROP TABLE} Syntax @findex DROP TABLE @example DROP TABLE [IF EXISTS] tbl_name [, tbl_name,...] [RESTRICT | CASCADE] @end example @code{DROP TABLE} removes one or more tables. All table data and the table definition are @emph{removed}, so @strong{be careful} with this command! In MySQL Version 3.22 or later, you can use the keywords @code{IF EXISTS} to prevent an error from occurring for tables that don't exist. @code{RESTRICT} and @code{CASCADE} are allowed to make porting easier. For the moment they don't do anything. @strong{NOTE}: @code{DROP TABLE} is not transaction-safe and will automatically commit any active transactions. @node CREATE INDEX, DROP INDEX, DROP TABLE, Data Definition @subsection @code{CREATE INDEX} Syntax @findex CREATE INDEX @cindex indexes @cindex indexes, multi-part @cindex multi-part index @example CREATE [UNIQUE|FULLTEXT] INDEX index_name ON tbl_name (col_name[(length)],... ) @end example The @code{CREATE INDEX} statement doesn't do anything in MySQL prior to Version 3.22. In Version 3.22 or later, @code{CREATE INDEX} is mapped to an @code{ALTER TABLE} statement to create indexes. @xref{ALTER TABLE, , @code{ALTER TABLE}}. Normally, you create all indexes on a table at the time the table itself is created with @code{CREATE TABLE}. @xref{CREATE TABLE, , @code{CREATE TABLE}}. @code{CREATE INDEX} allows you to add indexes to existing tables. A column list of the form @code{(col1,col2,...)} creates a multiple-column index. Index values are formed by concatenating the values of the given columns. For @code{CHAR} and @code{VARCHAR} columns, indexes can be created that use only part of a column, using @code{col_name(length)} syntax. (On @code{BLOB} and @code{TEXT} columns the length is required). The statement shown below creates an index using the first 10 characters of the @code{name} column: @example mysql> CREATE INDEX part_of_name ON customer (name(10)); @end example Because most names usually differ in the first 10 characters, this index should not be much slower than an index created from the entire @code{name} column. Also, using partial columns for indexes can make the index file much smaller, which could save a lot of disk space and might also speed up @code{INSERT} operations! Note that you can only add an index on a column that can have @code{NULL} values or on a @code{BLOB}/@code{TEXT} column if you are using MySQL Version 3.23.2 or newer and are using the @code{MyISAM} table type. For more information about how MySQL uses indexes, see @ref{MySQL indexes, , MySQL indexes}. @code{FULLTEXT} indexes can index only @code{VARCHAR} and @code{TEXT} columns, and only in @code{MyISAM} tables. @code{FULLTEXT} indexes are available in MySQL Version 3.23.23 and later. @ref{Fulltext Search}. @node DROP INDEX, , CREATE INDEX, Data Definition @subsection @code{DROP INDEX} Syntax @findex DROP INDEX @example DROP INDEX index_name ON tbl_name @end example @code{DROP INDEX} drops the index named @code{index_name} from the table @code{tbl_name}. @code{DROP INDEX} doesn't do anything in MySQL prior to Version 3.22. In Version 3.22 or later, @code{DROP INDEX} is mapped to an @code{ALTER TABLE} statement to drop the index. @xref{ALTER TABLE, , @code{ALTER TABLE}}. @node Basic User Commands, Transactional Commands, Data Definition, Reference @section Basic MySQL User Utility Commands @menu * USE:: @code{USE} Syntax * DESCRIBE:: @code{DESCRIBE} Syntax (Get Information About Columns) @end menu @node USE, DESCRIBE, Basic User Commands, Basic User Commands @subsection @code{USE} Syntax @findex USE @example USE db_name @end example The @code{USE db_name} statement tells MySQL to use the @code{db_name} database as the default database for subsequent queries. The database remains current until the end of the session or until another @code{USE} statement is issued: @example mysql> USE db1; mysql> SELECT count(*) FROM mytable; # selects from db1.mytable mysql> USE db2; mysql> SELECT count(*) FROM mytable; # selects from db2.mytable @end example Making a particular database current by means of the @code{USE} statement does not preclude you from accessing tables in other databases. The example below accesses the @code{author} table from the @code{db1} database and the @code{editor} table from the @code{db2} database: @example mysql> USE db1; mysql> SELECT author_name,editor_name FROM author,db2.editor WHERE author.editor_id = db2.editor.editor_id; @end example @cindex Sybase compatibility @cindex compatibility, with Sybase The @code{USE} statement is provided for Sybase compatibility. @node DESCRIBE, , USE, Basic User Commands @subsection @code{DESCRIBE} Syntax (Get Information About Columns) @findex DESC @findex DESCRIBE @example @{DESCRIBE | DESC@} tbl_name @{col_name | wild@} @end example @code{DESCRIBE} is a shortcut for @code{SHOW COLUMNS FROM}. @xref{SHOW DATABASE INFO}. @code{DESCRIBE} provides information about a table's columns. @code{col_name} may be a column name or a string containing the SQL @samp{%} and @samp{_} wild-card characters. If the column types are different than you expect them to be based on a @code{CREATE TABLE} statement, note that MySQL sometimes changes column types. @xref{Silent column changes}. @cindex Oracle compatibility @cindex compatibility, with Oracle This statement is provided for Oracle compatibility. The @code{SHOW} statement provides similar information. @xref{SHOW, , @code{SHOW}}. @node Transactional Commands, Fulltext Search, Basic User Commands, Reference @section MySQL Transactional and Locking Commands @menu * COMMIT:: @code{BEGIN/COMMIT/ROLLBACK} Syntax * LOCK TABLES:: @code{LOCK TABLES/UNLOCK TABLES} Syntax * SET TRANSACTION:: @code{SET TRANSACTION} Syntax @end menu @node COMMIT, LOCK TABLES, Transactional Commands, Transactional Commands @subsection @code{BEGIN/COMMIT/ROLLBACK} Syntax @findex BEGIN @findex COMMIT @findex ROLLBACK By default, MySQL runs in @code{autocommit} mode. This means that as soon as you execute an update, MySQL will store the update on disk. If you are using transactions safe tables (like @code{BDB}, @code{InnoDB}, you can put MySQL into non-@code{autocommit} mode with the following command: @example SET AUTOCOMMIT=0 @end example After this you must use @code{COMMIT} to store your changes to disk or @code{ROLLBACK} if you want to ignore the changes you have made since the beginning of your transaction. If you want to switch from @code{AUTOCOMMIT} mode for one series of statements, you can use the @code{BEGIN} or @code{BEGIN WORK} statement: @example BEGIN; SELECT @@A:=SUM(salary) FROM table1 WHERE type=1; UPDATE table2 SET summmary=@@A WHERE type=1; COMMIT; @end example Note that if you are using non-transaction-safe tables, the changes will be stored at once, independent of the status of the @code{autocommit} mode. If you do a @code{ROLLBACK} when you have updated a non-transactional table you will get an error (@code{ER_WARNING_NOT_COMPLETE_ROLLBACK}) as a warning. All transactional safe tables will be restored but any non-transactional table will not change. If you are using @code{BEGIN} or @code{SET AUTOCOMMIT=0}, you should use the MySQL binary log for backups instead of the older update log. Transactions are stored in the binary log in one chunk, upon @code{COMMIT}, to ensure that transactions which are rolled back are not stored. @xref{Binary log}. The following commands automatically end a transaction (as if you had done a @code{COMMIT} before executing the command): @multitable @columnfractions .33 .33 .33 @item @code{ALTER TABLE} @tab @code{BEGIN} @tab @code{CREATE INDEX} @item @code{DROP DATABASE} @tab @code{DROP TABLE} @tab @code{RENAME TABLE} @item @code{TRUNCATE} @end multitable You can change the isolation level for transactions with @code{SET TRANSACTION ISOLATION LEVEL ...}. @xref{SET TRANSACTION}. @node LOCK TABLES, SET TRANSACTION, COMMIT, Transactional Commands @subsection @code{LOCK TABLES/UNLOCK TABLES} Syntax @findex LOCK TABLES @findex UNLOCK TABLES @example LOCK TABLES tbl_name [AS alias] @{READ | [READ LOCAL] | [LOW_PRIORITY] WRITE@} [, tbl_name @{READ | [LOW_PRIORITY] WRITE@} ...] ... UNLOCK TABLES @end example @code{LOCK TABLES} locks tables for the current thread. @code{UNLOCK TABLES} releases any locks held by the current thread. All tables that are locked by the current thread are automatically unlocked when the thread issues another @code{LOCK TABLES}, or when the connection to the server is closed. The main reasons to use @code{LOCK TABLES} are for emulating transactions or getting more speed when updating tables. This is explained in more detail later. If a thread obtains a @code{READ} lock on a table, that thread (and all other threads) can only read from the table. If a thread obtains a @code{WRITE} lock on a table, then only the thread holding the lock can @code{READ} from or @code{WRITE} to the table. Other threads are blocked. The difference between @code{READ LOCAL} and @code{READ} is that @code{READ LOCAL} allows non-conflicting @code{INSERT} statements to execute while the lock is held. This can't however be used if you are going to manipulate the database files outside MySQL while you hold the lock. When you use @code{LOCK TABLES}, you must lock all tables that you are going to use and you must use the same alias that you are going to use in your queries! If you are using a table multiple times in a query (with aliases), you must get a lock for each alias! @code{WRITE} locks normally have higher priority than @code{READ} locks, to ensure that updates are processed as soon as possible. This means that if one thread obtains a @code{READ} lock and then another thread requests a @code{WRITE} lock, subsequent @code{READ} lock requests will wait until the @code{WRITE} thread has gotten the lock and released it. You can use @code{LOW_PRIORITY WRITE} locks to allow other threads to obtain @code{READ} locks while the thread is waiting for the @code{WRITE} lock. You should only use @code{LOW_PRIORITY WRITE} locks if you are sure that there will eventually be a time when no threads will have a @code{READ} lock. @code{LOCK TABLES} works as follows: @enumerate @item Sort all tables to be locked in a internally defined order (from the user standpoint the order is undefined). @item If a table is locked with a read and a write lock, put the write lock before the read lock. @item Lock one table at a time until the thread gets all locks. @end enumerate This policy ensures that table locking is deadlock free. There is however other things one needs to be aware of with this schema: If you are using a @code{LOW_PRIORITY_WRITE} lock for a table, this means only that MySQL will wait for this particlar lock until there is no threads that wants a @code{READ} lock. When the thread has got the @code{WRITE} lock and is waiting to get the lock for the next table in the lock table list, all other threads will wait for the @code{WRITE} lock to be released. If this becomes a serious problem with your application, you should consider converting some of your tables to transactions safe tables. You can safely kill a thread that is waiting for a table lock with @code{KILL}. @xref{KILL}. Note that you should @strong{NOT} lock any tables that you are using with @code{INSERT DELAYED}. This is because that in this case the @code{INSERT} is done by a separate thread. Normally, you don't have to lock tables, as all single @code{UPDATE} statements are atomic; no other thread can interfere with any other currently executing SQL statement. There are a few cases when you would like to lock tables anyway: @itemize @bullet @item If you are going to run many operations on a bunch of tables, it's much faster to lock the tables you are going to use. The downside is, of course, that no other thread can update a @code{READ}-locked table and no other thread can read a @code{WRITE}-locked table. The reason some things are faster under @code{LOCK TABLES} is that MySQL will not flush the key cache for the locked tables until @code{UNLOCK TABLES} is called (normally the key cache is flushed after each SQL statement). This speeds up inserting/updateing/deletes on @code{MyISAM} tables. @item If you are using a table handler in MySQL that doesn't support transactions, you must use @code{LOCK TABLES} if you want to ensure that no other thread comes between a @code{SELECT} and an @code{UPDATE}. The example shown below requires @code{LOCK TABLES} in order to execute safely: @example mysql> LOCK TABLES trans READ, customer WRITE; mysql> select sum(value) from trans where customer_id= some_id; mysql> update customer set total_value=sum_from_previous_statement where customer_id=some_id; mysql> UNLOCK TABLES; @end example Without @code{LOCK TABLES}, there is a chance that another thread might insert a new row in the @code{trans} table between execution of the @code{SELECT} and @code{UPDATE} statements. @end itemize By using incremental updates (@code{UPDATE customer SET value=value+new_value}) or the @code{LAST_INSERT_ID()} function, you can avoid using @code{LOCK TABLES} in many cases. You can also solve some cases by using the user-level lock functions @code{GET_LOCK()} and @code{RELEASE_LOCK()}. These locks are saved in a hash table in the server and implemented with @code{pthread_mutex_lock()} and @code{pthread_mutex_unlock()} for high speed. @xref{Miscellaneous functions}. See @ref{Internal locking}, for more information on locking policy. You can lock all tables in all databases with read locks with the @code{FLUSH TABLES WITH READ LOCK} command. @xref{FLUSH}. This is very convenient way to get backups if you have a file system, like Veritas, that can take snapshots in time. @strong{NOTE}: @code{LOCK TABLES} is not transaction-safe and will automatically commit any active transactions before attempting to lock the tables. @node SET TRANSACTION, , LOCK TABLES, Transactional Commands @subsection @code{SET TRANSACTION} Syntax @findex ISOLATION LEVEL @example SET [GLOBAL | SESSION] TRANSACTION ISOLATION LEVEL [READ UNCOMMITTED |�READ COMMITTED | REPEATABLE READ | SERIALIZABLE] @end example Sets the transaction isolation level for the global, whole session or the next transaction. The default behavior is to set the isolation level for the next (not started) transaction. If you set the @code{GLOBAL} privilege it will affect all new created threads. You will need the @code{PROCESS} privilege to do do this. Setting the @code{SESSION} privilege will affect the following and all future transactions. You can set the default isolation level for @code{mysqld} with @code{--transaction-isolation=...}. @xref{Command-line options}. @node Fulltext Search, , Transactional Commands, Reference @section MySQL Full-text Search @cindex searching, full-text @cindex full-text search @cindex FULLTEXT Since Version 3.23.23, MySQL has support for full-text indexing and searching. Full-text indexes in MySQL are an index of type @code{FULLTEXT}. @code{FULLTEXT} indexes can be created from @code{VARCHAR} and @code{TEXT} columns at @code{CREATE TABLE} time or added later with @code{ALTER TABLE} or @code{CREATE INDEX}. For large datasets, adding @code{FULLTEXT} index with @code{ALTER TABLE} (or @code{CREATE INDEX}) would be much faster than inserting rows into the empty table with a @code{FULLTEXT} index. Full-text search is performed with the @code{MATCH} function. @example mysql> CREATE TABLE articles ( -> id INT UNSIGNED AUTO_INCREMENT NOT NULL PRIMARY KEY, -> title VARCHAR(200), -> body TEXT, -> FULLTEXT (title,body) -> ); Query OK, 0 rows affected (0.00 sec) mysql> INSERT INTO articles VALUES -> (0,'MySQL Tutorial', 'DBMS stands for DataBase Management ...'), -> (0,'How To Use MySQL Efficiently', 'After you went through a ...'), -> (0,'Optimizing MySQL','In this tutorial we will show how to ...'), -> (0,'1001 MySQL Trick','1. Never run mysqld as root. 2. Normalize ...'), -> (0,'MySQL vs. YourSQL', 'In the following database comparison we ...'), -> (0,'MySQL Security', 'When configured properly, MySQL could be ...'); Query OK, 5 rows affected (0.00 sec) Records: 5 Duplicates: 0 Warnings: 0 mysql> SELECT * FROM articles WHERE MATCH (title,body) AGAINST ('database'); +----+-------------------+---------------------------------------------+ | id | title | body | +----+-------------------+---------------------------------------------+ | 5 | MySQL vs. YourSQL | In the following database comparison we ... | | 1 | MySQL Tutorial | DBMS stands for DataBase Management ... | +----+-------------------+---------------------------------------------+ 2 rows in set (0.00 sec) @end example The function @code{MATCH} matches a natural language query @code{AGAINST} a text collection (which is simply the set of columns covered by a @code{FULLTEXT} index). For every row in a table it returns relevance - a similarity measure between the text in that row (in the columns that are part of the collection) and the query. When it is used in a @code{WHERE} clause (see example above) the rows returned are automatically sorted with relevance decreasing. Relevance is a non-negative floating-point number. Zero relevance means no similarity. Relevance is computed based on the number of words in the row, the number of unique words in that row, the total number of words in the collection, and the number of documents (rows) that contain a particular word. The above is a basic example of using @code{MATCH} function. Rows are returned with relevance decreasing. @example mysql> SELECT id,MATCH (title,body) AGAINST ('Tutorial') FROM articles; +----+-----------------------------------------+ | id | MATCH (title,body) AGAINST ('Tutorial') | +----+-----------------------------------------+ | 1 | 0.64840710366884 | | 2 | 0 | | 3 | 0.66266459031789 | | 4 | 0 | | 5 | 0 | | 6 | 0 | +----+-----------------------------------------+ 5 rows in set (0.00 sec) @end example This example shows how to retrieve the relevances. As neither @code{WHERE} nor @code{ORDER BY} clauses are present, returned rows are not ordered. @example mysql> SELECT id, body, MATCH (title,body) AGAINST ( -> 'Security implications of running MySQL as root') AS score -> FROM articles WHERE MATCH (title,body) AGAINST -> ('Security implications of running MySQL as root'); +----+-----------------------------------------------+-----------------+ | id | body | score | +----+-----------------------------------------------+-----------------+ | 4 | 1. Never run mysqld as root. 2. Normalize ... | 1.5055546709332 | | 6 | When configured properly, MySQL could be ... | 1.31140957288 | +----+-----------------------------------------------+-----------------+ 2 rows in set (0.00 sec) @end example This is more complex example - the query returns the relevance and still sorts the rows with relevance decreasing. To achieve it one should specify @code{MATCH} twice. Note, that this will cause no additional overhead, as MySQL optimizer will notice that these two @code{MATCH} calls are identical and will call full-text search code only once. MySQL uses a very simple parser to split text into words. A ``word'' is any sequence of letters, numbers, @samp{'}, and @samp{_}. Any ``word'' that is present in the stopword list or just too short (3 characters or less) is ignored. Every correct word in the collection and in the query is weighted, according to its significance in the query or collection. This way, a word that is present in many documents will have lower weight (and may even have a zero weight), because it has lower semantic value in this particular collection. Otherwise, if the word is rare, it will receive a higher weight. The weights of the words are then combined to compute the relevance of the row. Such a technique works best with large collections (in fact, it was carefully tuned this way). For very small tables, word distribution does not reflect adequately their semantical value, and this model may sometimes produce bizarre results. @example mysql> SELECT * FROM articles WHERE MATCH (title,body) AGAINST ('MySQL'); Empty set (0.00 sec) @end example Search for the word @code{MySQL} produces no results in the above example. Word @code{MySQL} is present in more than half of rows, and as such, is effectively treated as a stopword (that is, with semantical value zero). It is, really, the desired behavior - a natural language query should not return every second row in 1GB table. A word that matches half of rows in a table is less likely to locate relevant documents. In fact, it will most likely find plenty of irrelevant documents. We all know this happens far too often when we are trying to find something on the Internet with a search engine. It is with this reasoning that such rows have been assigned a low semantical value in @strong{this particular dataset}. @menu * Fulltext restrictions:: Fulltext restrictions * Fulltext Fine-tuning:: Fine-tuning MySQL Full-text Search * Fulltext Features to Appear in MySQL 4.0:: New Features of Full-text Search to Appear in MySQL 4.0 * Fulltext TODO:: Full-text Search TODO @end menu @node Fulltext restrictions, Fulltext Fine-tuning, Fulltext Search, Fulltext Search @subsection Fulltext restrictions @itemize @bullet @item All parameters to the @code{MATCH} function must be columns from the same table that is part of the same fulltext index. @item The argument to @code{AGAINST} must be a constant string. @end itemize @node Fulltext Fine-tuning, Fulltext Features to Appear in MySQL 4.0, Fulltext restrictions, Fulltext Search @subsection Fine-tuning MySQL Full-text Search Unfortunately, full-text search has no user-tunable parameters yet, although adding some is very high on the TODO. However, if you have a MySQL source distribution (@xref{Installing source}.), you can somewhat alter the full-text search behavior. Note that full-text search was carefully tuned for the best searching effectiveness. Modifying the default behavior will, in most cases, only make the search results worse. Do not alter the MySQL sources unless you know what you are doing! @itemize @bullet @item Minimal length of word to be indexed is defined in @code{myisam/ftdefs.h} file by the line @example #define MIN_WORD_LEN 4 @end example Change it to the value you prefer, recompile MySQL, and rebuild your @code{FULLTEXT} indexes. @item The stopword list is defined in @code{myisam/ft_static.c} Modify it to your taste, recompile MySQL and rebuild your @code{FULLTEXT} indexes. @item The 50% threshold is caused by the particular weighting scheme chosen. To disable it, change the following line in @code{myisam/ftdefs.h}: @example #define GWS_IN_USE GWS_PROB @end example to @example #define GWS_IN_USE GWS_FREQ @end example and recompile MySQL. There is no need to rebuild the indexes in this case. @end itemize @node Fulltext Features to Appear in MySQL 4.0, Fulltext TODO, Fulltext Fine-tuning, Fulltext Search @subsection New Features of Full-text Search to Appear in MySQL 4.0 This section includes a list of the fulltext features that are already implemented in the 4.0 tree. It explains @strong{More functions for full-text search} entry of @ref{TODO MySQL 4.0}. @itemize @bullet @item @code{REPAIR TABLE} with @code{FULLTEXT} indexes, @code{ALTER TABLE} with @code{FULLTEXT} indexes, and @code{OPTIMIZE TABLE} with @code{FULLTEXT} indexes are now up to 100 times faster. @item @code{MATCH ... AGAINST} is going to supports the following @strong{boolean operators}: @itemize @bullet @item @code{+}word means the that word @strong{must} be present in every row returned. @item @code{-}word means the that word @strong{must not} be present in every row returned. @item @code{<} and @code{>} can be used to decrease and increase word weight in the query. @item @code{~} can be used to assign a @strong{negative} weight to a noise word. @item @code{*} is a truncation operator. @end itemize Boolean search utilizes a more simplistic way of calculating the relevance, that does not have a 50% threshold. @item Searches are now up to 2 times faster due to optimized search algorithm. @item Utility program @code{ft_dump} added for low-level @code{FULLTEXT} index operations (querying/dumping/statistics). @end itemize @node Fulltext TODO, , Fulltext Features to Appear in MySQL 4.0, Fulltext Search @subsection Full-text Search TODO @itemize @bullet @item Make all operations with @code{FULLTEXT} index @strong{faster}. @item Support for braces @code{()} in boolean full-text search. @item Phrase search, proximity operators @item Boolean search can work without @code{FULLTEXT} index (yes, @strong{very} slow). @item Support for "always-index words". They could be any strings the user wants to treat as words, examples are "C++", "AS/400", "TCP/IP", etc. @item Support for full-text search in @code{MERGE} tables. @item Support for multi-byte charsets. @item Make stopword list to depend of the language of the data. @item Stemming (dependent of the language of the data, of course). @item Generic user-supplyable UDF (?) preparser. @item Make the model more flexible (by adding some adjustable parameters to @code{FULLTEXT} in @code{CREATE/ALTER TABLE}). @end itemize @node Table types, Clients, Reference, Top @chapter MySQL Table Types @cindex table types, choosing @cindex @code{BDB} table type @cindex @code{Berkeley_db} table type @cindex @code{HEAP} table type @cindex @code{ISAM} table type @cindex @code{InnoDB} table type @cindex @code{MERGE} table type @cindex MySQL table types @cindex @code{MyISAM} table type @cindex types, of tables As of MySQL Version 3.23.6, you can choose between three basic table formats (@code{ISAM}, @code{HEAP} and @code{MyISAM}. Newer MySQL may support additional table type (@code{BDB}, or @code{InnoDB}), depending on how you compile it. When you create a new table, you can tell MySQL which table type it should use for the table. MySQL will always create a @code{.frm} file to hold the table and column definitions. Depending on the table type, the index and data will be stored in other files. Note that to use @code{InnoDB} tables you have to use at least the @code{innodb_data_file_path} startup option. @xref{InnoDB start}. The default table type in MySQL is @code{MyISAM}. If you are trying to use a table type that is not compiled-in or activated, MySQL will instead create a table of type @code{MyISAM}. This is a very useful feature when you want to copy tables between different SQL servers that supports different table types (like copying tables to a slave that is optimized for speed by not having transactional tables). This automatic table changing can however also be very confusing for new MySQL users. We plan to fix this by introducing warnings in MySQL 4.0 and giving a warning when a table type is automatically changed. You can convert tables between different types with the @code{ALTER TABLE} statement. @xref{ALTER TABLE, , @code{ALTER TABLE}}. Note that MySQL supports two different kinds of tables. Transaction-safe tables (@code{BDB}, @code{InnoDB} and not transaction-safe tables (@code{HEAP}, @code{ISAM}, @code{MERGE}, and @code{MyISAM}). Advantages of transaction-safe tables (TST): @itemize @bullet @item Safer. Even if MySQL crashes or you get hardware problems, you can get your data back, either by automatic recovery or from a backup + the transaction log. @item You can combine many statements and accept these all in one go with the @code{COMMIT} command. @item You can execute @code{ROLLBACK} to ignore your changes (if you are not running in auto commit mode). @item If an update fails, all your changes will be restored. (With NTST tables all changes that have taken place are permanent) @end itemize Advantages of not transaction-safe tables (NTST): @itemize @bullet @item Much faster as there is no transaction overhead. @item Will use less disk space as there is no overhead of transactions. @item Will use less memory to do updates. @end itemize You can combine TST and NTST tables in the same statements to get the best of both worlds. @menu * MyISAM:: MyISAM tables * MERGE:: MERGE tables * ISAM:: ISAM tables * HEAP:: HEAP tables * BDB:: BDB or Berkeley_db tables * InnoDB:: InnoDB tables @end menu @node MyISAM, MERGE, Table types, Table types @section MyISAM Tables @code{MyISAM} is the default table type in MySQL Version 3.23. It's based on the @code{ISAM} code and has a lot of useful extensions. The index is stored in a file with the @code{.MYI} (MYIndex) extension, and the data is stored in a file with the @code{.MYD} (MYData) extension. You can check/repair @code{MyISAM} tables with the @code{myisamchk} utility. @xref{Crash recovery}. You can compress @code{MyISAM} tables with @code{myisampack} to take up much less space. @xref{myisampack}. The following is new in @code{MyISAM}: @itemize @bullet @item There is a flag in the @code{MyISAM} file that indicates whether or not the table was closed correctly. If @code{mysqld} is started with @code{--myisam-recover}, @code{MyISAM} tables will automatically be checked and/or repaired on open if the table wasn't closed properly. @item You can @code{INSERT} new rows in a table that doesn't have free blocks in the middle of the data file, at the same time other threads are reading from the table (concurrent insert). An free block can come from an update of a dynamic length row with much data to a row with less data or when deleting rows. When all free blocks are used up, all future inserts will be concurrent again. @item Support for big files (63-bit) on filesystems/operating systems that support big files. @item All data is stored with the low byte first. This makes the data machine and OS independent. The only requirement is that the machine uses two's-complement signed integers (as every machine for the last 20 years has) and IEEE floating-point format (also totally dominant among mainstream machines). The only area of machines that may not support binary compatibility are embedded systems (because they sometimes have peculiar processors). There is no big speed penalty in storing data low byte first; The bytes in a table row is normally unaligned and it doesn't take that much more power to read an unaligned byte in order than in reverse order. The actual fetch-column-value code is also not time critical compared to other code. @item All number keys are stored with high byte first to give better index compression. @item Internal handling of one @code{AUTO_INCREMENT} column. @code{MyISAM} will automatically update this on @code{INSERT/UPDATE}. The @code{AUTO_INCREMENT} value can be reset with @code{myisamchk}. This will make @code{AUTO_INCREMENT} columns faster (at least 10 %) and old numbers will not be reused as with the old @code{ISAM}. Note that when an @code{AUTO_INCREMENT} is defined on the end of a multi-part-key the old behavior is still present. @item When inserted in sorted order (as when you are using an @code{AUTO_INCREMENT} column) the key tree will be split so that the high node only contains one key. This will improve the space utilization in the key tree. @item @code{BLOB} and @code{TEXT} columns can be indexed. @item @code{NULL} values are allowed in indexed columns. This takes 0-1 bytes/key. @item Maximum key length is 500 bytes by default (can be changed by recompiling). In cases of keys longer than 250 bytes, a bigger key block size than the default of 1024 bytes is used for this key. @item Maximum number of keys/table is 32 as default. This can be enlarged to 64 without having to recompile @code{myisamchk}. @item @code{myisamchk} will mark tables as checked if one runs it with @code{--update-state}. @code{myisamchk --fast} will only check those tables that don't have this mark. @item @code{myisamchk -a} stores statistics for key parts (and not only for whole keys as in @code{ISAM}). @item Dynamic size rows will now be much less fragmented when mixing deletes with updates and inserts. This is done by automatically combining adjacent deleted blocks and by extending blocks if the next block is deleted. @item @code{myisampack} can pack @code{BLOB} and @code{VARCHAR} columns. @item You can use put the datafile and index file on different directories to get more speed (with the @code{DATA/INDEX DIRECTORY="path"} option to @code{CREATE TABLE}). @xref{CREATE TABLE}. @end itemize @code{MyISAM} also supports the following things, which MySQL will be able to use in the near future: @itemize @bullet @item Support for a true @code{VARCHAR} type; A @code{VARCHAR} column starts with a length stored in 2 bytes. @item Tables with @code{VARCHAR} may have fixed or dynamic record length. @item @code{VARCHAR} and @code{CHAR} may be up to 64K. All key segments have their own language definition. This will enable MySQL to have different language definitions per column. @item A hashed computed index can be used for @code{UNIQUE}. This will allow you to have @code{UNIQUE} on any combination of columns in a table. (You can't search on a @code{UNIQUE} computed index, however.) @end itemize Note that index files are usually much smaller with @code{MyISAM} than with @code{ISAM}. This means that @code{MyISAM} will normally use less system resources than @code{ISAM}, but will need more CPU when inserting data into a compressed index. The following options to @code{mysqld} can be used to change the behavior of @code{MyISAM} tables. @xref{SHOW VARIABLES}. @multitable @columnfractions .40 .60 @item @strong{Option} @tab @strong{Meaning} @item @code{--myisam-recover=#} @tab Automatic recover of crashed tables. @item @code{-O myisam_sort_buffer_size=#} @tab Buffer used when recovering tables. @item @code{--delay-key-write-for-all-tables} @tab Don't flush key buffers between writes for any MyISAM table @item @code{-O myisam_max_extra_sort_file_size=#} @tab Used to help MySQL to decide when to use the slow but safe key cache index create method. @strong{NOTE} that this parameter is given in megabytes! @item @code{-O myisam_max_sort_file_size=#} @tab Don't use the fast sort index method to created index if the temporary file would get bigger than this. @strong{NOTE} that this paramter is given in megabytes! @end multitable The automatic recovery is activated if you start @code{mysqld} with @code{--myisam-recover=#}. @xref{Command-line options}. On open, the table is checked if it's marked as crashed or if the open count variable for the table is not 0 and you are running with @code{--skip-locking}. If either of the above is true the following happens. @itemize @bullet @item The table is checked for errors. @item If we found an error, try to do a fast repair (with sorting and without re-creating the data file) of the table. @item If the repair fails because of an error in the data file (for example a duplicate key error), we try again, but this time we re-create the data file. @item If the repair fails, retry once more with the old repair option method (write row by row without sorting) which should be able to repair any type of error with little disk requirements.. @end itemize If the recover wouldn't be able to recover all rows from a previous completed statement and you didn't specify @code{FORCE} as an option to @code{myisam-recover}, then the automatic repair will abort with an error message in the error file: @example Error: Couldn't repair table: test.g00pages @end example If you in this case had used the @code{FORCE} option you would instead have got a warning in the error file: @example Warning: Found 344 of 354 rows when repairing ./test/g00pages @end example Note that if you run automatic recover with the @code{BACKUP} option, you should have a cron script that automatically moves file with names like @file{tablename-datetime.BAK} from the database directories to a backup media. @xref{Command-line options}. @menu * Key space:: Space needed for keys * MyISAM table formats:: MyISAM table formats * MyISAM table problems:: MyISAM table problems. @end menu @node Key space, MyISAM table formats, MyISAM, MyISAM @subsection Space Needed for Keys @cindex key space, MyISAM MySQL can support different index types, but the normal type is ISAM or MyISAM. These use a B-tree index, and you can roughly calculate the size for the index file as @code{(key_length+4)/0.67}, summed over all keys. (This is for the worst case when all keys are inserted in sorted order and we don't have any compressed keys.) String indexes are space compressed. If the first index part is a string, it will also be prefix compressed. Space compression makes the index file smaller than the above figures if the string column has a lot of trailing space or is a @code{VARCHAR} column that is not always used to the full length. Prefix compression is used on keys that start with a string. Prefix compression helps if there are many strings with an identical prefix. In @code{MyISAM} tables, you can also prefix compress numbers by specifying @code{PACK_KEYS=1} when you create the table. This helps when you have many integer keys that have an identical prefix when the numbers are stored high-byte first. @node MyISAM table formats, MyISAM table problems, Key space, MyISAM @subsection MyISAM Table Formats @strong{MyISAM} supports 3 different table types. Two of them are chosen automatically depending on the type of columns you are using. The third, compressed tables, can only be created with the @code{myisampack} tool. @menu * Static format:: Static (Fixed-length) table characteristics * Dynamic format:: Dynamic table characteristics * Compressed format:: Compressed table characteristics @end menu @node Static format, Dynamic format, MyISAM table formats, MyISAM table formats @subsubsection Static (Fixed-length) Table Characteristics This is the default format. It's used when the table contains no @code{VARCHAR}, @code{BLOB}, or @code{TEXT} columns. This format is the simplest and most secure format. It is also the fastest of the on-disk formats. The speed comes from the easy way data can be found on disk. When looking up something with an index and static format it is very simple. Just multiply the row number by the row length. Also, when scanning a table it is very easy to read a constant number of records with each disk read. The security is evidenced if your computer crashes when writing to a fixed-size MyISAM file, in which case @code{myisamchk} can easily figure out where each row starts and ends. So it can usually reclaim all records except the partially written one. Note that in MySQL all indexes can always be reconstructed: @itemize @bullet @item All @code{CHAR}, @code{NUMERIC}, and @code{DECIMAL} columns are space-padded to the column width. @item Very quick. @item Easy to cache. @item Easy to reconstruct after a crash, because records are located in fixed positions. @item Doesn't have to be reorganized (with @code{myisamchk}) unless a huge number of records are deleted and you want to return free disk space to the operating system. @item Usually requires more disk space than dynamic tables. @end itemize @node Dynamic format, Compressed format, Static format, MyISAM table formats @subsubsection Dynamic Table Characteristics @cindex dynamic table characteristics @cindex tables, dynamic This format is used if the table contains any @code{VARCHAR}, @code{BLOB}, or @code{TEXT} columns or if the table was created with @code{ROW_FORMAT=dynamic}. This format is a little more complex because each row has to have a header that says how long it is. One record can also end up at more than one location when it is made longer at an update. @cindex tables, defragment You can use @code{OPTIMIZE table} or @code{myisamchk} to defragment a table. If you have static data that you access/change a lot in the same table as some @code{VARCHAR} or @code{BLOB} columns, it might be a good idea to move the dynamic columns to other tables just to avoid fragmentation: @itemize @bullet @item All string columns are dynamic (except those with a length less than 4). @item Each record is preceded by a bitmap indicating which columns are empty (@code{''}) for string columns, or zero for numeric columns. (This isn't the same as columns containing @code{NULL} values.) If a string column has a length of zero after removal of trailing spaces, or a numeric column has a value of zero, it is marked in the bit map and not saved to disk. Non-empty strings are saved as a length byte plus the string contents. @item Usually takes much less disk space than fixed-length tables. @item Each record uses only as much space as is required. If a record becomes larger, it is split into as many pieces as are required. This results in record fragmentation. @item If you update a row with information that extends the row length, the row will be fragmented. In this case, you may have to run @code{myisamchk -r} from time to time to get better performance. Use @code{myisamchk -ei tbl_name} for some statistics. @item Not as easy to reconstruct after a crash, because a record may be fragmented into many pieces and a link (fragment) may be missing. @item The expected row length for dynamic sized records is: @example 3 + (number of columns + 7) / 8 + (number of char columns) + packed size of numeric columns + length of strings + (number of NULL columns + 7) / 8 @end example There is a penalty of 6 bytes for each link. A dynamic record is linked whenever an update causes an enlargement of the record. Each new link will be at least 20 bytes, so the next enlargement will probably go in the same link. If not, there will be another link. You may check how many links there are with @code{myisamchk -ed}. All links may be removed with @code{myisamchk -r}. @end itemize @node Compressed format, , Dynamic format, MyISAM table formats @subsubsection Compressed Table Characteristics @cindex tables, compressed format This is a read-only type that is generated with the optional @code{myisampack} tool (@code{pack_isam} for @code{ISAM} tables): @itemize @bullet @item All MySQL distributions, even those that existed before MySQL went GPL, can read tables that were compressed with @code{myisampack}. @item Compressed tables take very little disk space. This minimizes disk usage, which is very nice when using slow disks (like CD-ROMs). @item Each record is compressed separately (very little access overhead). The header for a record is fixed (1-3 bytes) depending on the biggest record in the table. Each column is compressed differently. Some of the compression types are: @itemize @minus @item There is usually a different Huffman table for each column. @item Suffix space compression. @item Prefix space compression. @item Numbers with value @code{0} are stored using 1 bit. @item If values in an integer column have a small range, the column is stored using the smallest possible type. For example, a @code{BIGINT} column (8 bytes) may be stored as a @code{TINYINT} column (1 byte) if all values are in the range @code{0} to @code{255}. @item If a column has only a small set of possible values, the column type is converted to @code{ENUM}. @item A column may use a combination of the above compressions. @end itemize @item Can handle fixed- or dynamic-length records, but not @code{BLOB} or @code{TEXT} columns. @item Can be uncompressed with @code{myisamchk}. @end itemize @node MyISAM table problems, , MyISAM table formats, MyISAM @subsection MyISAM table problems. The file format that MySQL uses to store data has been extensively tested, but there are always circumstances that may cause database tables to become corrupted. @menu * Corrupted MyISAM tables:: Corrupted MyISAM tables. * MyISAM table close:: Clients is using or hasn't closed the table properly @end menu @node Corrupted MyISAM tables, MyISAM table close, MyISAM table problems, MyISAM table problems @subsubsection Corrupted MyISAM tables. Even if the MyISAM table format is very reliable (all changes to a table is written before the SQL statements returns) , you can still get corrupted tables if some of the following things happens: @itemize @bullet @item The @code{mysqld} process being killed in the middle of a write. @item Unexpected shutdown of the computer (for example, if the computer is turned off). @item A hardware error. @item You are using an external program (like myisamchk) on a live table. @item A software bug in the MySQL or MyISAM code. @end itemize Typial typical symptoms for a corrupt table is: @itemize @bullet @item You get the error @code{Incorrect key file for table: '...'. Try to repair it} while selecting data from the table. @item Queries doesn't find rows in the table or returns incomplete data. @end itemize You can check if a table is ok with the command @code{CHECK TABLE}. @xref{CHECK TABLE}. You can repair a corrupted table with @code{REPAIR TABLE}. @xref{REPAIR TABLE}. You can also repair a table, when @code{mysqld} is not running with the @code{myisamchk} command. @code{myisamchk syntax}. If your tables get corrupted a lot you should try to find the reason for this! @xref{Crashing}. In this case the most important thing to know is if the table got corrupted if the @code{mysqld} died (one can easily verify this by checking if there is a recent row @code{restarted mysqld} in the mysqld error file). If this isn't the case, then you should try to make a test case of this. @xref{Reproduceable test case}. @node MyISAM table close, , Corrupted MyISAM tables, MyISAM table problems @subsubsection Clients is using or hasn't closed the table properly Each @code{MyISAM} @code{.MYI} file has in the header a counter that can be used to check if a table has been closed properly. If you get the following warning from @code{CHECK TABLE} or @code{myisamchk}: @example # clients is using or hasn't closed the table properly @end example this means that this counter has come out of sync. This doesn't mean that the table is corrupted, but means that you should at least do a check on the table to verify that it's ok. The counter works as follows: @itemize @bullet @item The first time a table is updated in MySQL, a counter in the header of the index files is incremented. @item The counter is not changed during further updates. @item When the last instance of a table is closed (because of a @code{FLUSH} or because there isn't room in the table cache) the counter is decremented if the table has been updated at any point. @item When you repair the table or check the table and it was ok, the counter is reset to 0. @item To avoid problems with interaction with other processes that may do a check on the table, the counter is not decremented on close if it was 0. @end itemize In other words, the only ways this can go out of sync are: @itemize @bullet @item The @code{MyISAM} tables are copied without a @code{LOCK} and @code{FLUSH TABLES}. @item MySQL has crashed between an update and the final close (Note that the table may still be ok, as MySQL always issues writes for everything between each statement). @item Someone has done a @code{myisamchk --repair} or @code{myisamchk --update-state}on a table that was in use by @code{mysqld}. @item Many @code{mysqld} servers are using the table and one has done a @code{REPAIR} or @code{CHECK} of the table while it was in use by another server. In this setup the @code{CHECK} is safe to do (even if you will get the warning from other servers), but @code{REPAIR} should be avoided as it currently replaces the data file with a new one, which is not signaled to the other servers. @end itemize @node MERGE, ISAM, MyISAM, Table types @section MERGE Tables @cindex tables, merging @cindex MERGE tables, defined @code{MERGE} tables are new in MySQL Version 3.23.25. The code is still in gamma, but should be resonable stable. A @code{MERGE} table is a collection of identical @code{MyISAM} tables that can be used as one. You can only @code{SELECT}, @code{DELETE}, and @code{UPDATE} from the collection of tables. If you @code{DROP} the @code{MERGE} table, you are only dropping the @code{MERGE} specification. Note that @code{DELETE FROM merge_table} used without a @code{WHERE} will only clear the mapping for the table, not delete everything in the mapped tables. (We plan to fix this in 4.0). With identical tables we mean that all tables are created with identical column and key information. You can't put a MERGE over tables where the columns are packed differently, doesn't have exactly the same columns or have the keys in different order. Some of the tables can however be compressed with @code{myisampack}. @xref{myisampack}. When you create a @code{MERGE} table, you will get a @code{.frm} table definition file and a @code{.MRG} table list file. The @code{.MRG} just contains a list of the index files (@code{.MYI} files) that should be used as one. All used tables must be in the same database as the @code{MERGE} table itself. For the moment you need to have @code{SELECT}, @code{UPDATE}, and @code{DELETE} privileges on the tables you map to a @code{MERGE} table. @code{MERGE} tables can help you solve the following problems: @itemize @bullet @item Easily manage a set of log tables. For example, you can put data from different months into separate files, compress some of them with @code{myisampack}, and then create a @code{MERGE} to use these as one. @item Give you more speed. You can split a big read-only table based on some criteria and then put the different table part on different disks. A @code{MERGE} table on this could be much faster than using the big table. (You can, of course, also use a RAID to get the same kind of benefits.) @item Do more efficient searches. If you know exactly what you are looking after, you can search in just one of the split tables for some queries and use @strong{MERGE} table for others. You can even have many different @code{MERGE} tables active, with possible overlapping files. @item More efficient repairs. It's easier to repair the individual files that are mapped to a @code{MERGE} file than trying to repair a real big file. @item Instant mapping of many files as one. A @code{MERGE} table uses the index of the individual tables. It doesn't need to maintain an index of its one. This makes @code{MERGE} table collections VERY fast to make or remap. Note that you must specify the key definitions when you create a @code{MERGE} table!. @item If you have a set of tables that you join to a big table on demand or batch, you should instead create a @code{MERGE} table on them on demand. This is much faster and will save a lot of disk space. @item Go around the file size limit for the operating system. @item You can create an alias/synonym for a table by just using MERGE over one table. There shouldn't be any really notable performance impacts of doing this (only a couple of indirect calls and memcpy's for each read). @end itemize The disadvantages with @code{MERGE} tables are: @itemize @bullet @item You can't use @code{INSERT} on @code{MERGE} tables, as MySQL can't know in which of the tables we should insert the row. @item You can only use identical @code{MyISAM} tables for a @code{MERGE} table. @item @code{MERGE} tables uses more file descriptors. If you are using a @strong{MERGE} that maps over 10 tables and 10 users are using this, you are using 10*10 + 10 file descriptors. (10 data files for 10 users and 10 shared index files.) @item Key reads are slower. When you do a read on a key, the @code{MERGE} handler will need to issue a read on all underlying tables to check which one most closely matches the given key. If you then do a 'read-next' then the merge table handler will need to search the read buffers to find the next key. Only when one key buffer is used up, the handler will need to read the next key block. This makes @code{MERGE} keys much slower on @code{eq_ref} searches, but not much slower on @code{ref} searches. @xref{EXPLAIN}. @item You can't do @code{DROP TABLE}, @code{ALTER TABLE} or @code{DELETE FROM table_name} without a @code{WHERE} clause on any of the table that is mapped by a @code{MERGE} table that is 'open'. If you do this, the @code{MERGE} table may still refer to the original table and you will get unexpected results. @end itemize The following example shows you how to use @code{MERGE} tables: @example CREATE TABLE t1 (a INT AUTO_INCREMENT PRIMARY KEY, message CHAR(20)); CREATE TABLE t2 (a INT AUTO_INCREMENT PRIMARY KEY, message CHAR(20)); INSERT INTO t1 (message) VALUES ("Testing"),("table"),("t1"); INSERT INTO t2 (message) VALUES ("Testing"),("table"),("t2"); CREATE TABLE total (a INT NOT NULL, message CHAR(20), KEY(a)) TYPE=MERGE UNION=(t1,t2); @end example Note that we didn't create a @code{UNIQUE} or @code{PRIMARY KEY} in the @code{total} table as the key isn't going to be unique in the @code{total} table. Note that you can also manipulate the @code{.MRG} file directly from the outside of the MySQL server: @example shell> cd /mysql-data-directory/current-database shell> ls -1 t1.MYI t2.MYI > total.MRG shell> mysqladmin flush-tables @end example Now you can do things like: @example mysql> select * from total; +---+---------+ | a | message | +---+---------+ | 1 | Testing | | 2 | table | | 3 | t1 | | 1 | Testing | | 2 | table | | 3 | t2 | +---+---------+ @end example To remap a @code{MERGE} table you can do one of the following: @itemize @bullet @item @code{DROP} the table and re-create it @item Use @code{ALTER TABLE table_name UNION(...)} @item Change the @code{.MRG} file and issue a @code{FLUSH TABLE} on the @code{MERGE} table and all underlying tables to force the handler to read the new definition file. @end itemize @node ISAM, HEAP, MERGE, Table types @section ISAM Tables @cindex tables, ISAM You can also use the deprecated ISAM table type. This will disappear rather soon because @code{MyISAM} is a better implementation of the same thing. ISAM uses a @code{B-tree} index. The index is stored in a file with the @code{.ISM} extension, and the data is stored in a file with the @code{.ISD} extension. You can check/repair ISAM tables with the @code{isamchk} utility. @xref{Crash recovery}. @code{ISAM} has the following features/properties: @itemize @bullet @item Compressed and fixed-length keys @item Fixed and dynamic record length @item 16 keys with 16 key parts/key @item Max key length 256 (default) @item Data is stored in machine format; this is fast, but is machine/OS dependent. @end itemize Most of the things true for @code{MyISAM} tables are also true for @code{ISAM} tables. @xref{MyISAM}. The major differences compared to @code{MyISAM} tables are: @itemize @bullet @item ISAM tables are not binary portable across OS/Platforms. @item Can't handle tables > 4G. @item Only support prefix compression on strings. @item Smaller key limits. @item Dynamic tables get more fragmented. @item Tables are compressed with @code{pack_isam} rather than with @code{myisampack}. @end itemize If you want to convert an @code{ISAM} table to a @code{MyISAM} table so that you can use utilities such as @code{mysqlcheck}, use an @code{ALTER TABLE} statement: @example mysql> ALTER TABLE tbl_name TYPE = MYISAM; @end example @node HEAP, BDB, ISAM, Table types @section HEAP Tables @cindex tables, @code{HEAP} @code{HEAP} tables use a hashed index and are stored in memory. This makes them very fast, but if MySQL crashes you will lose all data stored in them. @code{HEAP} is very useful for temporary tables! The MySQL internal HEAP tables use 100% dynamic hashing without overflow areas. There is no extra space needed for free lists. @code{HEAP} tables also don't have problems with delete + inserts, which normally is common with hashed tables: @example mysql> CREATE TABLE test TYPE=HEAP SELECT ip,SUM(downloads) as down FROM log_table GROUP BY ip; mysql> SELECT COUNT(ip),AVG(down) FROM test; mysql> DROP TABLE test; @end example Here are some things you should consider when you use @code{HEAP} tables: @itemize @bullet @item You should always use specify @code{MAX_ROWS} in the @code{CREATE} statement to ensure that you accidentally do not use all memory. @item Indexes will only be used with @code{=} and @code{<=>} (but are VERY fast). @item @code{HEAP} tables can only use whole keys to search for a row; compare this to @code{MyISAM} tables where any prefix of the key can be used to find rows. @item @code{HEAP} tables use a fixed record length format. @item @code{HEAP} doesn't support @code{BLOB}/@code{TEXT} columns. @item @code{HEAP} doesn't support @code{AUTO_INCREMENT} columns. @item @code{HEAP} doesn't support an index on a @code{NULL} column. @item You can have non-unique keys in a @code{HEAP} table (this isn't common for hashed tables). @item @code{HEAP} tables are shared between all clients (just like any other table). @item You can't search for the next entry in order (that is, to use the index to do an @code{ORDER BY}). @item Data for @code{HEAP} tables are allocated in small blocks. The tables are 100% dynamic (on inserting). No overflow areas and no extra key space are needed. Deleted rows are put in a linked list and are reused when you insert new data into the table. @item You need enough extra memory for all HEAP tables that you want to use at the same time. @item To free memory, you should execute @code{DELETE FROM heap_table}, @code{TRUNCATE heap_table} or @code{DROP TABLE heap_table}. @item MySQL cannot find out approximately how many rows there are between two values (this is used by the range optimizer to decide which index to use). This may affect some queries if you change a @code{MyISAM} table to a @code{HEAP} table. @item To ensure that you accidentally don't do anything foolish, you can't create @code{HEAP} tables bigger than @code{max_heap_table_size}. @end itemize The memory needed for one row in a @code{HEAP} table is: @example SUM_OVER_ALL_KEYS(max_length_of_key + sizeof(char*) * 2) + ALIGN(length_of_row+1, sizeof(char*)) @end example @code{sizeof(char*)} is 4 on 32-bit machines and 8 on 64-bit machines. @node BDB, InnoDB, HEAP, Table types @section BDB or Berkeley_DB Tables @cindex tables, @code{BDB} @cindex tables, @code{Berkeley DB} @menu * BDB overview:: Overview of BDB Tables * BDB install:: Installing BDB * BDB start:: BDB startup options * BDB characteristic:: Some characteristic of @code{BDB} tables: * BDB TODO:: Some things we need to fix for BDB in the near future: * BDB portability:: Operating systems supported by @strong{BDB} * BDB errors:: Errors You May Get When Using BDB Tables @end menu @node BDB overview, BDB install, BDB, BDB @subsection Overview of BDB Tables Support for BDB tables is included in the MySQL source distribution starting from Version 3.23.34 and is activated in the MySQL-Max binary. BerkeleyDB, available at @uref{http://www.sleepycat.com/} has provided MySQL with a transactional table handler. By using BerkeleyDB tables, your tables may have a greater chance of surviving crashes, and also provides @code{COMMIT} and @code{ROLLBACK} on transactions. The MySQL source distribution comes with a BDB distribution that has a couple of small patches to make it work more smoothly with MySQL. You can't use a non-patched @code{BDB} version with MySQL. We at MySQL AB are working in close cooperation with Sleepycat to keep the quality of the MySQL/BDB interface high. When it comes to supporting BDB tables, we are committed to help our users to locate the problem and help creating a reproducable test case for any problems involving BDB tables. Any such test case will be forwarded to Sleepycat who in turn will help us find and fix the problem. As this is a two stage operation, any problems with BDB tables may take a little longer for us to fix than for other table handlers. However, as the BerkeleyDB code itself has been used by many other applications than MySQL, we don't envision any big problems with this. @xref{Table handler support}. @node BDB install, BDB start, BDB overview, BDB @subsection Installing BDB If you have downloaded a binary version of MySQL that includes support for BerkeleyDB, simply follow the instructions for installing a binary version of MySQL. @xref{Installing binary}. @xref{mysqld-max, , @code{mysqld-max}}. To compile MySQL with Berkeley DB support, download MySQL Version 3.23.34 or newer and configure @code{MySQL} with the @code{--with-berkeley-db} option. @xref{Installing source}. @example cd /path/to/source/of/mysql-3.23.34 ./configure --with-berkeley-db @end example Please refer to the manual provided with the @code{BDB} distribution for more updated information. Even though Berkeley DB is in itself very tested and reliable, the MySQL interface is still considered beta quality. We are actively improving and optimizing it to get it stable very soon. @node BDB start, BDB characteristic, BDB install, BDB @subsection BDB startup options If you are running with @code{AUTOCOMMIT=0} then your changes in @code{BDB} tables will not be updated until you execute @code{COMMIT}. Instead of commit you can execute @code{ROLLBACK} to forget your changes. @xref{COMMIT}. If you are running with @code{AUTOCOMMIT=1} (the default), your changes will be committed immediately. You can start an extended transaction with the @code{BEGIN WORK} SQL command, after which your changes will not be committed until you execute @code{COMMIT} (or decide to @code{ROLLBACK} the changes). The following options to @code{mysqld} can be used to change the behavior of BDB tables: @multitable @columnfractions .30 .70 @item @strong{Option} @tab @strong{Meaning} @item @code{--bdb-home=directory} @tab Base directory for BDB tables. This should be the same directory you use for --datadir. @item @code{--bdb-lock-detect=#} @tab Berkeley lock detect. One of (DEFAULT, OLDEST, RANDOM, or YOUNGEST). @item @code{--bdb-logdir=directory} @tab Berkeley DB log file directory. @item @code{--bdb-no-sync} @tab Don't synchronously flush logs. @item @code{--bdb-no-recover} @tab Don't start Berkeley DB in recover mode. @item @code{--bdb-shared-data} @tab Start Berkeley DB in multi-process mode (Don't use @code{DB_PRIVATE} when initializing Berkeley DB) @item @code{--bdb-tmpdir=directory} @tab Berkeley DB tempfile name. @item @code{--skip-bdb} @tab Don't use berkeley db. @item @code{-O bdb_max_lock=1000} @tab Set the maximum number of locks possible. @xref{SHOW VARIABLES}. @end multitable If you use @code{--skip-bdb}, MySQL will not initialize the Berkeley DB library and this will save a lot of memory. Of course, you cannot use @code{BDB} tables if you are using this option. Normally you should start @code{mysqld} without @code{--bdb-no-recover} if you intend to use BDB tables. This may, however, give you problems when you try to start @code{mysqld} if the BDB log files are corrupted. @xref{Starting server}. With @code{bdb_max_lock} you can specify the maximum number of locks (10000 by default) you can have active on a BDB table. You should increase this if you get errors of type @code{bdb: Lock table is out of available locks} or @code{Got error 12 from ...} when you have do long transactions or when @code{mysqld} has to examine a lot of rows to calculate the query. You may also want to change @code{binlog_cache_size} and @code{max_binlog_cache_size} if you are using big multi-line transactions. @xref{COMMIT}. @node BDB characteristic, BDB TODO, BDB start, BDB @subsection Some characteristic of @code{BDB} tables: @itemize @bullet @item To be able to rollback transactions BDB maintain log files. For maximum performance you should place these on another disk than your databases by using the @code{--bdb_log_dir} options. @item MySQL performs a checkpoint each time a new BDB log file is started, and removes any log files that are not needed for current transactions. One can also run @code{FLUSH LOGS} at any time to checkpoint the Berkeley DB tables. For disaster recovery, one should use table backups plus MySQL's binary log. @xref{Backup}. @strong{Warning}: If you delete old log files that are in use, BDB will not be able to do recovery at all and you may lose data if something goes wrong. @item MySQL requires a @code{PRIMARY KEY} in each BDB table to be able to refer to previously read rows. If you don't create one, MySQL will create an maintain a hidden @code{PRIMARY KEY} for you. The hidden key has a length of 5 bytes and is incremented for each insert attempt. @item If all columns you access in a @code{BDB} table are part of the same index or part of the primary key, then MySQL can execute the query without having to access the actual row. In a @code{MyISAM} table the above holds only if the columns are part of the same index. @item The @code{PRIMARY KEY} will be faster than any other key, as the @code{PRIMARY KEY} is stored together with the row data. As the other keys are stored as the key data + the @code{PRIMARY KEY}, it's important to keep the @code{PRIMARY KEY} as short as possible to save disk and get better speed. @item @code{LOCK TABLES} works on @code{BDB} tables as with other tables. If you don't use @code{LOCK TABLE}, MYSQL will issue an internal multiple-write lock on the table to ensure that the table will be properly locked if another thread issues a table lock. @item Internal locking in @code{BDB} tables is done on page level. @item @code{SELECT COUNT(*) FROM table_name} is slow as @code{BDB} tables doesn't maintain a count of the number of rows in the table. @item Scanning is slower than with @code{MyISAM} tables as one has data in BDB tables stored in B-trees and not in a separate data file. @item The application must always be prepared to handle cases where any change of a @code{BDB} table may make an automatic rollback and any read may fail with a deadlock error. @item Keys are not compressed to previous keys as with ISAM or MyISAM tables. In other words, the key information will take a little more space in @code{BDB} tables compared to MyISAM tables which don't use @code{PACK_KEYS=0}. @item There is often holes in the BDB table to allow you to insert new rows in the middle of the key tree. This makes BDB tables somewhat larger than MyISAM tables. @item The optimizer needs to know an approximation of the number of rows in the table. MySQL solves this by counting inserts and maintaining this in a separate segment in each BDB table. If you don't do a lot of @code{DELETE} or @code{ROLLBACK}:s this number should be accurate enough for the MySQL optimizer, but as MySQL only store the number on close, it may be wrong if MySQL dies unexpectedly. It should not be fatal even if this number is not 100 % correct. One can update the number of rows by executing @code{ANALYZE TABLE} or @code{OPTIMIZE TABLE}. @xref{ANALYZE TABLE} . @xref{OPTIMIZE TABLE}. @item If you get full disk with a @code{BDB} table, you will get an error (probably error 28) and the transaction should roll back. This is in contrast with @code{MyISAM} and @code{ISAM} tables where @code{mysqld} will wait for enough free disk before continuing. @end itemize @node BDB TODO, BDB portability, BDB characteristic, BDB @subsection Some things we need to fix for BDB in the near future: @itemize @bullet @item It's very slow to open many BDB tables at the same time. If you are going to use BDB tables, you should not have a very big table cache (> 256 ?) and you should use @code{--no-auto-rehash} with the @code{mysql} client. We plan to partly fix this in 4.0. @item @code{SHOW TABLE STATUS} doesn't yet provide that much information for BDB tables. @item Optimize performance. @item Change to not use page locks at all when we are scanning tables. @end itemize @node BDB portability, BDB errors, BDB TODO, BDB @subsection Operating systems supported by @strong{BDB} If you after having built MySQL with support for BDB tables get the following error in the log file when you start @code{mysqld}: @example bdb: architecture lacks fast mutexes: applications cannot be threaded Can't init dtabases @end example This means that @code{BDB} tables are not supported for your architecture. In this case you have to rebuild MySQL without BDB table support. NOTE: The following list is not complete; We will update this as we get more information about this. Currently we know that BDB tables works with the following operating system. @itemize @bullet @item Linux 2.x intel @item Solaris sparc @item SCO OpenServer @item SCO UnixWare 7.0.1 @end itemize It doesn't work with the following operating systems: @itemize @bullet @item Linux 2.x Alpha @item Max OS X @end itemize @node BDB errors, , BDB portability, BDB @subsection Errors You May Get When Using BDB Tables @itemize @bullet @item If you get the following error in the @code{hostname.err log} when starting @code{mysqld}: @example bdb: Ignoring log file: .../log.XXXXXXXXXX: unsupported log version # @end example it means that the new @code{BDB} version doesn't support the old log file format. In this case you have to delete all @code{BDB} log BDB from your database directory (the files that has the format @code{log.XXXXXXXXXX} ) and restart @code{mysqld}. We would also recommend you to do a @code{mysqldump --opt} of your old @code{BDB} tables, delete the old table and restore the dump. @item If you are running in not @code{auto_commit} mode and delete a table you are using by another thread you may get the following error messages in the MySQL error file: @example 001119 23:43:56 bdb: Missing log fileid entry 001119 23:43:56 bdb: txn_abort: Log undo failed for LSN: 1 3644744: Invalid @end example This is not fatal but we don't recommend that you delete tables if you are not in @code{auto_commit} mode, until this problem is fixed (the fix is not trivial). @end itemize @node InnoDB, , BDB, Table types @section InnoDB Tables @menu * InnoDB overview:: InnoDB tables overview * InnoDB start:: InnoDB startup options * InnoDB init:: Creating InnoDB table space. * Using InnoDB tables:: Creating InnoDB tables * Adding and removing:: Adding and removing InnoDB data and log files * Backing up:: Backing up and recovering an InnoDB database * Moving:: Moving an InnoDB database to another machine * InnoDB transaction model:: InnoDB transaction model. * Implementation:: Implementation of multiversioning * Table and index:: Table and index structures * File space management:: File space management and disk i/o * Error handling:: Error handling * InnoDB restrictions:: Some restrictions on InnoDB tables * InnoDB contact information:: InnoDB contact information. @end menu @node InnoDB overview, InnoDB start, InnoDB, InnoDB @subsection InnoDB tables overview InnoDB tables are included in the MySQL source distribution starting from 3.23.34a and are activated in the @strong{MySQL -max} binary. If you have downloaded a binary version of MySQL that includes support for InnoDB (mysqld-max), simply follow the instructions for installing a binary version of MySQL. @xref{Installing binary}. @xref{mysqld-max, , @code{mysqld-max}}. To compile MySQL with InnoDB support, download MySQL-3.23.37 or newer and configure MySQL with the @code{--with-innodb} option. @xref{Installing source}. @example cd /path/to/source/of/mysql-3.23.37 ./configure --with-innodb @end example To get InnoDB to work you have to specify where the data for InnoDB tables should be stored by specifying the @code{innodb_data_file_path} option on the command line or in an MySQL option file. @xref{InnoDB start}. If you have configured MySQL for InnoDB but you have not specified the above option, @code{mysqld} will print at start: @example Can't initialize InnoDB as 'innodb_data_file_path' is not set @end example InnoDB provides MySQL with a transaction-safe table handler with commit, rollback, and crash recovery capabilities. InnoDB does locking on row level, and also provides an Oracle-style consistent non-locking read in @code{SELECTS}, which increases transaction concurrency. There is not need for lock escalation in InnoDB, because row level locks in InnoDB fit in very small space. InnoDB has been designed for maximum performance when processing large data volumes. Its CPU efficiency is probably not matched by any other disk-based relational database engine. You can find the latest information about InnoDB at @uref{http://www.innodb.com}. The most up-to-date version of the InnoDB manual is always placed there, and you can also order commercial support for InnoDB. Technically, InnoDB is a database backend placed under MySQL. InnoDB has its own buffer pool for caching data and indexes in main memory. InnoDB stores its tables and indexes in a tablespace, which may consist of several files. This is different from, for example, @code{MyISAM} tables where each table is stored as a separate file. InnoDB is distributed under the GNU GPL License Version 2 (of June 1991). In the source distribution of MySQL, InnoDB appears as a subdirectory. @node InnoDB start, InnoDB init, InnoDB overview, InnoDB @subsection InnoDB startup options Beginning from MySQL-3.23.37 the prefix of the options is changed from @code{innobase_...} to @code{innodb_...}. To use InnoDB tables you @strong{MUST} specify configuration parameters in the MySQL configuration file in the @code{[mysqld]} section of the configuration file @file{my.cnf}. @xref{Option files}. The only required parameter to use InnoDB is @code{innodb_data_file_path}, but you should set others if you want to get a better performance. Suppose you have a Windows NT machine with 128 MB RAM and a single 10 GB hard disk. Below is an example of possible configuration parameters in @file{my.cnf} for InnoDB: @example innodb_data_file_path = ibdata1:2000M;ibdata2:2000M innodb_data_home_dir = c:\ibdata set-variable = innodb_mirrored_log_groups=1 innodb_log_group_home_dir = c:\iblogs set-variable = innodb_log_files_in_group=3 set-variable = innodb_log_file_size=30M set-variable = innodb_log_buffer_size=8M innodb_flush_log_at_trx_commit=1 innodb_log_arch_dir = c:\iblogs innodb_log_archive=0 set-variable = innodb_buffer_pool_size=80M set-variable = innodb_additional_mem_pool_size=10M set-variable = innodb_file_io_threads=4 set-variable = innodb_lock_wait_timeout=50 @end example Note that data files must be < 4G, and < 2G on some file systems! The total size of data files has to be >= 10 MB. InnoDB does not create directories: you have to create them yourself. Suppose you have a Linux machine with 512 MB RAM and three 20 GB hard disks (at directory paths @file{/}, @file{/dr2} and @file{/dr3}). Below is an example of possible configuration parameters in @file{my.cnf} for InnoDB: @example innodb_data_file_path = ibdata/ibdata1:2000M;dr2/ibdata/ibdata2:2000M innodb_data_home_dir = / set-variable = innodb_mirrored_log_groups=1 innodb_log_group_home_dir = /dr3 set-variable = innodb_log_files_in_group=3 set-variable = innodb_log_file_size=50M set-variable = innodb_log_buffer_size=8M innodb_flush_log_at_trx_commit=1 innodb_log_arch_dir = /dr3/iblogs innodb_log_archive=0 set-variable = innodb_buffer_pool_size=400M set-variable = innodb_additional_mem_pool_size=20M set-variable = innodb_file_io_threads=4 set-variable = innodb_lock_wait_timeout=50 @end example Note that we have placed the two data files on different disks. The reason for the name @code{innodb_data_file_path} is that you can also specify paths to your data files, and @code{innodb_data_home_dir} is just textually catenated before your data file paths, adding a possible slash or backslash in between. InnoDB will fill the tablespace formed by the data files from bottom up. In some cases it will improve the performance of the database if all data is not placed on the same physical disk. Putting log files on a different disk from data is very often beneficial for performance. The meanings of the configuration parameters are the following: @multitable @columnfractions .30 .70 @item @code{innodb_data_home_dir} @tab The common part of the directory path for all InnoDB data files. @item @code{innodb_data_file_path} @tab Paths to individual data files and their sizes. The full directory path to each data file is acquired by concatenating innodb_data_home_dir to the paths specified here. The file sizes are specified in megabytes, hence the 'M' after the size specification above. Do not set a file size bigger than 4000M, and on most operating systems not bigger than 2000M. InnoDB also understands the abbreviation 'G', 1G meaning 1024M. The sum of the sizes of the files must be at least 10 MB. @item @code{innodb_mirrored_log_groups} @tab Number of identical copies of log groups we keep for the database. Currently this should be set to 1. @item @code{innodb_log_group_home_dir} @tab Directory path to InnoDB log files. @item @code{innodb_log_files_in_group} @tab Number of log files in the log group. InnoDB writes to the files in a circular fashion. Value 3 is recommended here. @item @code{innodb_log_file_size} @tab Size of each log file in a log group in megabytes. Sensible values range from 1M to the size of the buffer pool specified below. The bigger the value, the less checkpoint flush activity is needed in the buffer pool, saving disk i/o. But bigger log files also mean that recovery will be slower in case of a crash. File size restriction as for a data file. @item @code{innodb_log_buffer_size} @tab The size of the buffer which InnoDB uses to write log to the log files on disk. Sensible values range from 1M to half the combined size of log files. A big log buffer allows large transactions to run without a need to write the log to disk until the transaction commit. Thus, if you have big transactions, making the log buffer big will save disk i/o. @item @code{innodb_flush_log_at_trx_commit} @tab Normally this is set to 1, meaning that at a transaction commit the log is flushed to disk, and the modifications made by the transaction become permanent, and survive a database crash. If you are willing to compromise this safety, and you are running small transactions, you may set this to 0 to reduce disk i/o to the logs. @item @code{innodb_log_arch_dir} @tab The directory where fully written log files would be archived if we used log archiving. The value of this parameter should currently be set the same as @code{innodb_log_group_home_dir}. @item @code{innodb_log_archive} @tab This value should currently be set to 0. As recovery from a backup is done by MySQL using its own log files, there is currently no need to archive InnoDB log files. @item @code{innodb_buffer_pool_size} @tab The size of the memory buffer InnoDB uses to cache data and indexes of its tables. The bigger you set this the less disk i/o is needed to access data in tables. On a dedicated database server you may set this parameter up to 90 % of the machine physical memory size. Do not set it too large, though, because competition of the physical memory may cause paging in the operating system. @item @code{innodb_additional_mem_pool_size} @tab Size of a memory pool InnoDB uses to store data dictionary information and other internal data structures. A sensible value for this might be 2M, but the more tables you have in your application the more you will need to allocate here. If InnoDB runs out of memory in this pool, it will start to allocate memory from the operating system, and write warning messages to the MySQL error log. @item @code{innodb_file_io_threads} @tab Number of file i/o threads in InnoDB. Normally, this should be 4, but on Windows NT disk i/o may benefit from a larger number. @item @code{innodb_lock_wait_timeout} @tab Timeout in seconds an InnoDB transaction may wait for a lock before being rolled back. InnoDB automatically detects transaction deadlocks in its own lock table and rolls back the transaction. If you use @code{LOCK TABLES} command, or other transaction-safe table handlers than InnoDB in the same transaction, then a deadlock may arise which InnoDB cannot notice. In cases like this the timeout is useful to resolve the situation. @item @code{innodb_flush_method} @tab (Available from 3.23.40 up.) The default value for this is @code{fdatasync}. Another option is @code{O_DSYNC}. @end multitable @node InnoDB init, Using InnoDB tables, InnoDB start, InnoDB @subsection Creating InnoDB table space Suppose you have installed MySQL and have edited @file{my.cnf} so that it contains the necessary InnoDB configuration parameters. Before starting MySQL you should check that the directories you have specified for InnoDB data files and log files exist and that you have access rights to those directories. InnoDB cannot create directories, only files. Check also you have enough disk space for the data and log files. When you now start MySQL, InnoDB will start creating your data files and log files. InnoDB will print something like the following: @example ~/mysqlm/sql > mysqld InnoDB: The first specified data file /home/heikki/data/ibdata1 did not exist: InnoDB: a new database to be created! InnoDB: Setting file /home/heikki/data/ibdata1 size to 134217728 InnoDB: Database physically writes the file full: wait... InnoDB: Data file /home/heikki/data/ibdata2 did not exist: new to be created InnoDB: Setting file /home/heikki/data/ibdata2 size to 262144000 InnoDB: Database physically writes the file full: wait... InnoDB: Log file /home/heikki/data/logs/ib_logfile0 did not exist: new to be c reated InnoDB: Setting log file /home/heikki/data/logs/ib_logfile0 size to 5242880 InnoDB: Log file /home/heikki/data/logs/ib_logfile1 did not exist: new to be c reated InnoDB: Setting log file /home/heikki/data/logs/ib_logfile1 size to 5242880 InnoDB: Log file /home/heikki/data/logs/ib_logfile2 did not exist: new to be c reated InnoDB: Setting log file /home/heikki/data/logs/ib_logfile2 size to 5242880 InnoDB: Started mysqld: ready for connections @end example A new InnoDB database has now been created. You can connect to the MySQL server with the usual MySQL client programs like @code{mysql}. When you shut down the MySQL server with @file{mysqladmin shutdown}, InnoDB output will be like the following: @example 010321 18:33:34 mysqld: Normal shutdown 010321 18:33:34 mysqld: Shutdown Complete InnoDB: Starting shutdown... InnoDB: Shutdown completed @end example You can now look at the data files and logs directories and you will see the files created. The log directory will also contain a small file named @file{ib_arch_log_0000000000}. That file resulted from the database creation, after which InnoDB switched off log archiving. When MySQL is again started, the output will be like the following: @example ~/mysqlm/sql > mysqld InnoDB: Started mysqld: ready for connections @end example @menu * Error creating InnoDB:: If something goes wrong in database creation @end menu @node Error creating InnoDB, , InnoDB init, InnoDB init @subsubsection If something goes wrong in database creation If something goes wrong in an InnoDB database creation, you should delete all files created by InnoDB. This means all data files, all log files, the small archived log file, and in the case you already did create some InnoDB tables, delete also the corresponding @file{.frm} files for these tables from the MySQL database directories. Then you can try the InnoDB database creation again. @node Using InnoDB tables, Adding and removing, InnoDB init, InnoDB @subsection Creating InnoDB tables Suppose you have started the MySQL client with the command @code{mysql test}. To create a table in the InnoDB format you must specify @code{TYPE = InnoDB} in the table creation SQL command: @example CREATE TABLE CUSTOMER (A INT, B CHAR (20), INDEX (A)) TYPE = InnoDB; @end example This SQL command will create a table and an index on column @code{A} into the InnoDB tablespace consisting of the data files you specified in @file{my.cnf}. In addition MySQL will create a file @file{CUSTOMER.frm} to the MySQL database directory @file{test}. Internally, InnoDB will add to its own data dictionary an entry for table @code{'test/CUSTOMER'}. Thus you can create a table of the same name @code{CUSTOMER} in another database of MySQL, and the table names will not collide inside InnoDB. You can query the amount of free space in the InnoDB tablespace by issuing the table status command of MySQL for any table you have created with @code{TYPE = InnoDB}. Then the amount of free space in the tablespace appears in the table comment section in the output of @code{SHOW}. An example: @example SHOW TABLE STATUS FROM test LIKE 'CUSTOMER' @end example Note that the statistics @code{SHOW} gives about InnoDB tables are only approximate: they are used in SQL optimization. Table and index reserved sizes in bytes are accurate, though. NOTE: @code{DROP DATABASE} does not currently work for InnoDB tables! You must drop the tables individually. Also take care not to delete or add @file{.frm} files to your InnoDB database manually: use @code{CREATE TABLE} and @code{DROP TABLE} commands. InnoDB has its own internal data dictionary, and you will get problems if the MySQL @file{.frm} files are out of 'sync' with the InnoDB internal data dictionary. @subsubsection Converting MyISAM tables to InnoDB InnoDB does not have a special optimization for separate index creation. Therefore it does not pay to export and import the table and create indexes afterwards. The fastest way to alter a table to InnoDB is to do the inserts directly to an InnoDB table, that is, use @code{ALTER TABLE ... TYPE=INNODB}, or create an empty InnoDB table with identical definitions and insert the rows with @code{INSERT INTO ... SELECT * FROM ...}. To get better control over the insertion process, it may be good to insert big tables in pieces: @example INSERT INTO newtable SELECT * FROM oldtable WHERE yourkey > something AND yourkey <= somethingelse; @end example After all data has been inserted you can rename the tables. During the conversion of big tables you should set the InnoDB buffer pool size big to reduce disk i/o. Not bigger than 80 % of the physical memory, though. You should set InnoDB log files big, and also the log buffer large. Make sure you do not run out of tablespace: InnoDB tables take a lot more space than MyISAM tables. If an @code{ALTER TABLE} runs out of space, it will start a rollback, and that can take hours if it is disk-bound. In inserts InnoDB uses the insert buffer to merge secondary index records to indexes in batches. That saves a lot of disk i/o. In rollback no such mechanism is used, and the rollback can take 30 times longer than the insertion. In the case of a runaway rollback, if you do not have valuable data in your database, it is better that you kill the database process and delete all InnoDB data and log files and all InnoDB table @file{.frm} files, and start your job again, rather than wait for millions of disk i/os to complete. @node Adding and removing, Backing up, Using InnoDB tables, InnoDB @subsection Adding and removing InnoDB data and log files You cannot increase the size of an InnoDB data file. To add more into your tablespace you have to add a new data file. To do this you have to shut down your MySQL database, edit the @file{my.cnf} file, adding a new file to @code{innodb_data_file_path}, and then start MySQL again. Currently you cannot remove a data file from InnoDB. To decrease the size of your database you have to use @code{mysqldump} to dump all your tables, create a new database, and import your tables to the new database. If you want to change the number or the size of your InnoDB log files, you have to shut down MySQL and make sure that it shuts down without errors. Then copy the old log files into a safe place just in case something went wrong in the shutdown and you will need them to recover the database. Delete then the old log files from the log file directory, edit @file{my.cnf}, and start MySQL again. InnoDB will tell you at the startup that it is creating new log files. @node Backing up, Moving, Adding and removing, InnoDB @subsection Backing up and recovering an InnoDB database The key to safe database management is taking regular backups. To take a 'binary' backup of your database you have to do the following: @itemize @bullet @item Shut down your MySQL database and make sure it shuts down without errors. @item Copy all your data files into a safe place. @item Copy all your InnoDB log files to a safe place. @item Copy your @file{my.cnf} configuration file(s) to a safe place. @item Copy all the @file{.frm} files for your InnoDB tables into a safe place. @end itemize There is currently no on-line or incremental backup tool available for InnoDB, though they are in the TODO list. In addition to taking the binary backups described above, you should also regularly take dumps of your tables with @file{mysqldump}. The reason to this is that a binary file may be corrupted without you noticing it. Dumped tables are stored into text files which are human-readable and much simpler than database binary files. Seeing table corruption from dumped files is easier, and since their format is simpler, the chance for serious data corruption in them is smaller. A good idea is to take the dumps at the same time you take a binary backup of your database. You have to shut out all clients from your database to get a consistent snapshot of all your tables into your dumps. Then you can take the binary backup, and you will then have a consistent snapshot of your database in two formats. To be able to recover your InnoDB database to the present from the binary backup described above, you have to run your MySQL database with the general logging and log archiving of MySQL switched on. Here by the general logging we mean the logging mechanism of the MySQL server which is independent of InnoDB logs. To recover from a crash of your MySQL server process, the only thing you have to do is to restart it. InnoDB will automatically check the logs and perform a roll-forward of the database to the present. InnoDB will automatically roll back uncommitted transactions which were present at the time of the crash. During recovery, InnoDB will print out something like the following: @example ~/mysqlm/sql > mysqld InnoDB: Database was not shut down normally. InnoDB: Starting recovery from log files... InnoDB: Starting log scan based on checkpoint at InnoDB: log sequence number 0 13674004 InnoDB: Doing recovery: scanned up to log sequence number 0 13739520 InnoDB: Doing recovery: scanned up to log sequence number 0 13805056 InnoDB: Doing recovery: scanned up to log sequence number 0 13870592 InnoDB: Doing recovery: scanned up to log sequence number 0 13936128 ... InnoDB: Doing recovery: scanned up to log sequence number 0 20555264 InnoDB: Doing recovery: scanned up to log sequence number 0 20620800 InnoDB: Doing recovery: scanned up to log sequence number 0 20664692 InnoDB: 1 uncommitted transaction(s) which must be rolled back InnoDB: Starting rollback of uncommitted transactions InnoDB: Rolling back trx no 16745 InnoDB: Rolling back of trx no 16745 completed InnoDB: Rollback of uncommitted transactions completed InnoDB: Starting an apply batch of log records to the database... InnoDB: Apply batch completed InnoDB: Started mysqld: ready for connections @end example If your database gets corrupted or your disk fails, you have to do the recovery from a backup. In the case of corruption, you should first find a backup which is not corrupted. From a backup do the recovery from the general log files of MySQL according to instructions in the MySQL manual. @menu * InnoDB checkpoints:: Checkpoints @end menu @node InnoDB checkpoints, , Backing up, Backing up @subsubsection Checkpoints InnoDB implements a checkpoint mechanism called a fuzzy checkpoint. InnoDB will flush modified database pages from the buffer pool in small batches, there is no need to flush the buffer pool in one single batch, which would in practice stop processing of user SQL statements for a while. In crash recovery InnoDB looks for a checkpoint label written to the log files. It knows that all modifications to the database before the label are already present on the disk image of the database. Then InnoDB scans the log files forward from the place of the checkpoint applying the logged modifications to the database. InnoDB writes to the log files in a circular fashion. All committed modifications which make the database pages in the buffer pool different from the images on disk must be available in the log files in case InnoDB has to do a recovery. This means that when InnoDB starts to reuse a log file in the circular fashion, it has to make sure that the database page images on disk already contain the modifications logged in the log file InnoDB is going to reuse. In other words, InnoDB has to make a checkpoint and often this involves flushing of modified database pages to disk. The above explains why making your log files very big may save disk i/o in checkpointing. It can make sense to set the total size of the log files as big as the buffer pool or even bigger. The drawback in big log files is that crash recovery can last longer because there will be more log to apply to the database. @node Moving, InnoDB transaction model, Backing up, InnoDB @subsection Moving an InnoDB database to another machine InnoDB data and log files are binary-compatible on all platforms if the floating point number format on the machines is the same. You can move an InnoDB database simply by copying all the relevant files, which we already listed in the previous section on backing up a database. If the floating point formats on the machines are different but you have not used @code{FLOAT} or @code{DOUBLE} data types in your tables then the procedure is the same: just copy the relevant files. If the formats are different and your tables contain floating point data, you have to use @file{mysqldump} and @file{mysqlimport} to move those tables. A performance tip is to switch off the auto commit when you import data into your database, assuming your tablespace has enough space for the big rollback segment the big import transaction will generate. Do the commit only after importing a whole table or a segment of a table. @node InnoDB transaction model, Implementation, Moving, InnoDB @subsection InnoDB transaction model In the InnoDB transaction model the goal has been to combine the best sides of a multiversioning database to traditional two-phase locking. InnoDB does locking on row level and runs queries by default as non-locking consistent reads, in the style of Oracle. The lock table in InnoDB is stored so space-efficiently that lock escalation is not needed: typically several users are allowed to lock every row in the database, or any random subset of the rows, without InnoDB running out of memory. In InnoDB all user activity happens inside transactions. If the auto commit mode is used in MySQL, then each SQL statement will form a single transaction. If the auto commit mode is switched off, then we can think that a user always has a transaction open. If he issues the SQL @code{COMMIT} or @code{ROLLBACK} statement, that ends the current transaction, and a new starts. Both statements will release all InnoDB locks that were set during the current transaction. A @code{COMMIT} means that the changes made in the current transaction are made permanent and become visible to other users. A @code{ROLLBACK} on the other hand cancels all modifications made by the current transaction. @menu * InnoDB consistent read:: Consistent read * InnoDB locking reads:: Locking reads * InnoDB Next-key locking:: Next-key locking: avoiding the phantom problem * InnoDB Locks set:: Locks set by different SQL statements in InnoDB * InnoDB Deadlock detection:: Deadlock detection and rollback * InnoDB Consistent read example:: An example of how the consistent read works in InnoDB @end menu @node InnoDB consistent read, InnoDB locking reads, InnoDB transaction model, InnoDB transaction model @subsubsection Consistent read A consistent read means that InnoDB uses its multiversioning to present to a query a snapshot of the database at a point in time. The query will see the changes made by exactly those transactions that committed before that point of time, and no changes made by later or uncommitted transactions. The exception to this rule is that the query will see the changes made by the transaction itself which issues the query. When a transaction issues its first consistent read, InnoDB assigns the snapshot, or the point of time, which all consistent reads in the same transaction will use. In the snapshot are all transactions that committed before assigning the snapshot. Thus the consistent reads within the same transaction will also be consistent with respect to each other. You can get a fresher snapshot for your queries by committing the current transaction and after that issuing new queries. Consistent read is the default mode in which InnoDB processes @code{SELECT} statements. A consistent read does not set any locks on the tables it accesses, and therefore other users are free to modify those tables at the same time a consistent read is being performed on the table. @node InnoDB locking reads, InnoDB Next-key locking, InnoDB consistent read, InnoDB transaction model @subsubsection Locking reads A consistent read is not convenient in some circumstances. Suppose you want to add a new row into your table @code{CHILD}, and make sure that the child already has a parent in table @code{PARENT}. Suppose you use a consistent read to read the table @code{PARENT} and indeed see the parent of the child in the table. Can you now safely add the child row to table @code{CHILD}? No, because it may happen that meanwhile some other user has deleted the parent row from the table @code{PARENT}, and you are not aware of that. The solution is to perform the @code{SELECT} in a locking mode, @code{LOCK IN SHARE MODE}. @example SELECT * FROM PARENT WHERE NAME = 'Jones' LOCK IN SHARE MODE; @end example Performing a read in share mode means that we read the latest available data, and set a shared mode lock on the rows we read. If the latest data belongs to a yet uncommitted transaction of another user, we will wait until that transaction commits. A shared mode lock prevents others from updating or deleting the row we have read. After we see that the above query returns the parent @code{'Jones'}, we can safely add his child to table @code{CHILD}, and commit our transaction. This example shows how to implement referential integrity in your application code. Let us look at another example: we have an integer counter field in a table @code{CHILD_CODES} which we use to assign a unique identifier to each child we add to table @code{CHILD}. Obviously, using a consistent read or a shared mode read to read the present value of the counter is not a good idea, since then two users of the database may see the same value for the counter, and we will get a duplicate key error when we add the two children with the same identifier to the table. In this case there are two good ways to implement the reading and incrementing of the counter: (1) update the counter first by incrementing it by 1 and only after that read it, or (2) read the counter first with a lock mode @code{FOR UPDATE}, and increment after that: @example SELECT COUNTER_FIELD FROM CHILD_CODES FOR UPDATE; UPDATE CHILD_CODES SET COUNTER_FIELD = COUNTER_FIELD + 1; @end example A @code{SELECT ... FOR UPDATE} will read the latest available data setting exclusive locks on each row it reads. Thus it sets the same locks a searched SQL @code{UPDATE} would set on the rows. @node InnoDB Next-key locking, InnoDB Locks set, InnoDB locking reads, InnoDB transaction model @subsubsection Next-key locking: avoiding the phantom problem In row level locking InnoDB uses an algorithm called next-key locking. InnoDB does the row level locking so that when it searches or scans an index of a table, it sets shared or exclusive locks on the index records in encounters. Thus the row level locks are more precisely called index record locks. The locks InnoDB sets on index records also affect the 'gap' before that index record. If a user has a shared or exclusive lock on record R in an index, then another user cannot insert a new index record immediately before R in the index order. This locking of gaps is done to prevent the so-called phantom problem. Suppose I want to read and lock all children with identifier bigger than 100 from table @code{CHILD}, and update some field in the selected rows. @example SELECT * FROM CHILD WHERE ID > 100 FOR UPDATE; @end example Suppose there is an index on table @code{CHILD} on column @code{ID}. Our query will scan that index starting from the first record where @code{ID} is bigger than 100. Now, if the locks set on the index records would not lock out inserts made in the gaps, a new child might meanwhile be inserted to the table. If now I in my transaction execute @example SELECT * FROM CHILD WHERE ID > 100 FOR UPDATE; @end example again, I will see a new child in the result set the query returns. This is against the isolation principle of transactions: a transaction should be able to run so that the data it has read does not change during the transaction. If we regard a set of rows as a data item, then the new 'phantom' child would break this isolation principle. When InnoDB scans an index it can also lock the gap after the last record in the index. Just that happens in the previous example: the locks set by InnoDB will prevent any insert to the table where @code{ID} would be bigger than 100. You can use the next-key locking to implement a uniqueness check in your application: if you read your data in share mode and do not see a duplicate for a row you are going to insert, then you can safely insert your row and know that the next-key lock set on the successor of your row during the read will prevent anyone meanwhile inserting a duplicate for your row. Thus the next-key locking allows you to 'lock' the non-existence of something in your table. @node InnoDB Locks set, InnoDB Deadlock detection, InnoDB Next-key locking, InnoDB transaction model @subsubsection Locks set by different SQL statements in InnoDB @itemize @bullet @item @code{SELECT ... FROM ...} : this is a consistent read, reading a snapshot of the database and setting no locks. @item @code{SELECT ... FROM ... LOCK IN SHARE MODE} : sets shared next-key locks on all index records the read encounters. @item @code{SELECT ... FROM ... FOR UPDATE} : sets exclusive next-key locks on all index records the read encounters. @item @code{INSERT INTO ... VALUES (...)} : sets an exclusive lock on the inserted row; note that this lock is not a next-key lock and does not prevent other users from inserting to the gap before the inserted row. If a duplicate key error occurs, sets a shared lock on the duplicate index record. @item @code{INSERT INTO T SELECT ... FROM S WHERE ...} sets an exclusive (non-next-key) lock on each row inserted into @code{T}. Does the search on @code{S} as a consistent read, but sets shared next-key locks on @code{S} if the MySQL logging is on. InnoDB has to set locks in the latter case because in roll-forward recovery from a backup every SQL statement has to be executed in exactly the same way as it was done originally. @item @code{CREATE TABLE ... SELECT ...} performs the @code{SELECT} as a consistent read or with shared locks, like in the previous item. @item @code{REPLACE} is done like an insert if there is no collision on a unique key. Otherwise, an exclusive next-key lock is placed on the row which has to be updated. @item @code{UPDATE ... SET ... WHERE ...} : sets an exclusive next-key lock on every record the search encounters. @item @code{DELETE FROM ... WHERE ...} : sets an exclusive next-key lock on every record the search encounters. @item @code{LOCK TABLES ... } : sets table locks. In the implementation the MySQL layer of code sets these locks. The automatic deadlock detection of InnoDB cannot detect deadlocks where such table locks are involved: see the next section below. See also section 13 'InnoDB restrictions' about the following: since MySQL does know about row level locks, it is possible that you get a table lock on a table where another user currently has row level locks. But that does not put transaction integerity into danger. @end itemize @node InnoDB Deadlock detection, InnoDB Consistent read example, InnoDB Locks set, InnoDB transaction model @subsubsection Deadlock detection and rollback InnoDB automatically detects a deadlock of transactions and rolls back the transaction whose lock request was the last one to build a deadlock, that is, a cycle in the waits-for graph of transactions. InnoDB cannot detect deadlocks where a lock set by a MySQL @code{LOCK TABLES} statement is involved, or if a lock set in another table handler than InnoDB is involved. You have to resolve these situations using @code{innodb_lock_wait_timeout} set in @file{my.cnf}. When InnoDB performs a complete rollback of a transaction, all the locks of the transaction are released. However, if just a single SQL statement is rolled back as a result of an error, some of the locks set by the SQL statement may be preserved. This is because InnoDB stores row locks in a format where it cannot afterwards know which was set by which SQL statement. @node InnoDB Consistent read example, , InnoDB Deadlock detection, InnoDB transaction model @subsubsection An example of how the consistent read works in InnoDB When you issue a consistent read, that is, an ordinary @code{SELECT} statement, InnoDB will give your transaction a timepoint according to which your query sees the database. Thus, if transaction B deletes a row and commits after your timepoint was assigned, then you will not see the row deleted. Similarly with inserts and updates. You can advance your timepoint by committing your transaction and then doing another @code{SELECT}. This is called multiversioned concurrency control. @example User A User B set autocommit=0; set autocommit=0; time | SELECT * FROM t; | empty set | INSERT INTO t VALUES (1, 2); | v SELECT * FROM t; empty set COMMIT; SELECT * FROM t; empty set; COMMIT; SELECT * FROM t; ---------------------- | 1 | 2 | ---------------------- @end example Thus user A sees the row inserted by B only when B has committed the insert, and A has committed his own transaction so that the timepoint is advanced past the the commit of B. If you want to see the 'freshest' state of the database, you should use a locking read: @example SELECT * FROM t LOCK IN SHARE MODE; @end example @subsection Performance tuning tips @strong{1.} If the Unix @file{top} or the Windows @file{Task Manager} shows that the CPU usage percentage with your workload is less than 70 %, your workload is probably disk-bound. Maybe you are making too many transaction commits, or the buffer pool is too small. Making the buffer pool bigger can help, but do not set it bigger than 80 % of physical memory. @strong{2.} Wrap several modifications into one transaction. InnoDB must flush the log to disk at each transaction commit, if that transaction made modifications to the database. Since the rotation speed of a disk is typically at most 167 revolutions/second, that constrains the number of commits to the same 167/second if the disk does not fool the operating system. @strong{3.} If you can afford the loss of some latest committed transactions, you can set the @file{my.cnf} parameter @code{innodb_flush_log_at_trx_commit} to zero. InnoDB tries to flush the log anyway once in a second, though the flush is not guaranteed. @strong{4.} Make your log files big, even as big as the buffer pool. When InnoDB has written the log files full, it has to write the modified contents of the buffer pool to disk in a checkpoint. Small log files will cause many unnecessary disk writes. The drawback in big log files is that recovery time will be longer. @strong{5.} Also the log buffer should be quite big, say 8 MB. @strong{6.} (Relevant from 3.23.39 up.) In some versions of Linux and Unix, flushing files to disk with the Unix @code{fdatasync} and other similar methods is surprisingly slow. The default method InnoDB uses is the @code{fdatasync} function. If you are not satisfied with the database write performance, you may try setting @code{innodb_flush_method} in @file{my.cnf} to @code{O_DSYNC}, though O_DSYNC seems to be slower on most systems. @strong{7.} In importing data to InnoDB, make sure that MySQL does not have @code{autocommit=1} on. Then every insert requires a log flush to disk. Put before your plain SQL import file line @example set autocommit=0; @end example and after it @example commit; @end example If you use the @file{mysqldump} option @code{--opt}, you will get dump files which are fast to import also to an InnoDB table, even without wrapping them to the above @code{set autocommit=0; ... commit;} wrappers. @strong{8.} Beware of big rollbacks of mass inserts: InnoDB uses the insert buffer to save disk i/o in inserts, but in a corresponding rollback no such mechanism is used. A disk-bound rollback can take 30 times the time of the corresponding insert. Killing the database process will not help because the rollback will start again at the database startup. The only way to get rid of a runaway rollback is to increase the buffer pool so that the rollback becomes CPU-bound and runs fast, or delete the whole InnoDB database. @strong{9.} Beware also of other big disk-bound operations. Use @code{DROP TABLE} or @code{TRUNCATE} (from MySQL-4.0 up) to empty a table, not @code{DELETE FROM yourtable}. @strong{10.} Use the multi-line @code{INSERT} to reduce communication overhead between the client and the server if you need to insert many rows: @example INSERT INTO yourtable VALUES (1, 2), (5, 5); @end example This tip is of course valid for inserts into any table type, not just InnoDB. @subsubsection The InnoDB Monitor Starting from version 3.23.41 InnoDB includes the InnoDB Monitor which prints information on the InnoDB internal state. When swithed on, InnoDB Monitor will make the MySQL server to print data to the standard output about once every 10 seconds. This data is useful in performance tuning. The printed information includes data on: @itemize @bullet @item table and record locks held by each active transaction, @item lock waits of a transactions, @item semaphore waits of threads, @item pending file i/o requests, @item buffer pool statistics, and @item purge and insert buffer merge activity of the main thread of InnoDB. @end itemize You can start InnoDB Monitor through the following SQL command: @example CREATE TABLE innodb_monitor(a int) type = innodb; @end example and stop it by @example DROP TABLE innodb_monitor; @end example The @code{CREATE TABLE} syntax is just a way to pass a command to the InnoDB engine through the MySQL SQL parser: the created table is not relevant at all for InnoDB Monitor. If you shut down the database when the monitor is running, and you want to start the monitor again, you have to drop the table before you can issue a new @code{CREATE TABLE} to start the monitor. This syntax may change in a future release. A sample output of the InnoDB Monitor: @example ================================ 010809 18:45:06 INNODB MONITOR OUTPUT ================================ -------------------------- LOCKS HELD BY TRANSACTIONS -------------------------- LOCK INFO: Number of locks in the record hash table 1294 LOCKS FOR TRANSACTION ID 0 579342744 TABLE LOCK table test/mytable trx id 0 582333343 lock_mode IX RECORD LOCKS space id 0 page no 12758 n bits 104 table test/mytable index PRIMARY trx id 0 582333343 lock_mode X Record lock, heap no 2 PHYSICAL RECORD: n_fields 74; 1-byte offs FALSE; info bits 0 0: len 4; hex 0001a801; asc ;; 1: len 6; hex 000022b5b39f; asc ";; 2: len 7; hex 000002001e03ec; asc ;; 3: len 4; hex 00000001; ... ----------------------------------------------- CURRENT SEMAPHORES RESERVED AND SEMAPHORE WAITS ----------------------------------------------- SYNC INFO: Sorry, cannot give mutex list info in non-debug version! Sorry, cannot give rw-lock list info in non-debug version! ----------------------------------------------------- SYNC ARRAY INFO: reservation count 6041054, signal count 2913432 4a239430 waited for by thread 49627477 op. S-LOCK file NOT KNOWN line 0 Mut ex 0 sp 5530989 r 62038708 sys 2155035; rws 0 8257574 8025336; rwx 0 1121090 1848344 ----------------------------------------------------- CURRENT PENDING FILE I/O'S -------------------------- Pending normal aio reads: Reserved slot, messages 40157658 4a4a40b8 Reserved slot, messages 40157658 4a477e28 ... Reserved slot, messages 40157658 4a4424a8 Reserved slot, messages 40157658 4a39ea38 Total of 36 reserved aio slots Pending aio writes: Total of 0 reserved aio slots Pending insert buffer aio reads: Total of 0 reserved aio slots Pending log writes or reads: Reserved slot, messages 40158c98 40157f98 Total of 1 reserved aio slots Pending synchronous reads or writes: Total of 0 reserved aio slots ----------- BUFFER POOL ----------- LRU list length 8034 Free list length 0 Flush list length 999 Buffer pool size in pages 8192 Pending reads 39 Pending writes: LRU 0, flush list 0, single page 0 Pages read 31383918, created 51310, written 2985115 ---------------------------- END OF INNODB MONITOR OUTPUT ============================ 010809 18:45:22 InnoDB starts purge 010809 18:45:22 InnoDB purged 0 pages @end example Some notes on the output: @itemize @bullet @item If the section LOCKS HELD BY TRANSACTIONS reports lock waits, then your application may have lock contention. The output can also help to trace reasons for transaction deadlocks. @item Section SYNC INFO will report reserved semaphores if you compile InnoDB with <code>UNIV_SYNC_DEBUG</code> defined in <tt>univ.i</tt>. @item Section SYNC ARRAY INFO reports threads waiting for a semaphore and statistics on how many times threads have needed a spin or a wait on a mutex or a rw-lock semaphore. A big number of threads waiting for semaphores may be a result of disk i/o, or contention problems inside InnoDB. Contention can be due to heavy parallelism of queries, or problems in operating system thread scheduling. @item Section CURRENT PENDING FILE I/O'S lists pending file i/o requests. A large number of these indicates that the workload is disk i/o -bound. @item Section BUFFER POOL gives you statistics on pages read and written. You can calculate from these numbers how many data file i/o's your queries are currently doing. @end itemize @node Implementation, Table and index, InnoDB transaction model, InnoDB @subsection Implementation of multiversioning Since InnoDB is a multiversioned database, it must keep information of old versions of rows in the tablespace. This information is stored in a data structure we call a rollback segment after an analogous data structure in Oracle. InnoDB internally adds two fields to each row stored in the database. A 6-byte field tells the transaction identifier for the last transaction which inserted or updated the row. Also a deletion is internally treated as an update where a special bit in the row is set to mark it as deleted. Each row also contains a 7-byte field called the roll pointer. The roll pointer points to an undo log record written to the rollback segment. If the row was updated, then the undo log record contains the information necessary to rebuild the content of the row before it was updated. InnoDB uses the information in the rollback segment to perform the undo operations needed in a transaction rollback. It also uses the information to build earlier versions of a row for a consistent read. Undo logs in the rollback segment are divided into insert and update undo logs. Insert undo logs are only needed in transaction rollback and can be discarded as soon as the transaction commits. Update undo logs are used also in consistent reads, and they can be discarded only after there is no transaction present for which InnoDB has assigned a snapshot that in a consistent read could need the information in the update undo log to build an earlier version of a database row. You must remember to commit your transactions regularly. Otherwise InnoDB cannot discard data from the update undo logs, and the rollback segment may grow too big, filling up your tablespace. The physical size of an undo log record in the rollback segment is typically smaller than the corresponding inserted or updated row. You can use this information to calculate the space need for your rollback segment. In our multiversioning scheme a row is not physically removed from the database immediately when you delete it with an SQL statement. Only when InnoDB can discard the update undo log record written for the deletion, it can also physically remove the corresponding row and its index records from the database. This removal operation is called a purge, and it is quite fast, usually taking the same order of time as the SQL statement which did the deletion. @node Table and index, File space management, Implementation, InnoDB @subsection Table and index structures Every InnoDB table has a special index called the clustered index where the data of the rows is stored. If you define a @code{PRIMARY KEY} on your table, then the index of the primary key will be the clustered index. If you do not define a primary key for your table, InnoDB will internally generate a clustered index where the rows are ordered by the row id InnoDB assigns to the rows in such a table. The row id is a 6-byte field which monotonically increases as new rows are inserted. Thus the rows ordered by the row id will be physically in the insertion order. Accessing a row through the clustered index is fast, because the row data will be on the same page where the index search leads us. In many databases the data is traditionally stored on a different page from the index record. If a table is large, the clustered index architecture often saves a disk i/o when compared to the traditional solution. The records in non-clustered indexes (we also call them secondary indexes), in InnoDB contain the primary key value for the row. InnoDB uses this primary key value to search for the row from the clustered index. Note that if the primary key is long, the secondary indexes will use more space. @menu * InnoDB physical structure:: Physical structure of an index * InnoDB Insert buffering:: Insert buffering * InnoDB Adaptive hash:: Adaptive hash indexes * InnoDB Physical record:: Physical record structure @end menu @node InnoDB physical structure, InnoDB Insert buffering, Table and index, Table and index @subsubsection Physical structure of an index All indexes in InnoDB are B-trees where the index records are stored in the leaf pages of the tree. The default size of an index page is 16 kB. When new records are inserted, InnoDB tries to leave 1 / 16 of the page free for future insertions and updates of the index records. If index records are inserted in a sequential (ascending or descending) order, the resulting index pages will be about 15/16 full. If records are inserted in a random order, then the pages will be 1/2 - 15/16 full. If the fillfactor of an index page drops below 1/2, InnoDB will try to contract the index tree to free the page. @node InnoDB Insert buffering, InnoDB Adaptive hash, InnoDB physical structure, Table and index @subsubsection Insert buffering It is a common situation in a database application that the primary key is a unique identifier and new rows are inserted in the ascending order of the primary key. Thus the insertions to the clustered index do not require random reads from a disk. On the other hand, secondary indexes are usually non-unique and insertions happen in a relatively random order into secondary indexes. This would cause a lot of random disk i/o's without a special mechanism used in InnoDB. If an index record should be inserted to a non-unique secondary index, InnoDB checks if the secondary index page is already in the buffer pool. If that is the case, InnoDB will do the insertion directly to the index page. But, if the index page is not found from the buffer pool, InnoDB inserts the record to a special insert buffer structure. The insert buffer is kept so small that it entirely fits in the buffer pool, and insertions can be made to it very fast. The insert buffer is periodically merged to the secondary index trees in the database. Often we can merge several insertions on the same page in of the index tree, and hence save disk i/o's. It has been measured that the insert buffer can speed up insertions to a table up to 15 times. @node InnoDB Adaptive hash, InnoDB Physical record, InnoDB Insert buffering, Table and index @subsubsection Adaptive hash indexes If a database fits almost entirely in main memory, then the fastest way to perform queries on it is to use hash indexes. InnoDB has an automatic mechanism which monitors index searches made to the indexes defined for a table, and if InnoDB notices that queries could benefit from building of a hash index, such an index is automatically built. But note that the hash index is always built based on an existing B-tree index on the table. InnoDB can build a hash index on a prefix of any length of the key defined for the B-tree, depending on what search pattern InnoDB observes on the B-tree index. A hash index can be partial: it is not required that the whole B-tree index is cached in the buffer pool. InnoDB will build hash indexes on demand to those pages of the index which are often accessed. In a sense, through the adaptive hash index mechanism InnoDB adapts itself to ample main memory, coming closer to the architecture of main memory databases. @node InnoDB Physical record, , InnoDB Adaptive hash, Table and index @subsubsection Physical record structure @itemize @bullet @item Each index record in InnoDB contains a header of 6 bytes. The header is used to link consecutive records together, and also in the row level locking. @item Records in the clustered index contain fields for all user-defined columns. In addition, there is a 6-byte field for the transaction id and a 7-byte field for the roll pointer. @item If the user has not defined a primary key for a table, then each clustered index record contains also a 6-byte row id field. @item Each secondary index record contains also all the fields defined for the clustered index key. @item A record contains also a pointer to each field of the record. If the total length of the fields in a record is < 128 bytes, then the pointer is 1 byte, else 2 bytes. @end itemize @subsubsection How an auto-increment column works in InnoDB After a database startup, when a user first does an insert to a table @code{T} where an auto-increment column has been defined, and the user does not provide an explicit value for the column, then InnoDB executes @code{SELECT MAX(auto-inc-column) FROM T}, and assigns that value incremented by one to the the column and the auto-increment counter of the table. We say that the auto-increment counter for table @code{T} has been initialized. InnoDB follows the same procedure in initializing the auto-increment counter for a freshly created table. Note that if the user specifies in an insert the value 0 to the auto-increment column, then InnoDB treats the row like the value would not have been specified. After the auto-increment counter has been initialized, if a user inserts a row where he explicitly specifies the column value, and the value is bigger than the current counter value, then the counter is set to the specified column value. If the user does not explicitly specify a value, then InnoDB increments the counter by one and assigns its new value to the column. The auto-increment mechanism, when assigning values from the counter, bypasses locking and transaction handling. Therefore you may also get gaps in the number sequence if you roll back transactions which have got numbers from the counter. The behavior of auto-increment is not defined if a user gives a negative value to the column or if the value becomes bigger than the maximum integer that can be stored in the specified integer type. @node File space management, Error handling, Table and index, InnoDB @subsection File space management and disk i/o @menu * InnoDB Disk i/o:: Disk i/o * InnoDB File space:: File space management * InnoDB File Defragmenting:: Defragmenting a table @end menu @node InnoDB Disk i/o, InnoDB File space, File space management, File space management @subsubsection Disk i/o In disk i/o InnoDB uses asynchronous i/o. On Windows NT it uses the native asynchronous i/o provided by the operating system. On Unix, InnoDB uses simulated asynchronous i/o built into InnoDB: InnoDB creates a number of i/o threads to take care of i/o operations, such as read-ahead. In a future version we will add support for simulated aio on Windows NT and native aio on those versions of Unix which have one. On Windows NT InnoDB uses non-buffered i/o. That means that the disk pages InnoDB reads or writes are not buffered in the operating system file cache. This saves some memory bandwidth. Starting from 3.23.41 InnoDB uses a novel file flush technique called doublewrite. It adds safety to crash recovery after an operating system crash or a power outage, and improves performance on most Unix flavors by reducing the need for fsync operations. Doublewrite means that InnoDB before writing pages to a data file first writes them to a contiguous tablespace area called the doublewrite buffer. Only after the write and the flush to the doublewrite buffer has completed, InnoDB writes the pages to their proper positions in the data file. If the operating system crashes in the middle of a page write, InnoDB will in recovery find a good copy of the page from the doublewrite buffer. Starting from 3.23.41 you can also use a raw disk partition as a data file, though this has not been tested yet. When you create a new data file you have to put the keyword @code{newraw} immediately after the data file size in @code{innodb_data_file_path}. The partition must be >= than you specify as the size. Note that 1M in InnoDB is 1024 x 1024 bytes, while in disk specifications 1 MB usually means 1000 000 bytes. @example innodb_data_file_path=hdd1:3Gnewraw;hdd2:2Gnewraw @end example When you start the database again you MUST change the keyword to @code{raw}. Otherwise InnoDB will write over your partition! @example innodb_data_file_path=hdd1:3Graw;hdd2:2Graw @end example Using a raw disk you can on some Unixes perform non-buffered i/o. There are two read-ahead heuristics in InnoDB: sequential read-ahead and random read-ahead. In sequential read-ahead InnoDB notices that the access pattern to a segment in the tablespace is sequential. Then InnoDB will post in advance a batch of reads of database pages to the i/o system. In random read-ahead InnoDB notices that some area in a tablespace seems to be in the process of being fully read into the buffer pool. Then InnoDB posts the remaining reads to the i/o system. @node InnoDB File space, InnoDB File Defragmenting, InnoDB Disk i/o, File space management @subsubsection File space management The data files you define in the configuration file form the tablespace of InnoDB. The files are simply catenated to form the tablespace, there is no striping in use. Currently you cannot directly instruct where the space is allocated for your tables, except by using the following fact: from a newly created tablespace InnoDB will allocate space starting from the low end. The tablespace consists of database pages whose default size is 16 kB. The pages are grouped into extents of 64 consecutive pages. The 'files' inside a tablespace are called segments in InnoDB. The name of the rollback segment is somewhat misleading because it actually contains many segments in the tablespace. For each index in InnoDB we allocate two segments: one is for non-leaf nodes of the B-tree, the other is for the leaf nodes. The idea here is to achieve better sequentiality for the leaf nodes, which contain the data. When a segment grows inside the tablespace, InnoDB allocates the first 32 pages to it individually. After that InnoDB starts to allocate whole extents to the segment. InnoDB can add to a large segment up to 4 extents at a time to ensure good sequentiality of data. Some pages in the tablespace contain bitmaps of other pages, and therefore a few extents in an InnoDB tablespace cannot be allocated to segments as a whole, but only as individual pages. When you issue a query @code{SHOW TABLE STATUS FROM ... LIKE ...} to ask for available free space in the tablespace, InnoDB will report you the space which is certainly usable in totally free extents of the tablespace. InnoDB always reserves some extents for clean-up and other internal purposes; these reserved extents are not included in the free space. When you delete data from a table, InnoDB will contract the corresponding B-tree indexes. It depends on the pattern of deletes if that frees individual pages or extents to the tablespace, so that the freed space is available for other users. Dropping a table or deleting all rows from it is guaranteed to release the space to other users, but remember that deleted rows can be physically removed only in a purge operation after they are no longer needed in transaction rollback or consistent read. @node InnoDB File Defragmenting, , InnoDB File space, File space management @subsubsection Defragmenting a table If there are random insertions or deletions in the indexes of a table, the indexes may become fragmented. By fragmentation we mean that the physical ordering of the index pages on the disk is not close to the alphabetical ordering of the records on the pages, or that there are many unused pages in the 64-page blocks which were allocated to the index. It can speed up index scans if you periodically use @code{mysqldump} to dump the table to a text file, drop the table, and reload it from the dump. Another way to do the defragmenting is to @code{ALTER} the table type to @code{MyISAM} and back to @code{InnoDB} again. Note that a @code{MyISAM} table must fit in a single file on your operating system. If the insertions to and index are always ascending and records are deleted only from the end, then the the file space management algorithm of InnoDB guarantees that fragmentation in the index will not occur. @node Error handling, InnoDB restrictions, File space management, InnoDB @subsection Error handling The error handling in InnoDB is not always the same as specified in the ANSI SQL standards. According to the ANSI standard, any error during an SQL statement should cause the rollback of that statement. InnoDB sometimes rolls back only part of the statement. The following list specifies the error handling of InnoDB. @itemize @bullet @item If you run out of file space in the tablespace, you will get the MySQL @code{'Table is full'} error and InnoDB rolls back the SQL statement. @item A transaction deadlock or a timeout in a lock wait will give @code{'Table handler error 1000000'} and InnoDB rolls back the SQL statement. @item A duplicate key error only rolls back the insert of that particular row, even in a statement like @code{INSERT INTO ... SELECT ...}. This will probably change so that the SQL statement will be rolled back if you have not specified the @code{IGNORE} option in your statement. @item A 'row too long' error rolls back the SQL statement. @item Other errors are mostly detected by the MySQL layer of code, and they roll back the corresponding SQL statement. @end itemize @node InnoDB restrictions, InnoDB contact information, Error handling, InnoDB @subsection Some restrictions on InnoDB tables @itemize @bullet @item @code{SHOW TABLE STATUS} does not give accurate statistics on InnoDB tables, except for the physical size reserved by the table. The row count is only a rough estimate used in SQL optimization. @item If you try to create an unique index on a prefix of a column you will get an error: @example CREATE TABLE T (A CHAR(20), B INT, UNIQUE (A(5))) TYPE = InnoDB; @end example If you create a non unique index on a prefix of a column, InnoDB will create an index over the whole column. @item @code{INSERT DELAYED} is not supported for InnoDB tables. @item The MySQL @code{LOCK TABLES} operation does not know of InnoDB row level locks set in already completed SQL statements: this means that you can get a table lock on a table even if there still exist transactions of other users which have row level locks on the same table. Thus your operations on the table may have to wait if they collide with these locks of other users. Also a deadlock is possible. However, this does not endanger transaction integrity, because the row level locks set by InnoDB will always take care of the integrity. Also, a table lock prevents other transactions from acquiring more row level locks (in a conflicting lock mode) on the table. @item You cannot have a key on a @code{BLOB} or @code{TEXT} column. @item A table cannot contain more than 1000 columns. @item @code{DELETE FROM TABLE} does not regenerate the table but instead deletes all rows, one by one, which is not that fast. In future versions of MySQL you can use @code{TRUNCATE} which is fast. @item Before dropping a database with InnoDB tables one has to drop the individual InnoDB tables first. @item The default database page size in InnoDB is 16 kB. By recompiling the code one can set it from 8 kB to 64 kB. The maximun row length is slightly less than half of a database page in versions <= 3.23.40 of InnoDB. Starting from source release 3.23.41 BLOB and TEXT columns are allowed to be < 4 GB, the total row length must also be < 4 GB. InnoDB does not store fields whose size is <= 30 bytes on separate pages. After InnoDB has modified the row by storing long fields on separate pages, the remaining length of the row must be slightly less than half a database page. @item The maximum data or log file size is 2 GB or 4 GB depending on how large files your operating system supports. Support for > 4 GB files will be added to InnoDB in a future version. @item The maximum tablespace size is 4 billion database pages. This is also the maximum size for a table. The minimum tablespace size is 10 MB. @end itemize @node InnoDB contact information, , InnoDB restrictions, InnoDB @subsection InnoDB contact information Contact information of Innobase Oy, producer of the InnoDB engine. Website: @uref{http://www.innodb.com}. Email: @email{Heikki.Tuuri@@innodb.com} @example phone: 358-9-6969 3250 (office) 358-40-5617367 (mobile) InnoDB Oy Inc. World Trade Center Helsinki Aleksanterinkatu 17 P.O.Box 800 00101 Helsinki Finland @end example @node Clients, Extending MySQL, Table types, Top @chapter MySQL APIs @cindex client tools @cindex APIs @cindex @code{mysqlclient} library @cindex buffer sizes, client @cindex library, @code{mysqlclient} @menu * PHP:: MySQL PHP API * Perl:: MySQL Perl API * ODBC:: MySQL ODBC Support * C:: MySQL C API * Cplusplus:: MySQL C++ APIs * Java:: MySQL Java connectivity (JDBC) * Python:: MySQL Python APIs * Tcl:: MySQL Tcl APIs * Eiffel:: MySQL Eiffel wrapper @end menu This chapter describes the APIs available for MySQL, where to get them, and how to use them. The C API is the most extensively covered, as it was developed by the MySQL team, and is the basis for most of the other APIs. @node PHP, Perl, Clients, Clients @section MySQL PHP API @cindex PHP API PHP is a server-side, HTML-embedded scripting language that may be used to create dynamic Web pages. It contains support for accessing several databases, including MySQL. PHP may be run as a separate program or compiled as a module for use with the Apache Web server. The distribution and documentation are available at the @uref{http://www.php.net/, PHP web site}. @menu * PHP problems:: Common problems with MySQL and PHP @end menu @node PHP problems, , PHP, PHP @subsection Common Problems with MySQL and PHP @itemize @bullet @item Error: "Maximum Execution Time Exceeded" This is a PHP limit; Go into the @file{php3.ini} file and set the maximum execution time up from 30 seconds to something higher, as needed. It is also not a bad idea to double the ram allowed per script to 16MB instead of 8 MB. @item Error: "Fatal error: Call to unsupported or undefined function mysql_connect() in .." This means that your PHP version isn't compiled with MySQL support. You can either compile a dynamic MySQL module and load it into PHP or recompile PHP with built-in MySQL support. This is described in detail in the PHP manual. @item Error: "undefined reference to `uncompress'" This means that the client library is compiled with support for a compressed client/server protocol. The fix is to add @code{-lz} last when linking with @code{-lmysqlclient}. @end itemize @node Perl, ODBC, PHP, Clients @section MySQL Perl API @cindex APIs, Perl @cindex Perl API This section documents the Perl @code{DBI} interface. The former interface was called @code{mysqlperl}. @code{DBI}/@code{DBD} now is the recommended Perl interface, so @code{mysqlperl} is obsolete and is not documented here. @menu * DBI with DBD:: @code{DBI} with @code{DBD::mysql} * Perl DBI Class:: The @code{DBI} interface * DBI-info:: More @code{DBI}/@code{DBD} information @end menu @node DBI with DBD, Perl DBI Class, Perl, Perl @subsection @code{DBI} with @code{DBD::mysql} @cindex @code{DBI} interface @code{DBI} is a generic interface for many databases. That means that you can write a script that works with many different database engines without change. You need a DataBase Driver (DBD) defined for each database type. For MySQL, this driver is called @code{DBD::mysql}. For more information on the Perl5 DBI, please visit the @code{DBI} Web page and read the documentation: @example @uref{http://www.symbolstone.org/technology/perl/DBI/index.html} @end example For more information on Object Oriented Programming (OOP) as defined in Perl5, see the Perl OOP page: @example @uref{http://language.perl.com/info/documentation.html} @end example Note that if you want to use transactions with Perl, you need to have @code{Msql-Mysql-modules} version 1.2216 or newer. Installation instructions for MySQL Perl support are given in @ref{Perl support}. @node Perl DBI Class, DBI-info, DBI with DBD, Perl @subsection The @code{DBI} Interface @cindex @code{DBI} Perl module @noindent @strong{Portable DBI Methods} @multitable @columnfractions .3 .7 @item @code{connect} @tab Establishes a connection to a database server. @item @code{disconnect} @tab Disconnects from the database server. @item @code{prepare} @tab Prepares a SQL statement for execution. @item @code{execute} @tab Executes prepared statements. @item @code{do} @tab Prepares and executes a SQL statement. @item @code{quote} @tab Quotes string or @code{BLOB} values to be inserted. @item @code{fetchrow_array} @tab Fetches the next row as an array of fields. @item @code{fetchrow_arrayref} @tab Fetches next row as a reference array of fields. @item @code{fetchrow_hashref} @tab Fetches next row as a reference to a hashtable. @item @code{fetchall_arrayref} @tab Fetches all data as an array of arrays. @item @code{finish} @tab Finishes a statement and lets the system free resources. @item @code{rows} @tab Returns the number of rows affected. @item @code{data_sources} @tab Returns an array of databases available on localhost. @item @code{ChopBlanks} @tab Controls whether @code{fetchrow_*} methods trim spaces. @item @code{NUM_OF_PARAMS} @tab The number of placeholders in the prepared statement. @item @code{NULLABLE} @tab Which columns can be @code{NULL}. @item @code{trace} @tab Perform tracing for debugging. @end multitable @noindent @strong{MySQL-specific Methods} @multitable @columnfractions .3 .7 @item @code{insertid} @tab The latest @code{AUTO_INCREMENT} value. @item @code{is_blob} @tab Which columns are @code{BLOB} values. @item @code{is_key} @tab Which columns are keys. @item @code{is_num} @tab Which columns are numeric. @item @code{is_pri_key} @tab Which columns are primary keys. @item @code{is_not_null} @tab Which columns CANNOT be @code{NULL}. See @code{NULLABLE}. @item @code{length} @tab Maximum possible column sizes. @item @code{max_length} @tab Maximum column sizes actually present in result. @item @code{NAME} @tab Column names. @item @code{NUM_OF_FIELDS} @tab Number of fields returned. @item @code{table} @tab Table names in returned set. @item @code{type} @tab All column types. @end multitable The Perl methods are described in more detail in the following sections. Variables used for method return values have these meanings: @table @code @item $dbh Database handle @item $sth Statement handle @item $rc Return code (often a status) @item $rv Return value (often a row count) @end table @noindent @strong{Portable DBI Methods} @table @code @findex DBI->connect() @findex connect() DBI method @item connect($data_source, $username, $password) Use the @code{connect} method to make a database connection to the data source. The @code{$data_source} value should begin with @code{DBI:driver_name:}. Example uses of @code{connect} with the @code{DBD::mysql} driver: @example $dbh = DBI->connect("DBI:mysql:$database", $user, $password); $dbh = DBI->connect("DBI:mysql:$database:$hostname", $user, $password); $dbh = DBI->connect("DBI:mysql:$database:$hostname:$port", $user, $password); @end example If the user name and/or password are undefined, @code{DBI} uses the values of the @code{DBI_USER} and @code{DBI_PASS} environment variables, respectively. If you don't specify a hostname, it defaults to @code{'localhost'}. If you don't specify a port number, it defaults to the default MySQL port (@value{default_port}). As of @code{Msql-Mysql-modules} Version 1.2009, the @code{$data_source} value allows certain modifiers: @table @code @item mysql_read_default_file=file_name Read @file{filename} as an option file. For information on option files, see @ref{Option files}. @item mysql_read_default_group=group_name The default group when reading an option file is normally the @code{[client]} group. By specifying the @code{mysql_read_default_group} option, the default group becomes the @code{[group_name]} group. @item mysql_compression=1 Use compressed communication between the client and server (MySQL Version 3.22.3 or later). @item mysql_socket=/path/to/socket Specify the pathname of the Unix socket that is used to connect to the server (MySQL Version 3.21.15 or later). @end table Multiple modifiers may be given; each must be preceded by a semicolon. For example, if you want to avoid hardcoding the user name and password into a @code{DBI} script, you can take them from the user's @file{~/.my.cnf} option file instead by writing your @code{connect} call like this: @example $dbh = DBI->connect("DBI:mysql:$database" . ";mysql_read_default_file=$ENV@{HOME@}/.my.cnf", $user, $password); @end example This call will read options defined for the @code{[client]} group in the option file. If you wanted to do the same thing but use options specified for the @code{[perl]} group as well, you could use this: @example $dbh = DBI->connect("DBI:mysql:$database" . ";mysql_read_default_file=$ENV@{HOME@}/.my.cnf" . ";mysql_read_default_group=perl", $user, $password); @end example @findex DBI->disconnect @findex disconnect DBI method @item disconnect The @code{disconnect} method disconnects the database handle from the database. This is typically called right before you exit from the program. Example: @example $rc = $dbh->disconnect; @end example @findex DBI->prepare() @findex prepare() DBI method @item prepare($statement) Prepares a SQL statement for execution by the database engine and returns a statement handle @code{($sth)}, which you can use to invoke the @code{execute} method. Typically you handle @code{SELECT} statements (and @code{SELECT}-like statements such as @code{SHOW}, @code{DESCRIBE}, and @code{EXPLAIN}) by means of @code{prepare} and @code{execute}. Example: @example $sth = $dbh->prepare($statement) or die "Can't prepare $statement: $dbh->errstr\n"; @end example @findex DBI->execute @findex execute DBI method @item execute The @code{execute} method executes a prepared statement. For non-@code{SELECT} statements, @code{execute} returns the number of rows affected. If no rows are affected, @code{execute} returns @code{"0E0"}, which Perl treats as zero but regards as true. If an error occurs, @code{execute} returns @code{undef}. For @code{SELECT} statements, @code{execute} only starts the SQL query in the database; you need to use one of the @code{fetch_*} methods described below to retrieve the data. Example: @example $rv = $sth->execute or die "can't execute the query: $sth->errstr; @end example @findex DBI->do() @findex do() DBI method @item do($statement) The @code{do} method prepares and executes a SQL statement and returns the number of rows affected. If no rows are affected, @code{do} returns @code{"0E0"}, which Perl treats as zero but regards as true. This method is generally used for non-@code{SELECT} statements that cannot be prepared in advance (due to driver limitations) or that do not need to be executed more than once (inserts, deletes, etc.). Example: @example $rv = $dbh->do($statement) or die "Can't execute $statement: $dbh- >errstr\n"; @end example Generally the 'do' statement is MUCH faster (and is preferable) than prepare/execute for statements that don't contain parameters. @findex DBI->quote() @findex quote() DBI method @cindex quoting strings @cindex strings, quoting @item quote($string) The @code{quote} method is used to "escape" any special characters contained in the string and to add the required outer quotation marks. Example: @example $sql = $dbh->quote($string) @end example @findex DBI->fetchrow_array @findex fetchrow_array DBI method @item fetchrow_array This method fetches the next row of data and returns it as an array of field values. Example: @example while(@@row = $sth->fetchrow_array) @{ print qw($row[0]\t$row[1]\t$row[2]\n); @} @end example @findex DBI->fetchrow_arrayref @findex fetchrow_arrayref DBI method @item fetchrow_arrayref This method fetches the next row of data and returns it as a reference to an array of field values. Example: @example while($row_ref = $sth->fetchrow_arrayref) @{ print qw($row_ref->[0]\t$row_ref->[1]\t$row_ref->[2]\n); @} @end example @findex DBI->fetchrow_hashref @findex fetchrow_hashref DBI method @item fetchrow_hashref This method fetches a row of data and returns a reference to a hash table containing field name/value pairs. This method is not nearly as efficient as using array references as demonstrated above. Example: @example while($hash_ref = $sth->fetchrow_hashref) @{ print qw($hash_ref->@{firstname@}\t$hash_ref->@{lastname@}\t\ $hash_ref- > title@}\n); @} @end example @findex DBI->fetchall_arrayref @findex fetchall_arrayref DBI method @item fetchall_arrayref This method is used to get all the data (rows) to be returned from the SQL statement. It returns a reference to an array of references to arrays for each row. You access or print the data by using a nested loop. Example: @example my $table = $sth->fetchall_arrayref or die "$sth->errstr\n"; my($i, $j); for $i ( 0 .. $#@{$table@} ) @{ for $j ( 0 .. $#@{$table->[$i]@} ) @{ print "$table->[$i][$j]\t"; @} print "\n"; @} @end example @findex DBI->finish @findex finish DBI method @item finish Indicates that no more data will be fetched from this statement handle. You call this method to free up the statement handle and any system resources associated with it. Example: @example $rc = $sth->finish; @end example @findex DBI->rows @findex rows DBI method @item rows Returns the number of rows changed (updated, deleted, etc.) by the last command. This is usually used after a non-@code{SELECT} @code{execute} statement. Example: @example $rv = $sth->rows; @end example @findex DBI->@{NULLABLE@} @findex NULLABLE DBI method @item NULLABLE Returns a reference to an array of boolean values; for each element of the array, a value of TRUE indicates that this column may contain @code{NULL} values. Example: @example $null_possible = $sth->@{NULLABLE@}; @end example @findex DBI->@{NUM_OF_FIELDS@} @findex NUM_OF_FIELDS DBI method @item NUM_OF_FIELDS This attribute indicates the number of fields returned by a @code{SELECT} or @code{SHOW FIELDS} statement. You may use this for checking whether a statement returned a result: A zero value indicates a non-@code{SELECT} statement like @code{INSERT}, @code{DELETE}, or @code{UPDATE}. Example: @example $nr_of_fields = $sth->@{NUM_OF_FIELDS@}; @end example @findex DBI->data_sources() @findex data_sources() DBI method @item data_sources($driver_name) This method returns an array containing names of databases available to the MySQL server on the host @code{'localhost'}. Example: @example @@dbs = DBI->data_sources("mysql"); @end example @findex DBI->@{ChopBlanks@} @findex ChopBlanks DBI method @item ChopBlanks This attribute determines whether the @code{fetchrow_*} methods will chop leading and trailing blanks from the returned values. Example: @example $sth->@{'ChopBlanks'@} =1; @end example @findex DBI->trace @findex trace DBI method @item trace($trace_level) @itemx trace($trace_level, $trace_filename) The @code{trace} method enables or disables tracing. When invoked as a @code{DBI} class method, it affects tracing for all handles. When invoked as a database or statement handle method, it affects tracing for the given handle (and any future children of the handle). Setting @code{$trace_level} to 2 provides detailed trace information. Setting @code{$trace_level} to 0 disables tracing. Trace output goes to the standard error output by default. If @code{$trace_filename} is specified, the file is opened in append mode and output for @emph{all} traced handles is written to that file. Example: @example DBI->trace(2); # trace everything DBI->trace(2,"/tmp/dbi.out"); # trace everything to # /tmp/dbi.out $dth->trace(2); # trace this database handle $sth->trace(2); # trace this statement handle @end example @tindex @code{DBI_TRACE} environment variable @tindex environment variable, @code{DBI_TRACE} You can also enable @code{DBI} tracing by setting the @code{DBI_TRACE} environment variable. Setting it to a numeric value is equivalent to calling @code{DBI->(value)}. Setting it to a pathname is equivalent to calling @code{DBI->(2,value)}. @end table @noindent @strong{MySQL-specific Methods} The methods shown below are MySQL-specific and not part of the @code{DBI} standard. Several of them are now deprecated: @code{is_blob}, @code{is_key}, @code{is_num}, @code{is_pri_key}, @code{is_not_null}, @code{length}, @code{max_length}, and @code{table}. Where @code{DBI}-standard alternatives exist, they are noted below: @table @code @findex DBI->@{insertid@} @findex insertid DBI method @tindex AUTO_INCREMENT, using with DBI @item insertid If you use the @code{AUTO_INCREMENT} feature of MySQL, the new auto-incremented values will be stored here. Example: @example $new_id = $sth->@{insertid@}; @end example As an alternative, you can use @code{$dbh->@{'mysql_insertid'@}}. @findex DBI->@{is_blob@} @findex is_blob DBI method @item is_blob Returns a reference to an array of boolean values; for each element of the array, a value of TRUE indicates that the respective column is a @code{BLOB}. Example: @example $keys = $sth->@{is_blob@}; @end example @findex DBI->@{is_key@} @findex is_key DBI method @item is_key Returns a reference to an array of boolean values; for each element of the array, a value of TRUE indicates that the respective column is a key. Example: @example $keys = $sth->@{is_key@}; @end example @findex DBI->@{is_num@} @findex is_num DBI method @item is_num Returns a reference to an array of boolean values; for each element of the array, a value of TRUE indicates that the respective column contains numeric values. Example: @example $nums = $sth->@{is_num@}; @end example @findex DBI->@{is_pri_key@} @findex is_pri_key DBI method @item is_pri_key Returns a reference to an array of boolean values; for each element of the array, a value of TRUE indicates that the respective column is a primary key. Example: @example $pri_keys = $sth->@{is_pri_key@}; @end example @findex DBI->@{is_not_null@} @findex is_not_null DBI method @item is_not_null Returns a reference to an array of boolean values; for each element of the array, a value of FALSE indicates that this column may contain @code{NULL} values. Example: @example $not_nulls = $sth->@{is_not_null@}; @end example @code{is_not_null} is deprecated; it is preferable to use the @code{NULLABLE} attribute (described above), because that is a DBI standard. @findex DBI->@{length@} @findex length DBI method @findex DBI->@{max_length@} @findex max_length DBI method @item length @itemx max_length Each of these methods returns a reference to an array of column sizes. The @code{length} array indicates the maximum possible sizes that each column may be (as declared in the table description). The @code{max_length} array indicates the maximum sizes actually present in the result table. Example: @example $lengths = $sth->@{length@}; $max_lengths = $sth->@{max_length@}; @end example @findex DBI->@{NAME@} @findex NAME DBI method @item NAME Returns a reference to an array of column names. Example: @example $names = $sth->@{NAME@}; @end example @findex DBI->@{table@} @findex table DBI method @item table Returns a reference to an array of table names. Example: @example $tables = $sth->@{table@}; @end example @findex DBI->@{type@} @findex type DBI method @item type Returns a reference to an array of column types. Example: @example $types = $sth->@{type@}; @end example @end table @node DBI-info, , Perl DBI Class, Perl @subsection More @code{DBI}/@code{DBD} Information @cindex @code{DBI/DBD} You can use the @code{perldoc} command to get more information about @code{DBI}. @example perldoc DBI perldoc DBI::FAQ perldoc DBD::mysql @end example You can also use the @code{pod2man}, @code{pod2html}, etc., tools to translate to other formats. You can find the latest @code{DBI} information at the @code{DBI} Web page: @example @uref{http://www.symbolstone.org/technology/perl/DBI/index.html} @end example @node ODBC, C, Perl, Clients @section MySQL ODBC Support @cindex ODBC @cindex Windows @cindex MyODBC @menu * Installing MyODBC:: How to install MyODBC * ODBC administrator:: How to fill in the various fields in the ODBC administrator program * MyODBC connect parameters:: Connect parameters for MyODBC * ODBC Problems:: How to report problems with MySQL ODBC * MyODBC clients:: Programs known to work with @strong{MyODBC} * ODBC and last_insert_id:: How to get the value of an @code{AUTO_INCREMENT} column in ODBC * MyODBC bug report:: Reporting problems with MyODBC @end menu MySQL provides support for ODBC by means of the @strong{MyODBC} program. This chapter will teach you how to install @strong{MyODBC}, and how to use it. Here, you will also find a list of common programs that are known to work with @strong{MyODBC}. @node Installing MyODBC, ODBC administrator, ODBC, ODBC @subsection How To Install MyODBC @strong{MyODBC} is a 32-bit ODBC (2.50) level 0 (with level 1 and level 2 features) driver for connecting an ODBC-aware application to MySQL. @strong{MyODBC} works on Windows95, Windows98, NT, and on most Unix platforms. @strong{MyODBC} is in public domain, and you can find the newest version at @uref{http://www.mysql.com/downloads/api-myodbc.html}. If you have problem with @strong{MyODBC} and your program also works with OLEDB, you should try the OLEDB driver that you can find in the Contrib section. @xref{Contrib}. Normally you only need to install @strong{MyODBC} on Windows machines. You only need @strong{MyODBC} for Unix if you have a program like ColdFusion that is running on the Unix machine and uses ODBC to connect to the databases. If you want to install @strong{MyODBC} on a Unix box, you will also need an @strong{ODBC} manager. @strong{MyODBC} is known to work with most of the Unix ODBC managers. You can find a list at these in the @strong{ODBC}-related links section on the MySQL useful links page. @xref{Useful Links}. To install @strong{MyODBC} on Windows, you should download the appropriate @strong{MyODBC} .zip file (for Windows or NT/Win2000), unpack it with @code{WINZIP}, or some similar program, and execute the @code{SETUP.EXE} file. On Windows/NT you may get the following error when trying to install @strong{MyODBC}: @example An error occurred while copying C:\WINDOWS\SYSTEM\MFC30.DLL. Restart Windows and try installing again (before running any applications which use ODBC) @end example The problem in this case is that some other program is using ODBC and because of how Windows is designed, you may not in this case be able to install a new ODBC drivers with Microsoft's ODBC setup program. In most cases you can continue by just pressing @code{Ignore} to copy the rest of the MyODBC files and the final installation should still work. If this doesn't work, the solution is to reboot your computer in ``safe mode`` (Choose this by pressing F8 just before your machine starts Windows during rebooting), install @strong{MyODBC}, and reboot to normal mode. @itemize @bullet @item To make a connection to a Unix box from a Windows box, with an ODBC application (one that doesn't support MySQL natively), you must first install @strong{MyODBC} on the Windows machine. @item The user and Windows machine must have the access privileges to the MySQL server on the Unix machine. This is set up with the @code{GRANT} command. @xref{GRANT,,@code{GRANT}}. @item You must create an ODBC DSN entry as follows: @itemize @minus @item Open the Control Panel on the Windows machine. @item Double-click the ODBC Data Sources 32 bits icon. @item Click the tab User DSN. @item Click the button Add. @item Select MySQL in the screen Create New Data Source and click the Finish button. @item The MySQL Driver default configuration screen is shown. @xref{ODBC administrator}. @end itemize @item Now start your application and select the ODBC driver with the DSN you specified in the ODBC administrator. @end itemize Notice that there are other configuration options on the screen of MySQL (trace, don't prompt on connect, etc) that you can try if you run into problems. @node ODBC administrator, MyODBC connect parameters, Installing MyODBC, ODBC @subsection How to Fill in the Various Fields in the ODBC Administrator Program @cindex ODBC, administrator There are three possibilities for specifying the server name on Windows95: @itemize @bullet @item Use the IP address of the server. @item Add a file @file{\windows\lmhosts} with the following information: @example ip hostname @end example For example: @example 194.216.84.21 my_hostname @end example @item Configure the PC to use DNS. @end itemize Example of how to fill in the @code{ODBC setup}: @example Windows DSN name: test Description: This is my test database MySql Database: test Server: 194.216.84.21 User: monty Password: my_password Port: @end example The value for the @code{Windows DSN name} field is any name that is unique in your Windows ODBC setup. You don't have to specify values for the @code{Server}, @code{User}, @code{Password}, or @code{Port} fields in the ODBC setup screen. However, if you do, the values will be used as the defaults later when you attempt to make a connection. You have the option of changing the values at that time. If the port number is not given, the default port (@value{default_port}) is used. If you specify the option @code{Read options from C:\my.cnf}, the groups @code{client} and @code{odbc} will be read from the @file{C:\my.cnf} file. You can use all options that are usable by @code{mysql_options()}. @xref{mysql_options, , @code{mysql_options}}. @node MyODBC connect parameters, ODBC Problems, ODBC administrator, ODBC @subsection Connect parameters for MyODBC One can specify the following parameters for @strong{MyODBC} on the @code{[Servername]} section of an @code{ODBC.INI} file or through the @code{InConnectionString} argument in the @code{SQLDriverConnect()} call. @multitable @columnfractions .2 .2 .6 @item @strong{Parameter} @tab @strong{Default value} @tab @strong{Comment} @item user @tab ODBC (on Windows) @tab The username used to connect to MySQL. @item server @tab localhost @tab The hostname of the MySQL server. @item database @tab @tab The default database @item option @tab 0 @tab A integer by which you can specify how @strong{MyODBC} should work. See below. @item port @tab 3306 @tab The TCP/IP port to use if @code{server} is not @code{localhost}. @item stmt @tab @tab A statement that will be executed when connection to @code{MySQL}. @item password @tab @tab The password for the @code{server} @code{user} combination. @item socket @tab @tab The socket or Windows pipe to connect to. @end multitable The option argument is used to tell @strong{MyODBC} that the client isn't 100% ODBC compliant. On Windows, one normally sets the option flag by toggling the different options on the connection screen but one can also set this in the opton argument. The following options are listed in the same order as they appear in the @strong{MyODBC} connect screen: @multitable @columnfractions .1 .9 @item @strong{Bit} @tab @strong{Meaning} @item 1 @tab The client can't handle that @strong{MyODBC} returns the real width of a column. @item 2 @tab The client can't handle that MySQL returns the true value of affected rows. If this flag is set then MySQL returns 'found rows' instead. One must have MySQL 3.21.14 or newer to get this to work. @item 4 @tab Make a debug log in c:\myodbc.log. This is the same as putting @code{MYSQL_DEBUG=d:t:O,c::\myodbc.log} in @file{AUTOEXEC.BAT} @item 8 @tab Don't set any packet limit for results and parameters. @item 16 @tab Don't prompt for questions even if driver would like to prompt @item 32 @tab Simulate a ODBC 1.0 driver in some context. @item 64 @tab Ignore use of database name in 'database.table.column'. @item 128 @tab Force use of ODBC manager cursors (experimental). @item 256 @tab Disable the use of extended fetch (experimental) @item 512 @tab Pad CHAR fields to full column length. @item 1024 @tab SQLDescribeCol() will return fully qualifed column names @item 2048 @tab Use the compressed server/client protocol @item 4096 @tab Tell server to ignore space after function name and before @code{'('} (needed by PowerBuilder). This will make all function names keywords! @item 8192 @tab Connect with named pipes to a @code{mysqld} server running on NT. @item 16384 @tab Change LONGLONG columns to INT columns (Some applications can't handle LONGLONG). @item 32768 @tab Return 'user' as Table_qualifier and Table_owner from SQLTables (experimental) @item 65536 @tab Read parameters from the @code{client} and @code{odbc} groups from @file{my.cnf} @item 131072 @tab Add some extra safety checks (should not bee needed but...) @end multitable If you want to have many options, you should add the above flags! For example setting option to 12 (4+8) gives you debugging without package limits! The default @file{MYODBC.DLL} is compiled for optimal performance. If you want to to debug @strong{MyODBC} (for example to enable tracing), you should instead use @code{MYODBCD.DLL}. To install this file, copy @file{MYODBCD.DLL} over the installed @code{MYODBC.DLL} file. @node ODBC Problems, MyODBC clients, MyODBC connect parameters, ODBC @subsection How to Report Problems with MyODBC @strong{MyODBC} has been tested with Access, Admndemo.exe, C++-Builder, Borland Builder 4, Centura Team Developer (formerly Gupta SQL/Windows), ColdFusion (on Solaris and NT with svc pack 5), Crystal Reports, DataJunction, Delphi, ERwin, Excel, iHTML, FileMaker Pro, FoxPro, Notes 4.5/4.6, SBSS, Perl DBD-ODBC, Paradox, Powerbuilder, Powerdesigner 32 bit, VC++, and Visual Basic. If you know of any other applications that work with @strong{MyODBC}, please send mail to @email{myodbc@@lists.mysql.com} about this! With some programs you may get an error like: @code{Another user has modifies the record that you have modified}. In most cases this can be solved by doing one of the following things: @itemize @bullet @item Add a primary key for the table if there isn't one already. @item Add a timestamp column if there isn't one already. @item Only use double float fields. Some programs may fail when they compare single floats. @end itemize If the above doesn't help, you should do a @code{MyODBC} trace file and try to figure out why things go wrong. @node MyODBC clients, ODBC and last_insert_id, ODBC Problems, ODBC @subsection Programs Known to Work with MyODBC Most programs should work with @strong{MyODBC}, but for each of those listed below, we have tested it ourselves or received confirmation from some user that it works: @table @asis @item @strong{Program} @strong{Comment} @cindex Access program @item Access To make Access work: @itemize @bullet @item If you are using Access 2000, you should get and install the newest (version 2.6 or above) Microsoft MDAC (@code{Microsoft Data Access Components}) from @uref{http://www.microsoft.com/data}. This will fix the following bug in Access: when you export data to MySQL, the table and column names aren't specified. Another way to around this bug is to upgrade to MyODBC Version 2.50.33 and MySQL Version 3.23.x, which together provide a workaround for this bug! You should also get and apply the Microsoft Jet 4.0 Service Pack 5 (SP5) which can be found here @uref{http://support.microsoft.com/support/kb/articles/Q 239/1/14.ASP}. This will fix some cases where columns are marked as @code{#deleted#} in Access. Note that if you are using MySQL Version 3.22, you must to apply the MDAC patch and use MyODBC 2.50.32 or 2.50.34 and above to go around this problem. @item Set the ``Return matching rows'' MyODBC option field when connecting to MySQL. @item You should have a primary key in the table. If not, new or updated rows may show up as @code{#Deleted#}. @item You should have a timestamp in all tables you want to be able to update. For maximum portability @code{TIMESTAMP(14)} or simple @code{TIMESTAMP} is recommended instead of other @code{TIMESTAMP(X)} variations. @item Only use double float fields. Access fails when comparing with single floats. The symptom usually is that new or updated rows may show up as @code{#Deleted#} or that you can't find or update rows. @item If you still get the error @code{Another user has changed your data} after adding a @code{TIMESTAMP} column, the following trick may help you: Don't use @code{table} data sheet view. Create instead a form with the fields you want, and use that @code{form} data sheet view. You should set the @code{DefaultValue} property for the @code{TIMESTAMP} column to @code{NOW()}. It may be a good idea to hide the @code{TIMESTAMP} column from view so your users are not confused. @item Access on NT will report @code{BLOB} columns as @code{OLE OBJECTS}. If you want to have @code{MEMO} columns instead, you should change the column to @code{TEXT} with @code{ALTER TABLE}. @item Access can't always handle @code{DATE} columns properly. If you have a problem with these, change the columns to @code{DATETIME}. @item In some cases, Access may generate illegal SQL queries that MySQL can't understand. You can fix this by selecting @code{"Query|SQLSpecific|Pass-Through"} from the Access menu. @item If you have in Access a column defined as BYTE, Access will try to export this as @code{TINYINT} instead of @code{TINYINT UNSIGNED}. This will give you problems if you have values > 127 in the column! @item If you are using Access 7.0, You should use the option flag @code{Return matching rows}. @item If you are using Access 2.0, You should use the option flags @code{Return matching rows} and @code{Simulate ODBC 1.0}. @end itemize @cindex ADO program @item ADO When you are coding with the ADO API and @strong{MyODBC} you need to put attention in some default properties that aren't supported by the MySQL server. For example, using the @code{CursorLocation Property} as @code{adUseServer} will return for the @code{RecordCount Property} a result of -1. To have the right value, you need to set this property to @code{adUseClient}, like is showing in the VB code below: @example Dim myconn As New ADODB.Connection Dim myrs As New Recordset Dim mySQL As String Dim myrows As Long myconn.Open "DSN=MyODBCsample" mySQL = "SELECT * from user" myrs.Source = mySQL Set myrs.ActiveConnection = myconn myrs.CursorLocation = adUseClient myrs.Open myrows = myrs.RecordCount myrs.Close myconn.Close @end example Another workaround is to use a @code{SELECT COUNT(*)} statement for a similar query to get the correct row count. @item Active server pages (ASP) You should use the option flag @code{Return matching rows}. @item BDE applications To get these to work, you should set the option flags @code{Don't optimize column widths} and @code{Return matching rows}. @cindex Borland Builder 4 program @item Borland Builder 4 When you start a query you can use the property @code{Active} or use the method @code{Open}. Note that @code{Active} will start by automatically issuing a @code{SELECT * FROM ...} query that may not be a good thing if your tables are big! @item ColdFusion (On Unix) The following information is taken from the ColdFusion documentation: Use the following information to configure ColdFusion Server for Linux to use the unixODBC driver with @strong{MyODBC} for MySQL data sources. Allaire has verified that @strong{MyODBC} Version 2.50.26 works with MySQL Version 3.22.27 and ColdFusion for Linux. (Any newer version should also work.) You can download @strong{MyODBC} at @uref{http://www.mysql.com/downloads/api-myodbc.html} @cindex ColdFusion program ColdFusion Version 4.5.1 allows you to us the ColdFusion Administrator to add the MySQL data source. However, the driver is not included with ColdFusion Version 4.5.1. Before the MySQL driver will appear in the ODBC datasources drop-down list, you must build and copy the @strong{MyODBC} driver to @file{/opt/coldfusion/lib/libmyodbc.so}. The Contrib directory contains the program mydsn-xxx.zip which allows you to build and remove the DSN registry file for the MyODBC driver on Coldfusion applications. @cindex DataJunction @item DataJunction You have to change it to output @code{VARCHAR} rather than @code{ENUM}, as it exports the latter in a manner that causes MySQL grief. @cindex Excel @item Excel Works. Some tips: @itemize @bullet @item If you have problems with dates, try to select them as strings using the @code{CONCAT()} function. For example: @example select CONCAT(rise_time), CONCAT(set_time) from sunrise_sunset; @end example Values retrieved as strings this way should be correctly recognized as time values by Excel97. The purpose of @code{CONCAT()} in this example is to fool ODBC into thinking the column is of ``string type''. Without the @code{CONCAT()}, ODBC knows the column is of time type, and Excel does not understand that. Note that this is a bug in Excel, because it automatically converts a string to a time. This would be great if the source was a text file, but is plain stupid when the source is an ODBC connection that reports exact types for each column. @end itemize @cindex Word program @item Word To retrieve data from MySQL to Word/Excel documents, you need to use the @code{MyODBC} driver and the Add-in Microsoft Query help. For example, create a db with a table containing 2 columns of text: @itemize @bullet @item Insert rows using the @code{mysql} client command-line tool. @item Create a DSN file using the MyODBC driver, for example, my for the db above. @item Open the Word application. @item Create a blank new documentation. @item Using the tool bar called Database, press the button insert database. @item Press the button Get Data. @item At the right hand of the screen Get Data, press the button Ms Query. @item In the Ms Query create a New Data Source using the DSN file my. @item Select the new query. @item Select the columns that you want. @item Make a filter if you want. @item Make a Sort if you want. @item Select Return Data to Microsoft Word. @item Click Finish. @item Click Insert data and select the records. @item Click OK and you see the rows in your Word document. @end itemize @cindex odbcadmin program @item odbcadmin Test program for ODBC. @cindex Delphi program @item Delphi You must use BDE Version 3.2 or newer. Set the `Don't optimize column width' option field when connecting to MySQL. Also, here is some potentially useful Delphi code that sets up both an ODBC entry and a BDE entry for @strong{MyODBC} (the BDE entry requires a BDE Alias Editor that is free at a Delphi Super Page near you. (Thanks to Bryan Brunton @email{bryan@@flesherfab.com} for this): @example fReg:= TRegistry.Create; fReg.OpenKey('\Software\ODBC\ODBC.INI\DocumentsFab', True); fReg.WriteString('Database', 'Documents'); fReg.WriteString('Description', ' '); fReg.WriteString('Driver', 'C:\WINNT\System32\myodbc.dll'); fReg.WriteString('Flag', '1'); fReg.WriteString('Password', ''); fReg.WriteString('Port', ' '); fReg.WriteString('Server', 'xmark'); fReg.WriteString('User', 'winuser'); fReg.OpenKey('\Software\ODBC\ODBC.INI\ODBC Data Sources', True); fReg.WriteString('DocumentsFab', 'MySQL'); fReg.CloseKey; fReg.Free; Memo1.Lines.Add('DATABASE NAME='); Memo1.Lines.Add('USER NAME='); Memo1.Lines.Add('ODBC DSN=DocumentsFab'); Memo1.Lines.Add('OPEN MODE=READ/WRITE'); Memo1.Lines.Add('BATCH COUNT=200'); Memo1.Lines.Add('LANGDRIVER='); Memo1.Lines.Add('MAX ROWS=-1'); Memo1.Lines.Add('SCHEMA CACHE DIR='); Memo1.Lines.Add('SCHEMA CACHE SIZE=8'); Memo1.Lines.Add('SCHEMA CACHE TIME=-1'); Memo1.Lines.Add('SQLPASSTHRU MODE=SHARED AUTOCOMMIT'); Memo1.Lines.Add('SQLQRYMODE='); Memo1.Lines.Add('ENABLE SCHEMA CACHE=FALSE'); Memo1.Lines.Add('ENABLE BCD=FALSE'); Memo1.Lines.Add('ROWSET SIZE=20'); Memo1.Lines.Add('BLOBS TO CACHE=64'); Memo1.Lines.Add('BLOB SIZE=32'); AliasEditor.Add('DocumentsFab','MySQL',Memo1.Lines); @end example @cindex C++ Builder @item C++ Builder Tested with BDE Version 3.0. The only known problem is that when the table schema changes, query fields are not updated. BDE, however, does not seem to recognize primary keys, only the index PRIMARY, though this has not been a problem. @item Vision You should use the option flag @code{Return matching rows}. @cindex Visual Basic @item Visual Basic To be able to update a table, you must define a primary key for the table. Visual Basic with ADO can't handle big integers. This means that some queries like @code{SHOW PROCESSLIST} will not work properly. The fix is to set add the option @code{OPTION=16834} in the ODBC connect string or set the @code{Change BIGINT columns to INT} option in the MyODBC connect screen. You may also want to set the @code{Return matching rows} option. @item VisualInterDev If you get the error @code{[Microsoft][ODBC Driver Manager] Driver does not support this parameter} the reason may be that you have a @code{BIGINT} in your result. Try setting the @code{Change BIGINT columns to INT} option in the MyODBC connect screen. @item Visual Objects You should use the option flag @code{Don't optimize column widths}. @end table @node ODBC and last_insert_id, MyODBC bug report, MyODBC clients, ODBC @subsection How to Get the Value of an @code{AUTO_INCREMENT} Column in ODBC @cindex AUTO-INCREMENT, ODBC A common problem is how to get the value of an automatically generated ID from an @code{INSERT}. With ODBC, you can do something like this (assuming that @code{auto} is an @code{AUTO_INCREMENT} field): @example INSERT INTO foo (auto,text) VALUES(NULL,'text'); SELECT LAST_INSERT_ID(); @end example Or, if you are just going to insert the ID into another table, you can do this: @example INSERT INTO foo (auto,text) VALUES(NULL,'text'); INSERT INTO foo2 (id,text) VALUES(LAST_INSERT_ID(),'text'); @end example @xref{Getting unique ID}. For the benefit of some ODBC applications (at least Delphi and Access), the following query can be used to find a newly inserted row: @example SELECT * FROM tbl_name WHERE auto IS NULL; @end example @node MyODBC bug report, , ODBC and last_insert_id, ODBC @subsection Reporting Problems with MyODBC @cindex reporting, MyODBC problems @cindex problems, ODBC @cindex MyODBC, reporting problems If you encounter difficulties with @strong{MyODBC}, you should start by making a log file from the ODBC manager (the log you get when requesting logs from ODBCADMIN) and a @strong{MyODBC} log. To get a @strong{MyODBC} log, you need to do the following: @enumerate @item Ensure that you are using @code{myodbcd.dll} and not @code{myodbc.dll}. The easiest way to do this is to get @code{myodbcd.dll} from the MyODBC distribution and copy it over the @code{myodbc.dll}, which is probably in your @code{C:\windows\system32} or @code{C:\winnt\system32} directory. Note that you probably want to restore the old myodbc.dll file when you have finished testing, as this is a lot faster than @code{myodbcd.dll}. @item Tag the `Trace MyODBC' option flag in the @strong{MyODBC} connect/configure screen. The log will be written to file @file{C:\myodbc.log}. If the trace option is not remembered when you are going back to the above screen, it means that you are not using the @code{myodbcd.dll} driver (see above). @item Start your application and try to get it to fail. @end enumerate Check the @code{MyODBC trace file}, to find out what could be wrong. You should be able to find out the issued queries by searching after the string @code{>mysql_real_query} in the @file{myodbc.log} file. You should also try duplicating the queries in the @code{mysql} monitor or @code{admndemo} to find out if the error is MyODBC or MySQL. If you find out something is wrong, please only send the relevant rows (max 40 rows) to @email{myodbc@@lists.mysql.com}. Please never send the whole MyODBC or ODBC log file! If you are unable to find out what's wrong, the last option is to make an archive (tar or zip) that contains a MyODBC trace file, the ODBC log file, and a README file that explains the problem. You can send this to @uref{ftp://support.mysql.com/pub/mysql/secret}. Only we at MySQL AB will have access to the files you upload, and we will be very discrete with the data! If you can create a program that also shows this problem, please upload this too! If the program works with some other SQL server, you should make an ODBC log file where you do exactly the same thing in the other SQL server. Remember that the more information you can supply to us, the more likely it is that we can fix the problem! @node C, Cplusplus, ODBC, Clients @section MySQL C API @cindex C API, datatypes @cindex datatypes, C API @menu * C API datatypes:: C API Datatypes * C API function overview:: C API Function Overview * C API functions:: C API Function Descriptions * C Thread functions:: * C API problems:: Common questions and problems when using the C API * Building clients:: Building Client Programs * Threaded clients:: How to Make a Threaded Client @end menu The C API code is distributed with MySQL. It is included in the @code{mysqlclient} library and allows C programs to access a database. Many of the clients in the MySQL source distribution are written in C. If you are looking for examples that demonstrate how to use the C API, take a look at these clients. You can find these in the @code{clients} directory in the MySQL source distribution. Most of the other client APIs (all except Java) use the @code{mysqlclient} library to communicate with the MySQL server. This means that, for example, you can take advantage of many of the same environment variables that are used by other client programs, because they are referenced from the library. See @ref{Client-Side Scripts}, for a list of these variables. The client has a maximum communication buffer size. The size of the buffer that is allocated initially (16K bytes) is automatically increased up to the maximum size (the maximum is 16M). Because buffer sizes are increased only as demand warrants, simply increasing the default maximum limit does not in itself cause more resources to be used. This size check is mostly a check for erroneous queries and communication packets. The communication buffer must be large enough to contain a single SQL statement (for client-to-server traffic) and one row of returned data (for server-to-client traffic). Each thread's communication buffer is dynamically enlarged to handle any query or row up to the maximum limit. For example, if you have @code{BLOB} values that contain up to 16M of data, you must have a communication buffer limit of at least 16M (in both server and client). The client's default maximum is 16M, but the default maximum in the server is 1M. You can increase this by changing the value of the @code{max_allowed_packet} parameter when the server is started. @xref{Server parameters}. The MySQL server shrinks each communication buffer to @code{net_buffer_length} bytes after each query. For clients, the size of the buffer associated with a connection is not decreased until the connection is closed, at which time client memory is reclaimed. For programming with threads, consult the 'how to make a thread-safe client' chapter. @xref{Threaded clients}. @node C API datatypes, C API function overview, C, C @subsection C API Datatypes @table @code @tindex MYSQL C type @item MYSQL This structure represents a handle to one database connection. It is used for almost all MySQL functions. @tindex MYSQL_RES C type @item MYSQL_RES This structure represents the result of a query that returns rows (@code{SELECT}, @code{SHOW}, @code{DESCRIBE}, @code{EXPLAIN}). The information returned from a query is called the @emph{result set} in the remainder of this section. @tindex MYSQL_ROW C type @item MYSQL_ROW This is a type-safe representation of one row of data. It is currently implemented as an array of counted byte strings. (You cannot treat these as null-terminated strings if field values may contain binary data, because such values may contain null bytes internally.) Rows are obtained by calling @code{mysql_fetch_row()}. @tindex MYSQL_FIELD C type @item MYSQL_FIELD This structure contains information about a field, such as the field's name, type, and size. Its members are described in more detail below. You may obtain the @code{MYSQL_FIELD} structures for each field by calling @code{mysql_fetch_field()} repeatedly. Field values are not part of this structure; they are contained in a @code{MYSQL_ROW} structure. @tindex MYSQL_FIELD_OFFSET C type @item MYSQL_FIELD_OFFSET This is a type-safe representation of an offset into a MySQL field list. (Used by @code{mysql_field_seek()}.) Offsets are field numbers within a row, beginning at zero. @tindex my_ulonglong C type @tindex my_ulonglong values, printing @item my_ulonglong The type used for the number of rows and for @code{mysql_affected_rows()}, @code{mysql_num_rows()}, and @code{mysql_insert_id()}. This type provides a range of @code{0} to @code{1.84e19}. On some systems, attempting to print a value of type @code{my_ulonglong} will not work. To print such a value, convert it to @code{unsigned long} and use a @code{%lu} print format. Example: @example printf (Number of rows: %lu\n", (unsigned long) mysql_num_rows(result)); @end example @end table @noindent The @code{MYSQL_FIELD} structure contains the members listed below: @table @code @item char * name The name of the field, as a null-terminated string. @item char * table The name of the table containing this field, if it isn't a calculated field. For calculated fields, the @code{table} value is an empty string. @item char * def The default value of this field, as a null-terminated string. This is set only if you use @code{mysql_list_fields()}. @item enum enum_field_types type The type of the field. The @code{type} value may be one of the following: @multitable @columnfractions .3 .55 @item @strong{Type value} @tab @strong{Type meaning} @item @code{FIELD_TYPE_TINY} @tab @code{TINYINT} field @item @code{FIELD_TYPE_SHORT} @tab @code{SMALLINT} field @item @code{FIELD_TYPE_LONG} @tab @code{INTEGER} field @item @code{FIELD_TYPE_INT24} @tab @code{MEDIUMINT} field @item @code{FIELD_TYPE_LONGLONG} @tab @code{BIGINT} field @item @code{FIELD_TYPE_DECIMAL} @tab @code{DECIMAL} or @code{NUMERIC} field @item @code{FIELD_TYPE_FLOAT} @tab @code{FLOAT} field @item @code{FIELD_TYPE_DOUBLE} @tab @code{DOUBLE} or @code{REAL} field @item @code{FIELD_TYPE_TIMESTAMP} @tab @code{TIMESTAMP} field @item @code{FIELD_TYPE_DATE} @tab @code{DATE} field @item @code{FIELD_TYPE_TIME} @tab @code{TIME} field @item @code{FIELD_TYPE_DATETIME} @tab @code{DATETIME} field @item @code{FIELD_TYPE_YEAR} @tab @code{YEAR} field @item @code{FIELD_TYPE_STRING} @tab String (@code{CHAR} or @code{VARCHAR}) field @item @code{FIELD_TYPE_BLOB} @tab @code{BLOB} or @code{TEXT} field (use @code{max_length} to determine the maximum length) @item @code{FIELD_TYPE_SET} @tab @code{SET} field @item @code{FIELD_TYPE_ENUM} @tab @code{ENUM} field @item @code{FIELD_TYPE_NULL} @tab @code{NULL}-type field @item @code{FIELD_TYPE_CHAR} @tab Deprecated; use @code{FIELD_TYPE_TINY} instead @end multitable You can use the @code{IS_NUM()} macro to test whether or not a field has a numeric type. Pass the @code{type} value to @code{IS_NUM()} and it will evaluate to TRUE if the field is numeric: @example if (IS_NUM(field->type)) printf("Field is numeric\n"); @end example @item unsigned int length The width of the field, as specified in the table definition. @item unsigned int max_length The maximum width of the field for the result set (the length of the longest field value for the rows actually in the result set). If you use @code{mysql_store_result()} or @code{mysql_list_fields()}, this contains the maximum length for the field. If you use @code{mysql_use_result()}, the value of this variable is zero. @item unsigned int flags Different bit-flags for the field. The @code{flags} value may have zero or more of the following bits set: @multitable @columnfractions .3 .55 @item @strong{Flag value} @tab @strong{Flag meaning} @item @code{NOT_NULL_FLAG} @tab Field can't be @code{NULL} @item @code{PRI_KEY_FLAG} @tab Field is part of a primary key @item @code{UNIQUE_KEY_FLAG} @tab Field is part of a unique key @item @code{MULTIPLE_KEY_FLAG} @tab Field is part of a non-unique key @item @code{UNSIGNED_FLAG} @tab Field has the @code{UNSIGNED} attribute @item @code{ZEROFILL_FLAG} @tab Field has the @code{ZEROFILL} attribute @item @code{BINARY_FLAG} @tab Field has the @code{BINARY} attribute @item @code{AUTO_INCREMENT_FLAG} @tab Field has the @code{AUTO_INCREMENT} attribute @item @code{ENUM_FLAG} @tab Field is an @code{ENUM} (deprecated) @item @code{BLOB_FLAG} @tab Field is a @code{BLOB} or @code{TEXT} (deprecated) @item @code{TIMESTAMP_FLAG} @tab Field is a @code{TIMESTAMP} (deprecated) @end multitable Use of the @code{BLOB_FLAG}, @code{ENUM_FLAG}, and @code{TIMESTAMP_FLAG} flags is deprecated because they indicate the type of a field rather than an attribute of its type. It is preferable to test @code{field->type} against @code{FIELD_TYPE_BLOB}, @code{FIELD_TYPE_ENUM}, or @code{FIELD_TYPE_TIMESTAMP} instead. @noindent The example below illustrates a typical use of the @code{flags} value: @example if (field->flags & NOT_NULL_FLAG) printf("Field can't be null\n"); @end example You may use the following convenience macros to determine the boolean status of the @code{flags} value: @multitable @columnfractions .3 .5 @item @code{IS_NOT_NULL(flags)} @tab True if this field is defined as @code{NOT NULL} @item @code{IS_PRI_KEY(flags)} @tab True if this field is a primary key @item @code{IS_BLOB(flags)} @tab True if this field is a @code{BLOB} or @code{TEXT} (deprecated; test @code{field->type} instead) @end multitable @item unsigned int decimals The number of decimals for numeric fields. @end table @node C API function overview, C API functions, C API datatypes, C @subsection C API Function Overview @cindex C API, functions @cindex functions, C API The functions available in the C API are listed below and are described in greater detail in the next section. @xref{C API functions}. @multitable @columnfractions .3 .7 @item @strong{mysql_affected_rows()} @tab Returns the number of rows changed/deleted/inserted by the last @code{UPDATE}, @code{DELETE}, or @code{INSERT} query. @item @strong{mysql_close()} @tab Closes a server connection. @item @strong{mysql_connect()} @tab Connects to a MySQL server. This function is deprecated; use @code{mysql_real_connect()} instead. @item @strong{mysql_change_user()} @tab Changes user and database on an open connection. @item @strong{mysql_character_set_name()} @tab Returns the name of the default character set for the connection. @item @strong{mysql_create_db()} @tab Creates a database. This function is deprecated; use the SQL command @code{CREATE DATABASE} instead. @item @strong{mysql_data_seek()} @tab Seeks to an arbitrary row in a query result set. @item @strong{mysql_debug()} @tab Does a @code{DBUG_PUSH} with the given string. @item @strong{mysql_drop_db()} @tab Drops a database. This function is deprecated; use the SQL command @code{DROP DATABASE} instead. @item @strong{mysql_dump_debug_info()} @tab Makes the server write debug information to the log. @item @strong{mysql_eof()} @tab Determines whether or not the last row of a result set has been read. This function is deprecated; @code{mysql_errno()} or @code{mysql_error()} may be used instead. @item @strong{mysql_errno()} @tab Returns the error number for the most recently invoked MySQL function. @item @strong{mysql_error()} @tab Returns the error message for the most recently invoked MySQL function. @item @strong{mysql_real_escape_string()} @tab Escapes special characters in a string for use in a SQL statement taking into account the current charset of the connection. @item @strong{mysql_escape_string()} @tab Escapes special characters in a string for use in a SQL statement. @item @strong{mysql_fetch_field()} @tab Returns the type of the next table field. @item @strong{mysql_fetch_field_direct()} @tab Returns the type of a table field, given a field number. @item @strong{mysql_fetch_fields()} @tab Returns an array of all field structures. @item @strong{mysql_fetch_lengths()} @tab Returns the lengths of all columns in the current row. @item @strong{mysql_fetch_row()} @tab Fetches the next row from the result set. @item @strong{mysql_field_seek()} @tab Puts the column cursor on a specified column. @item @strong{mysql_field_count()} @tab Returns the number of result columns for the most recent query. @item @strong{mysql_field_tell()} @tab Returns the position of the field cursor used for the last @code{mysql_fetch_field()}. @item @strong{mysql_free_result()} @tab Frees memory used by a result set. @item @strong{mysql_get_client_info()} @tab Returns client version information. @item @strong{mysql_get_host_info()} @tab Returns a string describing the connection. @item @strong{mysql_get_proto_info()} @tab Returns the protocol version used by the connection. @item @strong{mysql_get_server_info()} @tab Returns the server version number. @item @strong{mysql_info()} @tab Returns information about the most recently executed query. @item @strong{mysql_init()} @tab Gets or initializes a @code{MYSQL} structure. @item @strong{mysql_insert_id()} @tab Returns the ID generated for an @code{AUTO_INCREMENT} column by the previous query. @item @strong{mysql_kill()} @tab Kills a given thread. @item @strong{mysql_list_dbs()} @tab Returns database names matching a simple regular expression. @item @strong{mysql_list_fields()} @tab Returns field names matching a simple regular expression. @item @strong{mysql_list_processes()} @tab Returns a list of the current server threads. @item @strong{mysql_list_tables()} @tab Returns table names matching a simple regular expression. @item @strong{mysql_num_fields()} @tab Returns the number of columns in a result set. @item @strong{mysql_num_rows()} @tab Returns the number of rows in a result set. @item @strong{mysql_options()} @tab Sets connect options for @code{mysql_connect()}. @item @strong{mysql_ping()} @tab Checks whether or not the connection to the server is working, reconnecting as necessary. @item @strong{mysql_query()} @tab Executes a SQL query specified as a null-terminated string. @item @strong{mysql_real_connect()} @tab Connects to a MySQL server. @item @strong{mysql_real_query()} @tab Executes a SQL query specified as a counted string. @item @strong{mysql_reload()} @tab Tells the server to reload the grant tables. @item @strong{mysql_row_seek()} @tab Seeks to a row in a result set, using value returned from @code{mysql_row_tell()}. @item @strong{mysql_row_tell()} @tab Returns the row cursor position. @item @strong{mysql_select_db()} @tab Selects a database. @item @strong{mysql_shutdown()} @tab Shuts down the database server. @item @strong{mysql_stat()} @tab Returns the server status as a string. @item @strong{mysql_store_result()} @tab Retrieves a complete result set to the client. @item @strong{mysql_thread_id()} @tab Returns the current thread ID. @item @strong{mysql_thread_save()} @tab Returns 1 if the clients are compiled as thread-safe. @item @strong{mysql_use_result()} @tab Initiates a row-by-row result set retrieval. @end multitable To connect to the server, call @code{mysql_init()} to initialize a connection handler, then call @code{mysql_real_connect()} with that handler (along with other information such as the hostname, user name, and password). Upon connection, @code{mysql_real_connect()} sets the @code{reconnect} flag (part of the MYSQL structure) to a value of @code{1}. This flag indicates, in the event that a query cannot be performed because of a lost connection, to try reconnecting to the server before giving up. When you are done with the connection, call @code{mysql_close()} to terminate it. While a connection is active, the client may send SQL queries to the server using @code{mysql_query()} or @code{mysql_real_query()}. The difference between the two is that @code{mysql_query()} expects the query to be specified as a null-terminated string whereas @code{mysql_real_query()} expects a counted string. If the string contains binary data (which may include null bytes), you must use @code{mysql_real_query()}. For each non-@code{SELECT} query (for example, @code{INSERT}, @code{UPDATE}, @code{DELETE}), you can find out how many rows were changed (affected) by calling @code{mysql_affected_rows()}. For @code{SELECT} queries, you retrieve the selected rows as a result set. (Note that some statements are @code{SELECT}-like in that they return rows. These include @code{SHOW}, @code{DESCRIBE}, and @code{EXPLAIN}. They should be treated the same way as @code{SELECT} statements.) There are two ways for a client to process result sets. One way is to retrieve the entire result set all at once by calling @code{mysql_store_result()}. This function acquires from the server all the rows returned by the query and stores them in the client. The second way is for the client to initiate a row-by-row result set retrieval by calling @code{mysql_use_result()}. This function initializes the retrieval, but does not actually get any rows from the server. In both cases, you access rows by calling @code{mysql_fetch_row()}. With @code{mysql_store_result()}, @code{mysql_fetch_row()} accesses rows that have already been fetched from the server. With @code{mysql_use_result()}, @code{mysql_fetch_row()} actually retrieves the row from the server. Information about the size of the data in each row is available by calling @code{mysql_fetch_lengths()}. After you are done with a result set, call @code{mysql_free_result()} to free the memory used for it. The two retrieval mechanisms are complementary. Client programs should choose the approach that is most appropriate for their requirements. In practice, clients tend to use @code{mysql_store_result()} more commonly. An advantage of @code{mysql_store_result()} is that because the rows have all been fetched to the client, you not only can access rows sequentially, you can move back and forth in the result set using @code{mysql_data_seek()} or @code{mysql_row_seek()} to change the current row position within the result set. You can also find out how many rows there are by calling @code{mysql_num_rows()}. On the other hand, the memory requirements for @code{mysql_store_result()} may be very high for large result sets and you are more likely to encounter out-of-memory conditions. An advantage of @code{mysql_use_result()} is that the client requires less memory for the result set because it maintains only one row at a time (and because there is less allocation overhead, @code{mysql_use_result()} can be faster). Disadvantages are that you must process each row quickly to avoid tying up the server, you don't have random access to rows within the result set (you can only access rows sequentially), and you don't know how many rows are in the result set until you have retrieved them all. Furthermore, you @emph{must} retrieve all the rows even if you determine in mid-retrieval that you've found the information you were looking for. The API makes it possible for clients to respond appropriately to queries (retrieving rows only as necessary) without knowing whether or not the query is a @code{SELECT}. You can do this by calling @code{mysql_store_result()} after each @code{mysql_query()} (or @code{mysql_real_query()}). If the result set call succeeds, the query was a @code{SELECT} and you can read the rows. If the result set call fails, call @code{mysql_field_count()} to determine whether or not a result was actually to be expected. If @code{mysql_field_count()} returns zero, the query returned no data (indicating that it was an @code{INSERT}, @code{UPDATE}, @code{DELETE}, etc.), and was not expected to return rows. If @code{mysql_field_count()} is non-zero, the query should have returned rows, but didn't. This indicates that the query was a @code{SELECT} that failed. See the description for @code{mysql_field_count()} for an example of how this can be done. Both @code{mysql_store_result()} and @code{mysql_use_result()} allow you to obtain information about the fields that make up the result set (the number of fields, their names and types, etc.). You can access field information sequentially within the row by calling @code{mysql_fetch_field()} repeatedly, or by field number within the row by calling @code{mysql_fetch_field_direct()}. The current field cursor position may be changed by calling @code{mysql_field_seek()}. Setting the field cursor affects subsequent calls to @code{mysql_fetch_field()}. You can also get information for fields all at once by calling @code{mysql_fetch_fields()}. For detecting and reporting errors, MySQL provides access to error information by means of the @code{mysql_errno()} and @code{mysql_error()} functions. These return the error code or error message for the most recently invoked function that can succeed or fail, allowing you to determine when an error occurred and what it was. @node C API functions, C Thread functions, C API function overview, C @subsection C API Function Descriptions @menu * mysql_affected_rows:: @code{mysql_affected_rows()} * mysql_close:: @code{mysql_close()} * mysql_connect:: @code{mysql_connect()} * mysql_change_user:: @code{mysql_change_user()} * mysql_character_set_name:: @code{mysql_character_set_name()} * mysql_create_db:: @code{mysql_create_db()} * mysql_data_seek:: @code{mysql_data_seek()} * mysql_debug:: @code{mysql_debug()} * mysql_drop_db:: @code{mysql_drop_db()} * mysql_dump_debug_info:: @code{mysql_dump_debug_info()} * mysql_eof:: @code{mysql_eof()} * mysql_errno:: @code{mysql_errno()} * mysql_error:: @code{mysql_error()} * mysql_escape_string:: @code{mysql_escape_string()} * mysql_fetch_field:: @code{mysql_fetch_field()} * mysql_fetch_fields:: @code{mysql_fetch_fields()} * mysql_fetch_field_direct:: @code{mysql_fetch_field_direct()} * mysql_fetch_lengths:: @code{mysql_fetch_lengths()} * mysql_fetch_row:: @code{mysql_fetch_row()} * mysql_field_count:: @code{mysql_field_count()} * mysql_field_seek:: @code{mysql_field_seek()} * mysql_field_tell:: @code{mysql_field_tell()} * mysql_free_result:: @code{mysql_free_result()} * mysql_get_client_info:: @code{mysql_get_client_info()} * mysql_get_host_info:: @code{mysql_get_host_info()} * mysql_get_proto_info:: @code{mysql_get_proto_info()} * mysql_get_server_info:: @code{mysql_get_server_info()} * mysql_info:: @code{mysql_info()} * mysql_init:: @code{mysql_init()} * mysql_insert_id:: @code{mysql_insert_id()} * mysql_kill:: @code{mysql_kill()} * mysql_list_dbs:: @code{mysql_list_dbs()} * mysql_list_fields:: @code{mysql_list_fields()} * mysql_list_processes:: @code{mysql_list_processes()} * mysql_list_tables:: @code{mysql_list_tables()} * mysql_num_fields:: @code{mysql_num_fields()} * mysql_num_rows:: @code{mysql_num_rows()} * mysql_options:: @code{mysql_options()} * mysql_ping:: @code{mysql_ping()} * mysql_query:: @code{mysql_query()} * mysql_real_connect:: @code{mysql_real_connect()} * mysql_real_escape_string:: @code{mysql_real_escape_string()} * mysql_real_query:: @code{mysql_real_query()} * mysql_reload:: @code{mysql_reload()} * mysql_row_seek:: @code{mysql_row_seek()} * mysql_row_tell:: @code{mysql_row_tell()} * mysql_select_db:: @code{mysql_select_db()} * mysql_shutdown:: @code{mysql_shutdown()} * mysql_stat:: @code{mysql_stat()} * mysql_store_result:: @code{mysql_store_result()} * mysql_thread_id:: @code{mysql_thread_id()} * mysql_use_result:: @code{mysql_use_result()} @end menu In the descriptions below, a parameter or return value of @code{NULL} means @code{NULL} in the sense of the C programming language, not a MySQL @code{NULL} value. Functions that return a value generally return a pointer or an integer. Unless specified otherwise, functions returning a pointer return a non-@code{NULL} value to indicate success or a @code{NULL} value to indicate an error, and functions returning an integer return zero to indicate success or non-zero to indicate an error. Note that ``non-zero'' means just that. Unless the function description says otherwise, do not test against a value other than zero: @example if (result) /* correct */ ... error ... if (result < 0) /* incorrect */ ... error ... if (result == -1) /* incorrect */ ... error ... @end example When a function returns an error, the @strong{Errors} subsection of the function description lists the possible types of errors. You can find out which of these occurred by calling @code{mysql_errno()}. A string representation of the error may be obtained by calling @code{mysql_error()}. @node mysql_affected_rows, mysql_close, C API functions, C API functions @subsubsection @code{mysql_affected_rows()} @findex @code{mysql_affected_rows()} @code{my_ulonglong mysql_affected_rows(MYSQL *mysql)} @subsubheading Description Returns the number of rows changed by the last @code{UPDATE}, deleted by the last @code{DELETE} or inserted by the last @code{INSERT} statement. May be called immediately after @code{mysql_query()} for @code{UPDATE}, @code{DELETE}, or @code{INSERT} statements. For @code{SELECT} statements, @code{mysql_affected_rows()} works like @code{mysql_num_rows()}. @subsubheading Return Values An integer greater than zero indicates the number of rows affected or retrieved. Zero indicates that no records where updated for an @code{UPDATE} statement, no rows matched the @code{WHERE} clause in the query or that no query has yet been executed. -1 indicates that the query returned an error or that, for a @code{SELECT} query, @code{mysql_affected_rows()} was called prior to calling @code{mysql_store_result()}. @subsubheading Errors None. @subsubheading Example @example mysql_query(&mysql,"UPDATE products SET cost=cost*1.25 WHERE group=10"); printf("%ld products updated",(long) mysql_affected_rows(&mysql)); @end example If one specifies the flag @code{CLIENT_FOUND_ROWS} when connecting to @code{mysqld}, @code{mysql_affected_rows()} will return the number of rows matched by the @code{WHERE} statement for @code{UPDATE} statements. Note that when one uses a @code{REPLACE} command, @code{mysql_affected_rows()} will return 2 if the new row replaced and old row. This is because in this case one row was inserted and then the duplicate was deleted. @node mysql_close, mysql_connect, mysql_affected_rows, C API functions @subsubsection @code{mysql_close()} @findex @code{mysql_close()} @code{void mysql_close(MYSQL *mysql)} @subsubheading Description Closes a previously opened connection. @code{mysql_close()} also deallocates the connection handle pointed to by @code{mysql} if the handle was allocated automatically by @code{mysql_init()} or @code{mysql_connect()}. @subsubheading Return Values None. @subsubheading Errors None. @node mysql_connect, mysql_change_user, mysql_close, C API functions @subsubsection @code{mysql_connect()} @findex @code{mysql_connect()} @code{MYSQL *mysql_connect(MYSQL *mysql, const char *host, const char *user, const char *passwd)} @subsubheading Description This function is deprecated. It is preferable to use @code{mysql_real_connect()} instead. @code{mysql_connect()} attempts to establish a connection to a MySQL database engine running on @code{host}. @code{mysql_connect()} must complete successfully before you can execute any of the other API functions, with the exception of @code{mysql_get_client_info()}. The meanings of the parameters are the same as for the corresponding parameters for @code{mysql_real_connect()} with the difference that the connection parameter may be @code{NULL}. In this case the C API allocates memory for the connection structure automatically and frees it when you call @code{mysql_close()}. The disadvantage of this approach is that you can't retrieve an error message if the connection fails. (To get error information from @code{mysql_errno()} or @code{mysql_error()}, you must provide a valid @code{MYSQL} pointer.) @subsubheading Return Values Same as for @code{mysql_real_connect()}. @subsubheading Errors Same as for @code{mysql_real_connect()}. @node mysql_change_user, mysql_character_set_name, mysql_connect, C API functions @subsubsection @code{mysql_change_user()} @findex @code{mysql_change_user()} @code{my_bool mysql_change_user(MYSQL *mysql, const char *user, const char *password, const char *db)} @subsubheading Description Changes the user and causes the database specified by @code{db} to become the default (current) database on the connection specified by @code{mysql}. In subsequent queries, this database is the default for table references that do not include an explicit database specifier. This function was introduced in MySQL Version 3.23.3. @code{mysql_change_user()} fails unless the connected user can be authenticated or if he doesn't have permission to use the database. In this case the user and database are not changed The @code{db} parameter may be set to @code{NULL} if you don't want to have a default database. @subsubheading Return values Zero for success. Non-zero if an error occurred. @subsubheading Errors The same that you can get from @code{mysql_real_connect()}. @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @item ER_UNKNOWN_COM_ERROR The MySQL server doesn't implement this command (probably an old server) @item ER_ACCESS_DENIED_ERROR The user or password was wrong. @item ER_BAD_DB_ERROR The database didn't exist. @item ER_DBACCESS_DENIED_ERROR The user did not have access rights to the database. @item ER_WRONG_DB_NAME The database name was too long. @end table @subsubheading Example @example if (mysql_change_user(&mysql, "user", "password", "new_database")) @{ fprintf(stderr, "Failed to change user. Error: %s\n", mysql_error(&mysql)); @} @end example @node mysql_character_set_name, mysql_create_db, mysql_change_user, C API functions @subsubsection @code{mysql_character_set_name()} @findex @code{mysql_character_set_name()} @code{const char *mysql_character_set_name(MYSQL *mysql)} @subsubheading Description Returns the default character set for the current connection. @subsubheading Return Values The default character set @subsubheading Errors None. @node mysql_create_db, mysql_data_seek, mysql_character_set_name, C API functions @subsubsection @code{mysql_create_db()} @findex @code{mysql_create_db()} @code{int mysql_create_db(MYSQL *mysql, const char *db)} @subsubheading Description Creates the database named by the @code{db} parameter. This function is deprecated. It is preferable to use @code{mysql_query()} to issue a SQL @code{CREATE DATABASE} statement instead. @subsubheading Return Values Zero if the database was created successfully. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @subsubheading Example @example if(mysql_create_db(&mysql, "my_database")) @{ fprintf(stderr, "Failed to create new database. Error: %s\n", mysql_error(&mysql)); @} @end example @node mysql_data_seek, mysql_debug, mysql_create_db, C API functions @subsubsection @code{mysql_data_seek()} @findex @code{mysql_data_seek()} @code{void mysql_data_seek(MYSQL_RES *result, unsigned long long offset)} @subsubheading Description Seeks to an arbitrary row in a query result set. This requires that the result set structure contains the entire result of the query, so @code{mysql_data_seek()} may be used in conjunction only with @code{mysql_store_result()}, not with @code{mysql_use_result()}. The offset should be a value in the range from 0 to @code{mysql_num_rows(result)-1}. @subsubheading Return Values None. @subsubheading Errors None. @node mysql_debug, mysql_drop_db, mysql_data_seek, C API functions @subsubsection @code{mysql_debug()} @findex @code{mysql_debug()} @code{void mysql_debug(char *debug)} @subsubheading Description Does a @code{DBUG_PUSH} with the given string. @code{mysql_debug()} uses the Fred Fish debug library. To use this function, you must compile the client library to support debugging. @xref{Debugging server}. @xref{Debugging client}. @subsubheading Return Values None. @subsubheading Errors None. @subsubheading Example The call shown below causes the client library to generate a trace file in @file{/tmp/client.trace} on the client machine: @example mysql_debug("d:t:O,/tmp/client.trace"); @end example @node mysql_drop_db, mysql_dump_debug_info, mysql_debug, C API functions @subsubsection @code{mysql_drop_db()} @findex @code{mysql_drop_db()} @code{int mysql_drop_db(MYSQL *mysql, const char *db)} @subsubheading Description Drops the database named by the @code{db} parameter. This function is deprecated. It is preferable to use @code{mysql_query()} to issue a SQL @code{DROP DATABASE} statement instead. @subsubheading Return Values Zero if the database was dropped successfully. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @subsubheading Example @example if(mysql_drop_db(&mysql, "my_database")) fprintf(stderr, "Failed to drop the database: Error: %s\n", mysql_error(&mysql)); @end example @node mysql_dump_debug_info, mysql_eof, mysql_drop_db, C API functions @subsubsection @code{mysql_dump_debug_info()} @findex @code{mysql_dump_debug_info()} @code{int mysql_dump_debug_info(MYSQL *mysql)} @subsubheading Description Instructs the server to write some debug information to the log. The connected user must have the @strong{process} privilege for this to work. @subsubheading Return values Zero if the command was successful. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_eof, mysql_errno, mysql_dump_debug_info, C API functions @subsubsection @code{mysql_eof()} @findex @code{mysql_eof()} @code{my_bool mysql_eof(MYSQL_RES *result)} @subsubheading Description This function is deprecated. @code{mysql_errno()} or @code{mysql_error()} may be used instead. @code{mysql_eof()} determines whether or not the last row of a result set has been read. If you acquire a result set from a successful call to @code{mysql_store_result()}, the client receives the entire set in one operation. In this case, a @code{NULL} return from @code{mysql_fetch_row()} always means the end of the result set has been reached and it is unnecessary to call @code{mysql_eof()}. On the other hand, if you use @code{mysql_use_result()} to initiate a result set retrieval, the rows of the set are obtained from the server one by one as you call @code{mysql_fetch_row()} repeatedly. Because an error may occur on the connection during this process, a @code{NULL} return value from @code{mysql_fetch_row()} does not necessarily mean the end of the result set was reached normally. In this case, you can use @code{mysql_eof()} to determine what happened. @code{mysql_eof()} returns a non-zero value if the end of the result set was reached and zero if an error occurred. Historically, @code{mysql_eof()} predates the standard MySQL error functions @code{mysql_errno()} and @code{mysql_error()}. Because those error functions provide the same information, their use is preferred over @code{mysql_eof()}, which is now deprecated. (In fact, they provide more information, because @code{mysql_eof()} returns only a boolean value whereas the error functions indicate a reason for the error when one occurs.) @subsubheading Return Values Zero if no error occurred. Non-zero if the end of the result set has been reached. @subsubheading Errors None. @subsubheading Example The following example shows how you might use @code{mysql_eof()}: @example mysql_query(&mysql,"SELECT * FROM some_table"); result = mysql_use_result(&mysql); while((row = mysql_fetch_row(result))) @{ // do something with data @} if(!mysql_eof(result)) // mysql_fetch_row() failed due to an error @{ fprintf(stderr, "Error: %s\n", mysql_error(&mysql)); @} @end example However, you can achieve the same effect with the standard MySQL error functions: @example mysql_query(&mysql,"SELECT * FROM some_table"); result = mysql_use_result(&mysql); while((row = mysql_fetch_row(result))) @{ // do something with data @} if(mysql_errno(&mysql)) // mysql_fetch_row() failed due to an error @{ fprintf(stderr, "Error: %s\n", mysql_error(&mysql)); @} @end example @node mysql_errno, mysql_error, mysql_eof, C API functions @subsubsection @code{mysql_errno()} @findex @code{mysql_errno()} @code{unsigned int mysql_errno(MYSQL *mysql)} @subsubheading Description For the connection specified by @code{mysql}, @code{mysql_errno()} returns the error code for the most recently invoked API function that can succeed or fail. A return value of zero means that no error occurred. Client error message numbers are listed in the MySQL @file{errmsg.h} header file. Server error message numbers are listed in @file{mysqld_error.h}. In the MySQL source distribution you can find a complete list of error messages and error numbers in the file @file{Docs/mysqld_error.txt}. @subsubheading Return Values An error code value. Zero if no error occurred. @subsubheading Errors None. @node mysql_error, mysql_escape_string, mysql_errno, C API functions @subsubsection @code{mysql_error()} @findex @code{mysql_error()} @code{char *mysql_error(MYSQL *mysql)} @subsubheading Description For the connection specified by @code{mysql}, @code{mysql_error()} returns the error message for the most recently invoked API function that can succeed or fail. An empty string (@code{""}) is returned if no error occurred. This means the following two tests are equivalent: @example if(mysql_errno(&mysql)) @{ // an error occurred @} if(mysql_error(&mysql)[0] != '\0') @{ // an error occurred @} @end example The language of the client error messages may be changed by recompiling the MySQL client library. Currently you can choose error messages in several different languages. @xref{Languages}. @subsubheading Return Values A character string that describes the error. An empty string if no error occurred. @subsubheading Errors None. @node mysql_escape_string, mysql_fetch_field, mysql_error, C API functions @subsubsection @code{mysql_escape_string()} @findex @code{mysql_escape_string()} You should use @code{mysql_real_escape_string()} instead! This is identical to @code{mysql_real_escape_string()} except that it takes the connection as the first argument. @code{mysql_real_escape_string()} will escape the string according to the current character set while @code{mysql_escape_string()} does not respect the current charset setting. @node mysql_fetch_field, mysql_fetch_fields, mysql_escape_string, C API functions @subsubsection @code{mysql_fetch_field()} @findex @code{mysql_fetch_field()} @code{MYSQL_FIELD *mysql_fetch_field(MYSQL_RES *result)} @subsubheading Description Returns the definition of one column of a result set as a @code{MYSQL_FIELD} structure. Call this function repeatedly to retrieve information about all columns in the result set. @code{mysql_fetch_field()} returns @code{NULL} when no more fields are left. @code{mysql_fetch_field()} is reset to return information about the first field each time you execute a new @code{SELECT} query. The field returned by @code{mysql_fetch_field()} is also affected by calls to @code{mysql_field_seek()}. If you've called @code{mysql_query()} to perform a @code{SELECT} on a table but have not called @code{mysql_store_result()}, MySQL returns the default blob length (8K bytes) if you call @code{mysql_fetch_field()} to ask for the length of a @code{BLOB} field. (The 8K size is chosen because MySQL doesn't know the maximum length for the @code{BLOB}. This should be made configurable sometime.) Once you've retrieved the result set, @code{field->max_length} contains the length of the largest value for this column in the specific query. @subsubheading Return Values The @code{MYSQL_FIELD} structure for the current column. @code{NULL} if no columns are left. @subsubheading Errors None. @subsubheading Example @example MYSQL_FIELD *field; while((field = mysql_fetch_field(result))) @{ printf("field name %s\n", field->name); @} @end example @node mysql_fetch_fields, mysql_fetch_field_direct, mysql_fetch_field, C API functions @subsubsection @code{mysql_fetch_fields()} @findex @code{mysql_fetch_fields()} @code{MYSQL_FIELD *mysql_fetch_fields(MYSQL_RES *result)} @subsubheading Description Returns an array of all @code{MYSQL_FIELD} structures for a result set. Each structure provides the field definition for one column of the result set. @subsubheading Return Values An array of @code{MYSQL_FIELD} structures for all columns of a result set. @subsubheading Errors None. @subsubheading Example @example unsigned int num_fields; unsigned int i; MYSQL_FIELD *fields; num_fields = mysql_num_fields(result); fields = mysql_fetch_fields(result); for(i = 0; i < num_fields; i++) @{ printf("Field %u is %s\n", i, fields[i].name); @} @end example @node mysql_fetch_field_direct, mysql_fetch_lengths, mysql_fetch_fields, C API functions @subsubsection @code{mysql_fetch_field_direct()} @findex @code{mysql_fetch_field_direct()} @code{MYSQL_FIELD *mysql_fetch_field_direct(MYSQL_RES *result, unsigned int fieldnr)} @subsubheading Description Given a field number @code{fieldnr} for a column within a result set, returns that column's field definition as a @code{MYSQL_FIELD} structure. You may use this function to retrieve the definition for an arbitrary column. The value of @code{fieldnr} should be in the range from 0 to @code{mysql_num_fields(result)-1}. @subsubheading Return Values The @code{MYSQL_FIELD} structure for the specified column. @subsubheading Errors None. @subsubheading Example @example unsigned int num_fields; unsigned int i; MYSQL_FIELD *field; num_fields = mysql_num_fields(result); for(i = 0; i < num_fields; i++) @{ field = mysql_fetch_field_direct(result, i); printf("Field %u is %s\n", i, field->name); @} @end example @node mysql_fetch_lengths, mysql_fetch_row, mysql_fetch_field_direct, C API functions @subsubsection @code{mysql_fetch_lengths()} @findex @code{mysql_fetch_lengths()} @code{unsigned long *mysql_fetch_lengths(MYSQL_RES *result)} @subsubheading Description Returns the lengths of the columns of the current row within a result set. If you plan to copy field values, this length information is also useful for optimization, because you can avoid calling @code{strlen()}. In addition, if the result set contains binary data, you @emph{must} use this function to determine the size of the data, because @code{strlen()} returns incorrect results for any field containing null characters. The length for empty columns and for columns containing @code{NULL} values is zero. To see how to distinguish these two cases, see the description for @code{mysql_fetch_row()}. @subsubheading Return Values An array of unsigned long integers representing the size of each column (not including any terminating null characters). @code{NULL} if an error occurred. @subsubheading Errors @code{mysql_fetch_lengths()} is valid only for the current row of the result set. It returns @code{NULL} if you call it before calling @code{mysql_fetch_row()} or after retrieving all rows in the result. @subsubheading Example @example MYSQL_ROW row; unsigned long *lengths; unsigned int num_fields; unsigned int i; row = mysql_fetch_row(result); if (row) @{ num_fields = mysql_num_fields(result); lengths = mysql_fetch_lengths(result); for(i = 0; i < num_fields; i++) @{ printf("Column %u is %lu bytes in length.\n", i, lengths[i]); @} @} @end example @node mysql_fetch_row, mysql_field_count, mysql_fetch_lengths, C API functions @subsubsection @code{mysql_fetch_row()} @findex @code{mysql_fetch_row()} @code{MYSQL_ROW mysql_fetch_row(MYSQL_RES *result)} @subsubheading Description Retrieves the next row of a result set. When used after @code{mysql_store_result()}, @code{mysql_fetch_row()} returns @code{NULL} when there are no more rows to retrieve. When used after @code{mysql_use_result()}, @code{mysql_fetch_row()} returns @code{NULL} when there are no more rows to retrieve or if an error occurred. The number of values in the row is given by @code{mysql_num_fields(result)}. If @code{row} holds the return value from a call to @code{mysql_fetch_row()}, pointers to the values are accessed as @code{row[0]} to @code{row[mysql_num_fields(result)-1]}. @code{NULL} values in the row are indicated by @code{NULL} pointers. The lengths of the field values in the row may be obtained by calling @code{mysql_fetch_lengths()}. Empty fields and fields containing @code{NULL} both have length 0; you can distinguish these by checking the pointer for the field value. If the pointer is @code{NULL}, the field is @code{NULL}; otherwise the field is empty. @subsubheading Return Values A @code{MYSQL_ROW} structure for the next row. @code{NULL} if there are no more rows to retrieve or if an error occurred. @subsubheading Errors @table @code @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @subsubheading Example @example MYSQL_ROW row; unsigned int num_fields; unsigned int i; num_fields = mysql_num_fields(result); while ((row = mysql_fetch_row(result))) @{ unsigned long *lengths; lengths = mysql_fetch_lengths(result); for(i = 0; i < num_fields; i++) @{ printf("[%.*s] ", (int) lengths[i], row[i] ? row[i] : "NULL"); @} printf("\n"); @} @end example @node mysql_field_count, mysql_field_seek, mysql_fetch_row, C API functions @subsubsection @code{mysql_field_count()} @findex @code{mysql_field_count()} @code{unsigned int mysql_field_count(MYSQL *mysql)} If you are using a version of MySQL earlier than Version 3.22.24, you should use @code{unsigned int mysql_num_fields(MYSQL *mysql)} instead. @subsubheading Description Returns the number of columns for the most recent query on the connection. The normal use of this function is when @code{mysql_store_result()} returned @code{NULL} (and thus you have no result set pointer). In this case, you can call @code{mysql_field_count()} to determine whether or not @code{mysql_store_result()} should have produced a non-empty result. This allows the client program to take proper action without knowing whether or not the query was a @code{SELECT} (or @code{SELECT}-like) statement. The example shown below illustrates how this may be done. @xref{NULL mysql_store_result, , @code{NULL mysql_store_result()}}. @subsubheading Return Values An unsigned integer representing the number of fields in a result set. @subsubheading Errors None. @subsubheading Example @example MYSQL_RES *result; unsigned int num_fields; unsigned int num_rows; if (mysql_query(&mysql,query_string)) @{ // error @} else // query succeeded, process any data returned by it @{ result = mysql_store_result(&mysql); if (result) // there are rows @{ num_fields = mysql_num_fields(result); // retrieve rows, then call mysql_free_result(result) @} else // mysql_store_result() returned nothing; should it have? @{ if(mysql_field_count(&mysql) == 0) @{ // query does not return data // (it was not a SELECT) num_rows = mysql_affected_rows(&mysql); @} else // mysql_store_result() should have returned data @{ fprintf(stderr, "Error: %s\n", mysql_error(&mysql)); @} @} @} @end example An alternative is to replace the @code{mysql_field_count(&mysql)} call with @code{mysql_errno(&mysql)}. In this case, you are checking directly for an error from @code{mysql_store_result()} rather than inferring from the value of @code{mysql_field_count()} whether or not the statement was a @code{SELECT}. @node mysql_field_seek, mysql_field_tell, mysql_field_count, C API functions @subsubsection @code{mysql_field_seek()} @findex @code{mysql_field_seek()} @code{MYSQL_FIELD_OFFSET mysql_field_seek(MYSQL_RES *result, MYSQL_FIELD_OFFSET offset)} * Threaded clients:: How to Make a Threaded Client @subsubheading Description Sets the field cursor to the given offset. The next call to @code{mysql_fetch_field()} will retrieve the field definition of the column associated with that offset. To seek to the beginning of a row, pass an @code{offset} value of zero. @subsubheading Return Values The previous value of the field cursor. @subsubheading Errors None. @node mysql_field_tell, mysql_free_result, mysql_field_seek, C API functions @subsubsection @code{mysql_field_tell()} @findex @code{mysql_field_tell()} @code{MYSQL_FIELD_OFFSET mysql_field_tell(MYSQL_RES *result)} @subsubheading Description Returns the position of the field cursor used for the last @code{mysql_fetch_field()}. This value can be used as an argument to @code{mysql_field_seek()}. @subsubheading Return Values The current offset of the field cursor. @subsubheading Errors None. @node mysql_free_result, mysql_get_client_info, mysql_field_tell, C API functions @subsubsection @code{mysql_free_result()} @findex @code{mysql_free_result()} @code{void mysql_free_result(MYSQL_RES *result)} @subsubheading Description Frees the memory allocated for a result set by @code{mysql_store_result()}, @code{mysql_use_result()}, @code{mysql_list_dbs()}, etc. When you are done with a result set, you must free the memory it uses by calling @code{mysql_free_result()}. @subsubheading Return Values None. @subsubheading Errors None. @node mysql_get_client_info, mysql_get_host_info, mysql_free_result, C API functions @subsubsection @code{mysql_get_client_info()} @findex @code{mysql_get_client_info()} @code{char *mysql_get_client_info(void)} @subsubheading Description Returns a string that represents the client library version. @subsubheading Return Values A character string that represents the MySQL client library version. @subsubheading Errors None. @node mysql_get_host_info, mysql_get_proto_info, mysql_get_client_info, C API functions @subsubsection @code{mysql_get_host_info()} @findex @code{mysql_get_host_info()} @code{char *mysql_get_host_info(MYSQL *mysql)} @subsubheading Description Returns a string describing the type of connection in use, including the server host name. @subsubheading Return Values A character string representing the server host name and the connection type. @subsubheading Errors None. @node mysql_get_proto_info, mysql_get_server_info, mysql_get_host_info, C API functions @subsubsection @code{mysql_get_proto_info()} @findex @code{mysql_get_proto_info()} @code{unsigned int mysql_get_proto_info(MYSQL *mysql)} @subsubheading Description Returns the protocol version used by current connection. @subsubheading Return Values An unsigned integer representing the protocol version used by the current connection. @subsubheading Errors None. @node mysql_get_server_info, mysql_info, mysql_get_proto_info, C API functions @subsubsection @code{mysql_get_server_info()} @findex @code{mysql_get_server_info()} @code{char *mysql_get_server_info(MYSQL *mysql)} @subsubheading Description Returns a string that represents the server version number. @subsubheading Return Values A character string that represents the server version number. @subsubheading Errors None. @node mysql_info, mysql_init, mysql_get_server_info, C API functions @subsubsection @code{mysql_info()} @findex @code{mysql_info()} @code{char *mysql_info(MYSQL *mysql)} @subsubheading Description Retrieves a string providing information about the most recently executed query, but only for the statements listed below. For other statements, @code{mysql_info()} returns @code{NULL}. The format of the string varies depending on the type of query, as described below. The numbers are illustrative only; the string will contain values appropriate for the query. @table @code @item INSERT INTO ... SELECT ... String format: @code{Records: 100 Duplicates: 0 Warnings: 0} @item INSERT INTO ... VALUES (...),(...),(...)... String format: @code{Records: 3 Duplicates: 0 Warnings: 0} @item LOAD DATA INFILE ... String format: @code{Records: 1 Deleted: 0 Skipped: 0 Warnings: 0} @item ALTER TABLE String format: @code{Records: 3 Duplicates: 0 Warnings: 0} @item UPDATE String format: @code{Rows matched: 40 Changed: 40 Warnings: 0} @end table Note that @code{mysql_info()} returns a non-@code{NULL} value for the @code{INSERT ... VALUES} statement only if multiple value lists are specified in the statement. @subsubheading Return Values A character string representing additional information about the most recently executed query. @code{NULL} if no information is available for the query. @subsubheading Errors None. @node mysql_init, mysql_insert_id, mysql_info, C API functions @subsubsection @code{mysql_init()} @findex @code{mysql_init()} @code{MYSQL *mysql_init(MYSQL *mysql)} @subsubheading Description Allocates or initializes a @code{MYSQL} object suitable for @code{mysql_real_connect()}. If @code{mysql} is a @code{NULL} pointer, the function allocates, initializes, and returns a new object. Otherwise the object is initialized and the address of the object is returned. If @code{mysql_init()} allocates a new object, it will be freed when @code{mysql_close()} is called to close the connection. @subsubheading Return Values An initialized @code{MYSQL*} handle. @code{NULL} if there was insufficient memory to allocate a new object. @subsubheading Errors In case of insufficient memory, @code{NULL} is returned. @node mysql_insert_id, mysql_kill, mysql_init, C API functions @subsubsection @code{mysql_insert_id()} @findex @code{mysql_insert_id()} @code{my_ulonglong mysql_insert_id(MYSQL *mysql)} @subsubheading Description Returns the ID generated for an @code{AUTO_INCREMENT} column by the previous query. Use this function after you have performed an @code{INSERT} query into a table that contains an @code{AUTO_INCREMENT} field. Note that @code{mysql_insert_id()} returns @code{0} if the previous query does not generate an @code{AUTO_INCREMENT} value. If you need to save the value for later, be sure to call @code{mysql_insert_id()} immediately after the query that generates the value. @code{mysql_insert_id()} is updated after @code{INSERT} and @code{UPDATE} statements that generate an @code{AUTO_INCREMENT} value or that set a column value to @code{LAST_INSERT_ID(expr)}. @xref{Miscellaneous functions}. Also note that the value of the SQL @code{LAST_INSERT_ID()} function always contains the most recently generated @code{AUTO_INCREMENT} value, and is not reset between queries because the value of that function is maintained in the server. @subsubheading Return Values The value of the @code{AUTO_INCREMENT} field that was updated by the previous query. Returns zero if there was no previous query on the connection or if the query did not update an @code{AUTO_INCREMENT} value. @subsubheading Errors None. @node mysql_kill, mysql_list_dbs, mysql_insert_id, C API functions @subsubsection @code{mysql_kill()} @findex @code{mysql_kill()} @code{int mysql_kill(MYSQL *mysql, unsigned long pid)} @subsubheading Description Asks the server to kill the thread specified by @code{pid}. @subsubheading Return Values Zero for success. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_list_dbs, mysql_list_fields, mysql_kill, C API functions @subsubsection @code{mysql_list_dbs()} @findex @code{mysql_list_dbs()} @code{MYSQL_RES *mysql_list_dbs(MYSQL *mysql, const char *wild)} @subsubheading Description Returns a result set consisting of database names on the server that match the simple regular expression specified by the @code{wild} parameter. @code{wild} may contain the wild-card characters @samp{%} or @samp{_}, or may be a @code{NULL} pointer to match all databases. Calling @code{mysql_list_dbs()} is similar to executing the query @code{SHOW databases [LIKE wild]}. You must free the result set with @code{mysql_free_result()}. @subsubheading Return Values A @code{MYSQL_RES} result set for success. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_OUT_OF_MEMORY Out of memory. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_list_fields, mysql_list_processes, mysql_list_dbs, C API functions @subsubsection @code{mysql_list_fields()} @findex @code{mysql_list_fields()} @code{MYSQL_RES *mysql_list_fields(MYSQL *mysql, const char *table, const char *wild)} @subsubheading Description Returns a result set consisting of field names in the given table that match the simple regular expression specified by the @code{wild} parameter. @code{wild} may contain the wild-card characters @samp{%} or @samp{_}, or may be a @code{NULL} pointer to match all fields. Calling @code{mysql_list_fields()} is similar to executing the query @code{SHOW COLUMNS FROM tbl_name [LIKE wild]}. Note that it's recommended that you use @code{SHOW COLUMNS FROM tbl_name} instead of @code{mysql_list_fields()}. You must free the result set with @code{mysql_free_result()}. @subsubheading Return Values A @code{MYSQL_RES} result set for success. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_list_processes, mysql_list_tables, mysql_list_fields, C API functions @subsubsection @code{mysql_list_processes()} @findex @code{mysql_list_processes()} @code{MYSQL_RES *mysql_list_processes(MYSQL *mysql)} @subsubheading Description Returns a result set describing the current server threads. This is the same kind of information as that reported by @code{mysqladmin processlist} or a @code{SHOW PROCESSLIST} query. You must free the result set with @code{mysql_free_result()}. @subsubheading Return Values A @code{MYSQL_RES} result set for success. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_list_tables, mysql_num_fields, mysql_list_processes, C API functions @subsubsection @code{mysql_list_tables()} @findex @code{mysql_list_tables()} @code{MYSQL_RES *mysql_list_tables(MYSQL *mysql, const char *wild)} @subsubheading Description Returns a result set consisting of table names in the current database that match the simple regular expression specified by the @code{wild} parameter. @code{wild} may contain the wild-card characters @samp{%} or @samp{_}, or may be a @code{NULL} pointer to match all tables. Calling @code{mysql_list_tables()} is similar to executing the query @code{SHOW tables [LIKE wild]}. You must free the result set with @code{mysql_free_result()}. @subsubheading Return Values A @code{MYSQL_RES} result set for success. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_num_fields, mysql_num_rows, mysql_list_tables, C API functions @subsubsection @code{mysql_num_fields()} @findex @code{mysql_num_fields()} @findex @code{mysql_field_count()} @code{unsigned int mysql_num_fields(MYSQL_RES *result)} or @code{unsigned int mysql_num_fields(MYSQL *mysql)} The second form doesn't work on MySQL Version 3.22.24 or newer. To pass a @code{MYSQL*} argument, you must use @code{unsigned int mysql_field_count(MYSQL *mysql)} instead. @subsubheading Description Returns the number of columns in a result set. Note that you can get the number of columns either from a pointer to a result set or to a connection handle. You would use the connection handle if @code{mysql_store_result()} or @code{mysql_use_result()} returned @code{NULL} (and thus you have no result set pointer). In this case, you can call @code{mysql_field_count()} to determine whether or not @code{mysql_store_result()} should have produced a non-empty result. This allows the client program to take proper action without knowing whether or not the query was a @code{SELECT} (or @code{SELECT}-like) statement. The example shown below illustrates how this may be done. @xref{NULL mysql_store_result, , @code{NULL mysql_store_result()}}. @subsubheading Return Values An unsigned integer representing the number of fields in a result set. @subsubheading Errors None. @subsubheading Example @example MYSQL_RES *result; unsigned int num_fields; unsigned int num_rows; if (mysql_query(&mysql,query_string)) @{ // error @} else // query succeeded, process any data returned by it @{ result = mysql_store_result(&mysql); if (result) // there are rows @{ num_fields = mysql_num_fields(result); // retrieve rows, then call mysql_free_result(result) @} else // mysql_store_result() returned nothing; should it have? @{ if (mysql_errno(&mysql)) @{ fprintf(stderr, "Error: %s\n", mysql_error(&mysql)); @} else if (mysql_field_count(&mysql) == 0) @{ // query does not return data // (it was not a SELECT) num_rows = mysql_affected_rows(&mysql); @} @} @} @end example An alternative (if you KNOW that your query should have returned a result set) is to replace the @code{mysql_errno(&mysql)} call with a check if @code{mysql_field_count(&mysql)} is = 0. This will only happen if something went wrong. @node mysql_num_rows, mysql_options, mysql_num_fields, C API functions @subsubsection @code{mysql_num_rows()} @findex @code{mysql_num_rows()} @code{my_ulonglong mysql_num_rows(MYSQL_RES *result)} @subsubheading Description Returns the number of rows in the result set. The use of @code{mysql_num_rows()} depends on whether you use @code{mysql_store_result()} or @code{mysql_use_result()} to return the result set. If you use @code{mysql_store_result()}, @code{mysql_num_rows()} may be called immediately. If you use @code{mysql_use_result()}, @code{mysql_num_rows()} will not return the correct value until all the rows in the result set have been retrieved. @subsubheading Return Values The number of rows in the result set. @subsubheading Errors None. @node mysql_options, mysql_ping, mysql_num_rows, C API functions @subsubsection @code{mysql_options()} @findex @code{mysql_options()} @code{int mysql_options(MYSQL *mysql, enum mysql_option option, const char *arg)} @subsubheading Description Can be used to set extra connect options and affect behavior for a connection. This function may be called multiple times to set several options. @code{mysql_options()} should be called after @code{mysql_init()} and before @code{mysql_connect()} or @code{mysql_real_connect()}. The @code{option} argument is the option that you want to set; the @code{arg} argument is the value for the option. If the option is an integer, then @code{arg} should point to the value of the integer. Possible options values: @multitable @columnfractions .25 .25 .5 @item @strong{Option} @tab @strong{Argument type} @tab @strong{Function} @item @code{MYSQL_OPT_CONNECT_TIMEOUT} @tab @code{unsigned int *} @tab Connect timeout in seconds. @item @code{MYSQL_OPT_COMPRESS} @tab Not used @tab Use the compressed client/server protocol. @item @code{MYSQL_OPT_NAMED_PIPE} @tab Not used @tab Use named pipes to connect to a MySQL server on NT. @item @code{MYSQL_INIT_COMMAND} @tab @code{char *} @tab Command to execute when connecting to the MySQL server. Will automatically be re-executed when reconnecting. @item @code{MYSQL_READ_DEFAULT_FILE} @tab @code{char *} @tab Read options from the named option file instead of from @file{my.cnf}. @item @code{MYSQL_READ_DEFAULT_GROUP} @tab @code{char *} @tab Read options from the named group from @file{my.cnf} or the file specified with @code{MYSQL_READ_DEFAULT_FILE}. @end multitable Note that the group @code{client} is always read if you use @code{MYSQL_READ_DEFAULT_FILE} or @code{MYSQL_READ_DEFAULT_GROUP}. The specified group in the option file may contain the following options: @multitable @columnfractions .3 .7 @item @code{connect_timeout} @tab Connect timeout in seconds. On Linux this timeout is also used for waiting for the first answer from the server. @item @code{compress} @tab Use the compressed client/server protocol. @item @code{database} @tab Connect to this database if no database was specified in the connect command. @item @code{debug} @tab Debug options. @item @code{host} @tab Default host name. @item @code{init-command} @tab Command to execute when connecting to MySQL server. Will automatically be re-executed when reconnecting. @item @code{interactive-timeout} @tab Same as specifying @code{CLIENT_INTERACTIVE} to @code{mysql_real_connect()}. @xref{mysql_real_connect}. @item @code{password} @tab Default password. @item @code{pipe} @tab Use named pipes to connect to a MySQL server on NT. @item @code{port} @tab Default port number. @item @code{return-found-rows} @tab Tell @code{mysql_info()} to return found rows instead of updated rows when using @code{UPDATE}. @item @code{socket} @tab Default socket number. @item @item @code{user} @tab Default user. @end multitable Note that @code{timeout} has been replaced by @code{connect_timeout}, but @code{timeout} will still work for a while. For more information about option files, see @ref{Option files}. @subsubheading Return Values Zero for success. Non-zero if you used an unknown option. @subsubheading Example @example MYSQL mysql; mysql_init(&mysql); mysql_options(&mysql,MYSQL_OPT_COMPRESS,0); mysql_options(&mysql,MYSQL_READ_DEFAULT_GROUP,"odbc"); if (!mysql_real_connect(&mysql,"host","user","passwd","database",0,NULL,0)) @{ fprintf(stderr, "Failed to connect to database: Error: %s\n", mysql_error(&mysql)); @} @end example The above requests the client to use the compressed client/server protocol and read the additional options from the @code{odbc} section in the @code{my.cnf} file. @node mysql_ping, mysql_query, mysql_options, C API functions @subsubsection @code{mysql_ping()} @findex @code{mysql_ping()} @code{int mysql_ping(MYSQL *mysql)} @subsubheading Description Checks whether or not the connection to the server is working. If it has gone down, an automatic reconnection is attempted. This function can be used by clients that remain idle for a long while, to check whether or not the server has closed the connection and reconnect if necessary. @subsubheading Return Values Zero if the server is alive. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_query, mysql_real_connect, mysql_ping, C API functions @subsubsection @code{mysql_query()} @findex @code{mysql_query()} @code{int mysql_query(MYSQL *mysql, const char *query)} @subsubheading Description Executes the SQL query pointed to by the null-terminated string @code{query}. The query must consist of a single SQL statement. You should not add a terminating semicolon (@samp{;}) or @code{\g} to the statement. @code{mysql_query()} cannot be used for queries that contain binary data; you should use @code{mysql_real_query()} instead. (Binary data may contain the @samp{\0} character, which @code{mysql_query()} interprets as the end of the query string.) If you want to know if the query should return a result set or not, you can use @code{mysql_field_count()} to check for this. @xref{mysql_field_count, , @code{mysql_field_count}}. @subsubheading Return Values Zero if the query was successful. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_real_connect, mysql_real_escape_string, mysql_query, C API functions @subsubsection @code{mysql_real_connect()} @findex @code{mysql_real_connect()} @code{MYSQL *mysql_real_connect(MYSQL *mysql, const char *host, const char *user, const char *passwd, const char *db, unsigned int port, const char *unix_socket, unsigned int client_flag)} @subsubheading Description @code{mysql_real_connect()} attempts to establish a connection to a MySQL database engine running on @code{host}. @code{mysql_real_connect()} must complete successfully before you can execute any of the other API functions, with the exception of @code{mysql_get_client_info()}. The parameters are specified as follows: @itemize @bullet @item The first parameter should be the address of an existing @code{MYSQL} structure. Before calling @code{mysql_real_connect()} you must call @code{mysql_init()} to initialize the @code{MYSQL} structure. You can change a lot of connect options with the @code{mysql_options()} call. @xref{mysql_options}. @item The value of @code{host} may be either a hostname or an IP address. If @code{host} is @code{NULL} or the string @code{"localhost"}, a connection to the local host is assumed. If the OS supports sockets (Unix) or named pipes (Windows), they are used instead of TCP/IP to connect to the server. @item The @code{user} parameter contains the user's MySQL login ID. If @code{user} is @code{NULL}, the current user is assumed. Under Unix, this is the current login name. Under Windows ODBC, the current user name must be specified explicitly. @xref{ODBC administrator}. @item The @code{passwd} parameter contains the password for @code{user}. If @code{passwd} is @code{NULL}, only entries in the @code{user} table for the user that have a blank (empty) password field will be checked for a match. This allows the database administrator to set up the MySQL privilege system in such a way that users get different privileges depending on whether or not they have specified a password. NOTE: Do not attempt to encrypt the password before calling @code{mysql_real_connect()}; password encryption is handled automatically by the client API. @item @code{db} is the database name. If @code{db} is not @code{NULL}, the connection will set the default database to this value. @item If @code{port} is not 0, the value will be used as the port number for the TCP/IP connection. Note that the @code{host} parameter determines the type of the connection. @item If @code{unix_socket} is not @code{NULL}, the string specifies the socket or named pipe that should be used. Note that the @code{host} parameter determines the type of the connection. @item The value of @code{client_flag} is usually 0, but can be set to a combination of the following flags in very special circumstances: @multitable @columnfractions .25 .7 @item @strong{Flag name} @tab @strong{Flag meaning} @code{mysqld} to be more ODBC-friendly. @item @code{CLIENT_COMPRESS} @tab Use compression protocol. @item @code{CLIENT_FOUND_ROWS} @tab Return the number of found (matched) rows, not the number of affected rows. @item @code{CLIENT_IGNORE_SPACE} @tab Allow spaces after function names. Makes all functions names reserved words. @item @code{CLIENT_INTERACTIVE} @tab Allow @code{interactive_timeout} seconds (instead of @code{wait_timeout} seconds) of inactivity before closing the connection. @item @code{CLIENT_NO_SCHEMA} @tab Don't allow the @code{db_name.tbl_name.col_name} syntax. This is for ODBC. It causes the parser to generate an error if you use that syntax, which is useful for trapping bugs in some ODBC programs. @item @code{CLIENT_ODBC} @tab The client is an ODBC client. This changes @item @code{CLIENT_SSL} @tab Use SSL (encrypted protocol). @end multitable @end itemize @subsubheading Return Values A @code{MYSQL*} connection handle if the connection was successful, @code{NULL} if the connection was unsuccessful. For a successful connection, the return value is the same as the value of the first parameter. @subsubheading Errors @table @code @item CR_CONN_HOST_ERROR Failed to connect to the MySQL server. @item CR_CONNECTION_ERROR Failed to connect to the local MySQL server. @item CR_IPSOCK_ERROR Failed to create an IP socket. @item CR_OUT_OF_MEMORY Out of memory. @item CR_SOCKET_CREATE_ERROR Failed to create a Unix socket. @item CR_UNKNOWN_HOST Failed to find the IP address for the hostname. @item CR_VERSION_ERROR A protocol mismatch resulted from attempting to connect to a server with a client library that uses a different protocol version. This can happen if you use a very old client library to connect to a new server that wasn't started with the @code{--old-protocol} option. @item CR_NAMEDPIPEOPEN_ERROR Failed to create a named pipe on Windows. @item CR_NAMEDPIPEWAIT_ERROR Failed to wait for a named pipe on Windows. @item CR_NAMEDPIPESETSTATE_ERROR Failed to get a pipe handler on Windows. @item CR_SERVER_LOST If @code{connect_timeout} > 0 and it took longer then @code{connect_timeout} seconds to connect to the server or if the server died while executing the @code{init-command}. @end table @subsubheading Example @example MYSQL mysql; mysql_init(&mysql); mysql_options(&mysql,MYSQL_READ_DEFAULT_GROUP,"your_prog_name"); if (!mysql_real_connect(&mysql,"host","user","passwd","database",0,NULL,0)) @{ fprintf(stderr, "Failed to connect to database: Error: %s\n", mysql_error(&mysql)); @} @end example By using @code{mysql_options()} the MySQL library will read the @code{[client]} and @code{your_prog_name} sections in the @code{my.cnf} file which will ensure that your program will work, even if someone has set up MySQL in some non-standard way. Note that upon connection, @code{mysql_real_connect()} sets the @code{reconnect} flag (part of the MYSQL structure) to a value of @code{1}. This flag indicates, in the event that a query cannot be performed because of a lost connection, to try reconnecting to the server before giving up. @node mysql_real_escape_string, mysql_real_query, mysql_real_connect, C API functions @subsubsection @code{mysql_real_escape_string()} @findex @code{mysql_real_escape_string()} @code{unsigned int mysql_real_escape_string(MYSQL *mysql, char *to, const char *from, unsigned int length)} @subsubheading Description This function is used to create a legal SQL string that you can use in a SQL statement. @xref{String syntax}. The string in @code{from} is encoded to an escaped SQL string, taking into account the current character set of the connection. The result is placed in @code{to} and a terminating null byte is appended. Characters encoded are @code{NUL} (ASCII 0), @samp{\n}, @samp{\r}, @samp{\}, @samp{'}, @samp{"}, and Control-Z (@pxref{Literals}). The string pointed to by @code{from} must be @code{length} bytes long. You must allocate the @code{to} buffer to be at least @code{length*2+1} bytes long. (In the worse case, each character may need to be encoded as using two bytes, and you need room for the terminating null byte.) When @code{mysql_escape_string()} returns, the contents of @code{to} will be a null-terminated string. The return value is the length of the encoded string, not including the terminating null character. @subsubheading Example @example char query[1000],*end; end = strmov(query,"INSERT INTO test_table values("); *end++ = '\''; end += mysql_real_escape_string(&mysql, end,"What's this",11); *end++ = '\''; *end++ = ','; *end++ = '\''; end += mysql_real_escape_string(&mysql, end,"binary data: \0\r\n",16); *end++ = '\''; *end++ = ')'; if (mysql_real_query(&mysql,query,(unsigned int) (end - query))) @{ fprintf(stderr, "Failed to insert row, Error: %s\n", mysql_error(&mysql)); @} @end example The @code{strmov()} function used in the example is included in the @code{mysqlclient} library and works like @code{strcpy()} but returns a pointer to the terminating null of the first parameter. @subsubheading Return Values The length of the value placed into @code{to}, not including the terminating null character. @subsubheading Errors None. @node mysql_real_query, mysql_reload, mysql_real_escape_string, C API functions @subsubsection @code{mysql_real_query()} @findex @code{mysql_real_query()} @code{int mysql_real_query(MYSQL *mysql, const char *query, unsigned int length)} @subsubheading Description Executes the SQL query pointed to by @code{query}, which should be a string @code{length} bytes long. The query must consist of a single SQL statement. You should not add a terminating semicolon (@samp{;}) or @code{\g} to the statement. You @emph{must} use @code{mysql_real_query()} rather than @code{mysql_query()} for queries that contain binary data, because binary data may contain the @samp{\0} character. In addition, @code{mysql_real_query()} is faster than @code{mysql_query()} because it does not call @code{strlen()} on the query string. If you want to know if the query should return a result set or not, you can use @code{mysql_field_count()} to check for this. @xref{mysql_field_count, @code{mysql_field_count}}. @subsubheading Return Values Zero if the query was successful. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_reload, mysql_row_seek, mysql_real_query, C API functions @subsubsection @code{mysql_reload()} @findex @code{mysql_reload()} @code{int mysql_reload(MYSQL *mysql)} @subsubheading Description Asks the MySQL server to reload the grant tables. The connected user must have the @strong{reload} privilege. This function is deprecated. It is preferable to use @code{mysql_query()} to issue a SQL @code{FLUSH PRIVILEGES} statement instead. @subsubheading Return Values Zero for success. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_row_seek, mysql_row_tell, mysql_reload, C API functions @subsubsection @code{mysql_row_seek()} @findex @code{mysql_row_seek()} @code{MYSQL_ROW_OFFSET mysql_row_seek(MYSQL_RES *result, MYSQL_ROW_OFFSET offset)} @subsubheading Description Sets the row cursor to an arbitrary row in a query result set. This requires that the result set structure contains the entire result of the query, so @code{mysql_row_seek()} may be used in conjunction only with @code{mysql_store_result()}, not with @code{mysql_use_result()}. The offset should be a value returned from a call to @code{mysql_row_tell()} or to @code{mysql_row_seek()}. This value is not simply a row number; if you want to seek to a row within a result set using a row number, use @code{mysql_data_seek()} instead. @subsubheading Return Values The previous value of the row cursor. This value may be passed to a subsequent call to @code{mysql_row_seek()}. @subsubheading Errors None. @node mysql_row_tell, mysql_select_db, mysql_row_seek, C API functions @subsubsection @code{mysql_row_tell()} @findex @code{mysql_row_tell()} @code{MYSQL_ROW_OFFSET mysql_row_tell(MYSQL_RES *result)} @subsubheading Description Returns the current position of the row cursor for the last @code{mysql_fetch_row()}. This value can be used as an argument to @code{mysql_row_seek()}. You should use @code{mysql_row_tell()} only after @code{mysql_store_result()}, not after @code{mysql_use_result()}. @subsubheading Return Values The current offset of the row cursor. @subsubheading Errors None. @node mysql_select_db, mysql_shutdown, mysql_row_tell, C API functions @subsubsection @code{mysql_select_db()} @findex @code{mysql_select_db()} @code{int mysql_select_db(MYSQL *mysql, const char *db)} @subsubheading Description Causes the database specified by @code{db} to become the default (current) database on the connection specified by @code{mysql}. In subsequent queries, this database is the default for table references that do not include an explicit database specifier. @code{mysql_select_db()} fails unless the connected user can be authenticated as having permission to use the database. @subsubheading Return Values Zero for success. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_shutdown, mysql_stat, mysql_select_db, C API functions @subsubsection @code{mysql_shutdown()} @findex @code{mysql_shutdown()} @code{int mysql_shutdown(MYSQL *mysql)} @subsubheading Description Asks the database server to shut down. The connected user must have @strong{shutdown} privileges. @subsubheading Return Values Zero for success. Non-zero if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_stat, mysql_store_result, mysql_shutdown, C API functions @subsubsection @code{mysql_stat()} @findex @code{mysql_stat()} @code{char *mysql_stat(MYSQL *mysql)} @subsubheading Description Returns a character string containing information similar to that provided by the @code{mysqladmin status} command. This includes uptime in seconds and the number of running threads, questions, reloads, and open tables. @subsubheading Return Values A character string describing the server status. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_store_result, mysql_thread_id, mysql_stat, C API functions @subsubsection @code{mysql_store_result()} @findex @code{mysql_store_result()} @code{MYSQL_RES *mysql_store_result(MYSQL *mysql)} @subsubheading Description You must call @code{mysql_store_result()} or @code{mysql_use_result()} for every query that successfully retrieves data (@code{SELECT}, @code{SHOW}, @code{DESCRIBE}, @code{EXPLAIN}). You don't have to call @code{mysql_store_result()} or @code{mysql_use_result()} for other queries, but it will not do any harm or cause any notable performance if you call @code{mysql_store_result()} in all cases. You can detect if the query didn't have a result set by checking if @code{mysql_store_result()} returns 0 (more about this later one). If you want to know if the query should return a result set or not, you can use @code{mysql_field_count()} to check for this. @xref{mysql_field_count, @code{mysql_field_count}}. @code{mysql_store_result()} reads the entire result of a query to the client, allocates a @code{MYSQL_RES} structure, and places the result into this structure. @code{mysql_store_results()} returns a null pointer if the query didn't return a result set (if the query was, for example, an @code{INSERT} statement). @code{mysql_store_results()} also returns a null pointer if reading of the result set failed. You can check if you got an error by checking if @code{mysql_error()} doesn't return a null pointer, if @code{mysql_errno()} returns <> 0, or if @code{mysql_field_count()} returns <> 0. An empty result set is returned if there are no rows returned. (An empty result set differs from a null pointer as a return value.) Once you have called @code{mysql_store_result()} and got a result back that isn't a null pointer, you may call @code{mysql_num_rows()} to find out how many rows are in the result set. You can call @code{mysql_fetch_row()} to fetch rows from the result set, or @code{mysql_row_seek()} and @code{mysql_row_tell()} to obtain or set the current row position within the result set. You must call @code{mysql_free_result()} once you are done with the result set. @xref{NULL mysql_store_result, , @code{NULL mysql_store_result()}}. @subsubheading Return Values A @code{MYSQL_RES} result structure with the results. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_OUT_OF_MEMORY Out of memory. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node mysql_thread_id, mysql_use_result, mysql_store_result, C API functions @subsubsection @code{mysql_thread_id()} @findex @code{mysql_thread_id()} @code{unsigned long mysql_thread_id(MYSQL *mysql)} @subsubheading Description Returns the thread ID of the current connection. This value can be used as an argument to @code{mysql_kill()} to kill the thread. If the connection is lost and you reconnect with @code{mysql_ping()}, the thread ID will change. This means you should not get the thread ID and store it for later. You should get it when you need it. @subsubheading Return Values The thread ID of the current connection. @subsubheading Errors None. @node mysql_use_result, , mysql_thread_id, C API functions @subsubsection @code{mysql_use_result()} @findex @code{mysql_use_result()} @code{MYSQL_RES *mysql_use_result(MYSQL *mysql)} @subsubheading Description You must call @code{mysql_store_result()} or @code{mysql_use_result()} for every query that successfully retrieves data (@code{SELECT}, @code{SHOW}, @code{DESCRIBE}, @code{EXPLAIN}). @code{mysql_use_result()} initiates a result set retrieval but does not actually read the result set into the client like @code{mysql_store_result()} does. Instead, each row must be retrieved individually by making calls to @code{mysql_fetch_row()}. This reads the result of a query directly from the server without storing it in a temporary table or local buffer, which is somewhat faster and uses much less memory than @code{mysql_store_result()}. The client will only allocate memory for the current row and a communication buffer that may grow up to @code{max_allowed_packet} bytes. On the other hand, you shouldn't use @code{mysql_use_result()} if you are doing a lot of processing for each row on the client side, or if the output is sent to a screen on which the user may type a @code{^S} (stop scroll). This will tie up the server and prevent other threads from updating any tables from which the data is being fetched. When using @code{mysql_use_result()}, you must execute @code{mysql_fetch_row()} until a @code{NULL} value is returned, otherwise the unfetched rows will be returned as part of the result set for your next query. The C API will give the error @code{Commands out of sync; You can't run this command now} if you forget to do this! You may not use @code{mysql_data_seek()}, @code{mysql_row_seek()}, @code{mysql_row_tell()}, @code{mysql_num_rows()}, or @code{mysql_affected_rows()} with a result returned from @code{mysql_use_result()}, nor may you issue other queries until the @code{mysql_use_result()} has finished. (However, after you have fetched all the rows, @code{mysql_num_rows()} will accurately return the number of rows fetched.) You must call @code{mysql_free_result()} once you are done with the result set. @subsubheading Return Values A @code{MYSQL_RES} result structure. @code{NULL} if an error occurred. @subsubheading Errors @table @code @item CR_COMMANDS_OUT_OF_SYNC Commands were executed in an improper order. @item CR_OUT_OF_MEMORY Out of memory. @item CR_SERVER_GONE_ERROR The MySQL server has gone away. @item CR_SERVER_LOST The connection to the server was lost during the query. @item CR_UNKNOWN_ERROR An unknown error occurred. @end table @node C Thread functions, C API problems, C API functions, C @subsection C Threaded Function Descriptions You need to use the following functions when you want to create a threaded client. @xref{Threaded clients}. @menu * my_init:: * my_thread_init():: * my_thread_end():: @end menu @node my_init, my_thread_init(), C Thread functions, C Thread functions @subsubsection @code{my_init()} @findex @code{my_init()} @subsubheading Description This function needs to be called once in the program before calling any MySQL function. This initializes some global variables that MySQL needs. If you are using a thread safe client library, this will also call @code{my_thread_init()} for this thread. This is automatically called by @code{mysql_init()} and @code{mysql_connect()}. @subsubheading Return Values none. @node my_thread_init(), my_thread_end(), my_init, C Thread functions @subsubsection @code{my_thread_init()} @findex @code{my_thread_init()} @subsubheading Description This function needs to be called for each created thread to initialize thread specific variables. This is automatically called by @code{my_init()} and @code{mysql_connect()}. @subsubheading Return Values none. @node my_thread_end(), , my_thread_init(), C Thread functions @subsubsection @code{my_thread_end()} @findex @code{my_thread_end()} @subsubheading Description This function needs to be called before calling @code{pthread_exit()} to freed memory allocated by @code{my_thread_init()}. Note that this function is NOT invoked automatically by the client library! @subsubheading Return Values none. @node C API problems, Building clients, C Thread functions, C @subsection Common questions and problems when using the C API @tindex @code{mysql_query()} @tindex @code{mysql_store_result()} @menu * NULL mysql_store_result:: Why Is It that After @code{mysql_query()} Returns Success, @code{mysql_store_result()} Sometimes Returns @code{NULL?} * Query results:: What Results Can I Get From a Query? * Getting unique ID:: How Can I Get the Unique ID for the Last Inserted Row? * C API linking problems:: Problems Linking with the C API @end menu @node NULL mysql_store_result, Query results, C API problems, C API problems @subsubsection Why Is It that After @code{mysql_query()} Returns Success, @code{mysql_store_result()} Sometimes Returns @code{NULL?} It is possible for @code{mysql_store_result()} to return @code{NULL} following a successful call to @code{mysql_query()}. When this happens, it means one of the following conditions occurred: @itemize @bullet @item There was a @code{malloc()} failure (for example, if the result set was too large). @item The data couldn't be read (an error occurred on the connection). @item The query returned no data (for example, it was an @code{INSERT}, @code{UPDATE}, or @code{DELETE}). @end itemize You can always check whether or not the statement should have produced a non-empty result by calling @code{mysql_field_count()}. If @code{mysql_field_count()} returns zero, the result is empty and the last query was a statement that does not return values (for example, an @code{INSERT} or a @code{DELETE}). If @code{mysql_field_count()} returns a non-zero value, the statement should have produced a non-empty result. See the description of the @code{mysql_field_count()} function for an example. You can test for an error by calling @code{mysql_error()} or @code{mysql_errno()}. @cindex queries, C API results @menu * Query results:: What Results Can I Get From a Query? * Getting unique ID:: How Can I Get the Unique ID for the Last Inserted Row? * C API linking problems:: Problems Linking with the C API @end menu @node Query results, Getting unique ID, NULL mysql_store_result, C API problems @subsubsection What Results Can I Get From a Query? In addition to the result set returned by a query, you can also get the following information: @itemize @bullet @item @code{mysql_affected_rows()} returns the number of rows affected by the last query when doing an @code{INSERT}, @code{UPDATE}, or @code{DELETE}. An exception is that if @code{DELETE} is used without a @code{WHERE} clause, the table is re-created empty, which is much faster! In this case, @code{mysql_affected_rows()} returns zero for the number of records affected. @item @code{mysql_num_rows()} returns the number of rows in a result set. With @code{mysql_store_result()}, @code{mysql_num_rows()} may be called as soon as @code{mysql_store_result()} returns. With @code{mysql_use_result()}, @code{mysql_num_rows()} may be called only after you have fetched all the rows with @code{mysql_fetch_row()}. @item @code{mysql_insert_id()} returns the ID generated by the last query that inserted a row into a table with an @code{AUTO_INCREMENT} index. @xref{mysql_insert_id, , @code{mysql_insert_id()}}. @item Some queries (@code{LOAD DATA INFILE ...}, @code{INSERT INTO ... SELECT ...}, @code{UPDATE}) return additional information. The result is returned by @code{mysql_info()}. See the description for @code{mysql_info()} for the format of the string that it returns. @code{mysql_info()} returns a @code{NULL} pointer if there is no additional information. @end itemize @node Getting unique ID, C API linking problems, Query results, C API problems @subsubsection How Can I Get the Unique ID for the Last Inserted Row? @cindex unique ID @cindex last row, unique ID @cindex ID, unique @cindex tables, unique ID for last row If you insert a record in a table containing a column that has the @code{AUTO_INCREMENT} attribute, you can get the most recently generated ID by calling the @code{mysql_insert_id()} function. You can also retrieve the ID by using the @code{LAST_INSERT_ID()} function in a query string that you pass to @code{mysql_query()}. You can check if an @code{AUTO_INCREMENT} index is used by executing the following code. This also checks if the query was an @code{INSERT} with an @code{AUTO_INCREMENT} index: @example if (mysql_error(&mysql)[0] == 0 && mysql_num_fields(result) == 0 && mysql_insert_id(&mysql) != 0) @{ used_id = mysql_insert_id(&mysql); @} @end example The most recently generated ID is maintained in the server on a per-connection basis. It will not be changed by another client. It will not even be changed if you update another @code{AUTO_INCREMENT} column with a non-magic value (that is, a value that is not @code{NULL} and not @code{0}). If you want to use the ID that was generated for one table and insert it into a second table, you can use SQL statements like this: @example INSERT INTO foo (auto,text) VALUES(NULL,'text'); # generate ID by inserting NULL INSERT INTO foo2 (id,text) VALUES(LAST_INSERT_ID(),'text'); # use ID in second table @end example @node C API linking problems, , Getting unique ID, C API problems @subsubsection Problems Linking with the C API @cindex linking, problems @cindex C API, linking problems When linking with the C API, the following errors may occur on some systems: @example gcc -g -o client test.o -L/usr/local/lib/mysql -lmysqlclient -lsocket -lnsl Undefined first referenced symbol in file floor /usr/local/lib/mysql/libmysqlclient.a(password.o) ld: fatal: Symbol referencing errors. No output written to client @end example If this happens on your system, you must include the math library by adding @code{-lm} to the end of the compile/link line. @node Building clients, Threaded clients, C API problems, C @subsection Building Client Programs @cindex client programs, building @cindex linking @cindex building, client programs @cindex programs, client If you compile MySQL clients that you've written yourself or that you obtain from a third party, they must be linked using the @code{-lmysqlclient -lz} option on the link command. You may also need to specify a @code{-L} option to tell the linker where to find the library. For example, if the library is installed in @file{/usr/local/mysql/lib}, use @code{-L/usr/local/mysql/lib -lmysqlclient -lz} on the link command. For clients that use MySQL header files, you may need to specify a @code{-I} option when you compile them (for example, @code{-I/usr/local/mysql/include}), so the compiler can find the header files. @node Threaded clients, , Building clients, C @subsection How to Make a Threaded Client @cindex clients, threaded @cindex threaded clients The client library is almost thread safe. The biggest problem is that the subroutines in @file{net.c} that read from sockets are not interrupt safe. This was done with the thought that you might want to have your own alarm that can break a long read to a server. If you install interrupt handlers for the @code{SIGPIPE} interrupt, the socket handling should be thread safe. In the older binaries we distribute on our Web site, the client libraries are not normally compiled with the thread-safe option (the Windows binaries are by default compiled to be thread safe). Newer binary distributions should have both a normal and a thread-safe client library. To get a threaded client where you can interrupt the client from other threads and set timeouts when talking with the MySQL server, you should use the @code{-lmysys}, @code{-lstring}, and @code{-ldbug} libraries and the @code{net_serv.o} code that the server uses. If you don't need interrupts or timeouts, you can just compile a thread safe client library @code{(mysqlclient_r)} and use this. @xref{C,, MySQL C API}. In this case you don't have to worry about the @code{net_serv.o} object file or the other MySQL libraries. When using a threaded client and you want to use timeouts and interrupts, you can make great use of the routines in the @file{thr_alarm.c} file. If you are using routines from the @code{mysys} library, the only thing you must remember is to call @code{my_init()} first! @xref{C Thread functions}. All functions except @code{mysql_real_connect()} are by default thread safe. The following notes describe how to compile a thread safe client library and use it in a thread-safe manner. (The notes below for @code{mysql_real_connect()} actually apply to @code{mysql_connect()} as well, but because @code{mysql_connect()} is deprecated, you should be using @code{mysql_real_connect()} anyway.) To make @code{mysql_real_connect()} thread safe, you must recompile the client library with this command: @example shell> ./configure --enable-thread-safe-client @end example This will create a thread-safe client library @code{libmysqlclient_r}. @code{--enable-thread-safe-client}. This library is thread safe per connection. You can let two threads share the same connection as long as you do the following: @itemize @bullet @item Two threads can't send a query to the MySQL at the same time on the same connection. In particular, you have to ensure that between a @code{mysql_query()} and @code{mysql_store_result()} no other thread is using the same connection. @item Many threads can access different result sets that are retrieved with @code{mysql_store_result()}. @item If you use @code{mysql_use_result}, you have to ensure that no other thread is asking anything on the same connection until the result set is closed. However, it really is best for threaded clients that share the same connection to use @code{mysql_use_result()}. @item If you want to use multiple threads on the same connection, you must have a mutex lock around your @code{mysql_query()} and @code{mysql_store_result()} call combination. Once @code{mysql_store_result()} is ready, the lock can be released and other threads may query the same connection. @item If you program with POSIX threads, you can use @code{pthread_mutex_lock()} and @code{pthread_mutex_unlock()} to establish and release a mutex lock. @end itemize You need to know the following if you have a thread that is calling MySQL functions, but that thread has not created the connection to the MySQL database: When you call @code{mysql_init()} or @code{mysql_connect()}, MySQL will create a thread specific variable for the thread that is used by the debug library (among other things). If you have in a thread call a MySQL function, before a thread has called @code{mysql_init()} or @code{mysql_connect()}, the thread will not have the necessary thread specific variables in place and you are likely to end up with a core dump sooner or later. The get things to work smoothly you have to do the following: @enumerate @item Call @code{my_init()} at the start of your program if it calls any other MySQL function before calling @code{mysql_real_connect()}. @item Call @code{my_thread_init()} in the thread handler before calling any MySQL function. @item In the thread, call @code{my_thread_end()} before calling @code{pthread_exit()}. This will free the memory used by MySQL thread specific variables. @end enumerate You may get some errors because of undefined symbols when linking your client with @code{mysqlclient_r}. In most cases this is because you haven't included the thread libraries on the link/compile line. @node Cplusplus, Java, C, Clients @section MySQL C++ APIs @menu * Borland C++:: Borland C++ @end menu @cindex C++ APIs Two APIs are available in the MySQL @uref{http://www.mysql.com/Downloads/Contrib/,Contrib directory}. @node Borland C++, , Cplusplus, Cplusplus @subsection Borland C++ @cindex Borland C++ compiler You can compile the MySQL Windows source with Borland C++ 5.02. (The Windows source includes only projects for Microsoft VC++, for Borland C++ you have to do the project files yourself). One known problem with Borland C++ is that it uses a different structure alignment than VC++. This means that you will run into problems if you try to use the default @code{libmysql.dll} libraries (that was compiled with VC++) with Borland C++. You can do one of the following to avoid this problem. @itemize @bullet @item You can use the static MySQL libraries for Borland C++ that you can find on @uref{http://www.mysql.com/downloads/os-win32.html}. @item Only call @code{mysql_init()} with @code{NULL} as an argument, not a pre-allocated MYSQL struct. @end itemize @node Java, Python, Cplusplus, Clients @section MySQL Java Connectivity (JDBC) @cindex Java connectivity @cindex JDBC There are 2 supported JDBC drivers for MySQL (the mm driver and the Reisin JDBC driver). You can find a copy of the mm driver at @uref{http://mmmysql.sourceforge.net/} or @uref{http://www.mysql.com/Downloads/Contrib/} and the Reisin driver at @uref{http://www.caucho.com/projects/jdbc-mysql/index.xtp} For documentation consult any JDBC documentation and the driver's own documentation for MySQL-specific features. @node Python, Tcl, Java, Clients @section MySQL Python APIs @cindex Python APIs The MySQL @uref{http://www.mysql.com/Downloads/Contrib/,Contrib directory} contains a Python interface written by Joseph Skinner. You can also use the Python interface to iODBC to access a MySQL server. @uref{http://starship.skyport.net/~lemburg/,mxODBC} @node Tcl, Eiffel, Python, Clients @section MySQL Tcl APIs @cindex Tcl APIs @uref{http://www.binevolve.com/~tdarugar/tcl-sql/, Tcl at binevolve}. The @uref{http://www.mysql.com/Downloads/Contrib,Contrib directory} contains a Tcl interface that is based on msqltcl 1.50. @node Eiffel, , Tcl, Clients @section MySQL Eiffel wrapper @cindex Eiffel Wrapper @cindex wrappers, Eiffel The MySQL @uref{http://www.mysql.com/Downloads/Contrib/,Contrib directory} contains an Eiffel wrapper written by Michael Ravits. You can also find this at: @url{http://www.netpedia.net/hosting/newplayer/} @node Extending MySQL, Problems, Clients, Top @chapter Extending MySQL @menu * Adding functions:: Adding New Functions to MySQL * Adding procedures:: Adding New Procedures to MySQL * MySQL internals:: MySQL Internals @end menu @node Adding functions, Adding procedures, Extending MySQL, Extending MySQL @section Adding New Functions to MySQL @cindex functions, new @cindex adding, new functions @cindex user-defined functions, adding @cindex UDFs, defined @cindex functions, user-defined There are two ways to add new functions to MySQL: @itemize @bullet @item You can add the function through the user-definable function (UDF) interface. User-definable functions are added and removed dynamically using the @code{CREATE FUNCTION} and @code{DROP FUNCTION} statements. @xref{CREATE FUNCTION, , @code{CREATE FUNCTION}}. @item You can add the function as a native (built in) MySQL function. Native functions are compiled into the @code{mysqld} server and become available on a permanent basis. @end itemize Each method has advantages and disadvantages: @itemize @bullet @item If you write a user-definable function, you must install the object file in addition to the server itself. If you compile your function into the server, you don't need to do that. @item You can add UDFs to a binary MySQL distribution. Native functions require you to modify a source distribution. @item If you upgrade your MySQL distribution, you can continue to use your previously installed UDFs. For native functions, you must repeat your modifications each time you upgrade. @end itemize Whichever method you use to add new functions, they may be used just like native functions such as @code{ABS()} or @code{SOUNDEX()}. @menu * CREATE FUNCTION:: @code{CREATE FUNCTION/DROP FUNCTION} Syntax * Adding UDF:: Adding a new user-definable function * Adding native function:: Adding a new native function @end menu @node CREATE FUNCTION, Adding UDF, Adding functions, Adding functions @subsection @code{CREATE FUNCTION/DROP FUNCTION} Syntax @findex CREATE FUNCTION @findex DROP FUNCTION @findex UDF functions @findex User-defined functions @findex Functions, user-defined @example CREATE [AGGREGATE] FUNCTION function_name RETURNS @{STRING|REAL|INTEGER@} SONAME shared_library_name DROP FUNCTION function_name @end example A user-definable function (UDF) is a way to extend MySQL with a new function that works like native (built in) MySQL functions such as @code{ABS()} and @code{CONCAT()}. @code{AGGREGATE} is a new option for MySQL Version 3.23. An @code{AGGREGATE} function works exactly like a native MySQL @code{GROUP} function like @code{SUM} or @code{COUNT()}. @code{CREATE FUNCTION} saves the function's name, type, and shared library name in the @code{mysql.func} system table. You must have the @strong{insert} and @strong{delete} privileges for the @code{mysql} database to create and drop functions. All active functions are reloaded each time the server starts, unless you start @code{mysqld} with the @code{--skip-grant-tables} option. In this case, UDF initialization is skipped and UDFs are unavailable. (An active function is one that has been loaded with @code{CREATE FUNCTION} and not removed with @code{DROP FUNCTION}.) For instructions on writing user-definable functions, see @ref{Adding functions}. For the UDF mechanism to work, functions must be written in C or C++, your operating system must support dynamic loading and you must have compiled @code{mysqld} dynamically (not statically). @node Adding UDF, Adding native function, CREATE FUNCTION, Adding functions @subsection Adding a New User-definable Function @cindex adding, user-definable functions @cindex user-defined functions, adding @cindex functions, user-definable, adding @menu * UDF calling sequences:: UDF calling sequences * UDF arguments:: Argument processing * UDF return values:: Return values and error handling * UDF compiling:: Compiling and installing user-definable functions @end menu For the UDF mechanism to work, functions must be written in C or C++ and your operating system must support dynamic loading. The MySQL source distribution includes a file @file{sql/udf_example.cc} that defines 5 new functions. Consult this file to see how UDF calling conventions work. For @code{mysqld} to be able to use UDF functions, you should configure MySQL with @code{--with-mysqld-ldflags=-rdynamic} The reason is that to on many platforms (including Linux) you can load a dynamic library (with @code{dlopen()}) from a static linked program, which you would get if you are using @code{--with-mysqld-ldflags=-all-static} If you want to use an UDF that needs to access symbols from @code{mysqld} (like the @code{methaphone} example in @file{sql/udf_example.cc} that uses @code{default_charset_info}), you must link the program with @code{-rdynamic}. (see @code{man dlopen}). For each function that you want to use in SQL statements, you should define corresponding C (or C++) functions. In the discussion below, the name ``xxx'' is used for an example function name. To distinquish between SQL and C/C++ usage, @code{XXX()} (uppercase) indicates a SQL function call, and @code{xxx()} (lowercase) indicates a C/C++ function call. The C/C++ functions that you write to implement the interface for @code{XXX()} are: @table @asis @item @code{xxx()} (required) The main function. This is where the function result is computed. The correspondence between the SQL type and return type of your C/C++ function is shown below: @multitable @columnfractions .2 .8 @item @strong{SQL type} @tab @strong{C/C++ type} @item @code{STRING} @tab @code{char *} @item @code{INTEGER} @tab @code{long long} @item @code{REAL} @tab @code{double} @end multitable @item @code{xxx_init()} (optional) The initialization function for @code{xxx()}. It can be used to: @itemize @bullet @item Check the number of arguments to @code{XXX()}. @item Check that the arguments are of a required type or, alternatively, tell MySQL to coerce arguments to the types you want when the main function is called. @item Allocate any memory required by the main function. @item Specify the maximum length of the result. @item Specify (for @code{REAL} functions) the maximum number of decimals. @item Specify whether or not the result can be @code{NULL}. @end itemize @item @code{xxx_deinit()} (optional) The deinitialization function for @code{xxx()}. It should deallocate any memory allocated by the initialization function. @end table When a SQL statement invokes @code{XXX()}, MySQL calls the initialization function @code{xxx_init()} to let it perform any required setup, such as argument checking or memory allocation. If @code{xxx_init()} returns an error, the SQL statement is aborted with an error message and the main and deinitialization functions are not called. Otherwise, the main function @code{xxx()} is called once for each row. After all rows have been processed, the deinitialization function @code{xxx_deinit()} is called so it can perform any required cleanup. All functions must be thread safe (not just the main function, but the initialization and deinitialization functions as well). This means that you are not allowed to allocate any global or static variables that change! If you need memory, you should allocate it in @code{xxx_init()} and free it in @code{xxx_deinit()}. @node UDF calling sequences, UDF arguments, Adding UDF, Adding UDF @subsubsection UDF Calling Sequences @cindex calling sequences, UDF The main function should be declared as shown below. Note that the return type and parameters differ, depending on whether you will declare the SQL function @code{XXX()} to return @code{STRING}, @code{INTEGER}, or @code{REAL} in the @code{CREATE FUNCTION} statement: @noindent For @code{STRING} functions: @example char *xxx(UDF_INIT *initid, UDF_ARGS *args, char *result, unsigned long *length, char *is_null, char *error); @end example @noindent For @code{INTEGER} functions: @example long long xxx(UDF_INIT *initid, UDF_ARGS *args, char *is_null, char *error); @end example @noindent For @code{REAL} functions: @example double xxx(UDF_INIT *initid, UDF_ARGS *args, char *is_null, char *error); @end example The initialization and deinitialization functions are declared like this: @example my_bool xxx_init(UDF_INIT *initid, UDF_ARGS *args, char *message); void xxx_deinit(UDF_INIT *initid); @end example The @code{initid} parameter is passed to all three functions. It points to a @code{UDF_INIT} structure that is used to communicate information between functions. The @code{UDF_INIT} structure members are listed below. The initialization function should fill in any members that it wishes to change. (To use the default for a member, leave it unchanged.): @table @code @item my_bool maybe_null @code{xxx_init()} should set @code{maybe_null} to @code{1} if @code{xxx()} can return @code{NULL}. The default value is @code{1} if any of the arguments are declared @code{maybe_null}. @item unsigned int decimals Number of decimals. The default value is the maximum number of decimals in the arguments passed to the main function. (For example, if the function is passed @code{1.34}, @code{1.345}, and @code{1.3}, the default would be 3, because @code{1.345} has 3 decimals. @item unsigned int max_length The maximum length of the string result. The default value differs depending on the result type of the function. For string functions, the default is the length of the longest argument. For integer functions, the default is 21 digits. For real functions, the default is 13 plus the number of decimals indicated by @code{initid->decimals}. (For numeric functions, the length includes any sign or decimal point characters.) @item char *ptr A pointer that the function can use for its own purposes. For example, functions can use @code{initid->ptr} to communicate allocated memory between functions. In @code{xxx_init()}, allocate the memory and assign it to this pointer: @example initid->ptr = allocated_memory; @end example In @code{xxx()} and @code{xxx_deinit()}, refer to @code{initid->ptr} to use or deallocate the memory. @end table @node UDF arguments, UDF return values, UDF calling sequences, Adding UDF @subsubsection Argument Processing @cindex argument processing @cindex processing, arguments The @code{args} parameter points to a @code{UDF_ARGS} structure that thas the members listed below: @table @code @item unsigned int arg_count The number of arguments. Check this value in the initialization function if you want your function to be called with a particular number of arguments. For example: @example if (args->arg_count != 2) @{ strcpy(message,"XXX() requires two arguments"); return 1; @} @end example @item enum Item_result *arg_type The types for each argument. The possible type values are @code{STRING_RESULT}, @code{INT_RESULT}, and @code{REAL_RESULT}. To make sure that arguments are of a given type and return an error if they are not, check the @code{arg_type} array in the initialization function. For example: @example if (args->arg_type[0] != STRING_RESULT || args->arg_type[1] != INT_RESULT) @{ strcpy(message,"XXX() requires a string and an integer"); return 1; @} @end example As an alternative to requiring your function's arguments to be of particular types, you can use the initialization function to set the @code{arg_type} elements to the types you want. This causes MySQL to coerce arguments to those types for each call to @code{xxx()}. For example, to specify coercion of the first two arguments to string and integer, do this in @code{xxx_init()}: @example args->arg_type[0] = STRING_RESULT; args->arg_type[1] = INT_RESULT; @end example @item char **args @code{args->args} communicates information to the initialization function about the general nature of the arguments your function was called with. For a constant argument @code{i}, @code{args->args[i]} points to the argument value. (See below for instructions on how to access the value properly.) For a non-constant argument, @code{args->args[i]} is @code{0}. A constant argument is an expression that uses only constants, such as @code{3} or @code{4*7-2} or @code{SIN(3.14)}. A non-constant argument is an expression that refers to values that may change from row to row, such as column names or functions that are called with non-constant arguments. For each invocation of the main function, @code{args->args} contains the actual arguments that are passed for the row currently being processed. Functions can refer to an argument @code{i} as follows: @itemize @bullet @item An argument of type @code{STRING_RESULT} is given as a string pointer plus a length, to allow handling of binary data or data of arbitrary length. The string contents are available as @code{args->args[i]} and the string length is @code{args->lengths[i]}. You should not assume that strings are null-terminated. @item For an argument of type @code{INT_RESULT}, you must cast @code{args->args[i]} to a @code{long long} value: @example long long int_val; int_val = *((long long*) args->args[i]); @end example @item For an argument of type @code{REAL_RESULT}, you must cast @code{args->args[i]} to a @code{double} value: @example double real_val; real_val = *((double*) args->args[i]); @end example @end itemize @item unsigned long *lengths For the initialization function, the @code{lengths} array indicates the maximum string length for each argument. For each invocation of the main function, @code{lengths} contains the actual lengths of any string arguments that are passed for the row currently being processed. For arguments of types @code{INT_RESULT} or @code{REAL_RESULT}, @code{lengths} still contains the maximum length of the argument (as for the initialization function). @end table @node UDF return values, UDF compiling, UDF arguments, Adding UDF @subsubsection Return Values and Error Handling @cindex UDFs, return values @cindex return values, UDFs @cindex errors, handling for UDFs @cindex handling, errors The initialization function should return @code{0} if no error occurred and @code{1} otherwise. If an error occurs, @code{xxx_init()} should store a null-terminated error message in the @code{message} parameter. The message will be returned to the client. The message buffer is @code{MYSQL_ERRMSG_SIZE} characters long, but you should try to keep the message to less than 80 characters so that it fits the width of a standard terminal screen. The return value of the main function @code{xxx()} is the function value, for @code{long long} and @code{double} functions. A string functions should return a pointer to the result and store the length of the string in the @code{length} arguments. @code{result} is a buffer at least 255 bytes long. Set these to the contents and length of the return value. For example: @example memcpy(result, "result string", 13); *length = 13; @end example If your string functions that needs to return a string longer than 255 bytes, you must allocate the space for it with @code{malloc()} in your @code{xxx_init()} function or your @code{xxx()} function and free it in your @code{xxx_deinit()} function. You can store the allocated memory in the @code{ptr} slot in the @code{UDF_INIT} structure for reuse by future @code{xxx()} calls. @xref{UDF calling sequences}. To indicate a return value of @code{NULL} in the main function, set @code{is_null} to @code{1}: @example *is_null = 1; @end example To indicate an error return in the main function, set the @code{error} parameter to @code{1}: @example *error = 1; @end example If @code{xxx()} sets @code{*error} to @code{1} for any row, the function value is @code{NULL} for the current row and for any subsequent rows processed by the statement in which @code{XXX()} was invoked. (@code{xxx()} will not even be called for subsequent rows.) @strong{NOTE:} In MySQL versions prior to 3.22.10, you should set both @code{*error} and @code{*is_null}: @example *error = 1; *is_null = 1; @end example @node UDF compiling, , UDF return values, Adding UDF @subsubsection Compiling and Installing User-definable Functions @cindex compiling, user-defined functions @cindex UDFs, compiling @cindex installing, user-defined functions Files implementing UDFs must be compiled and installed on the host where the server runs. This process is described below for the example UDF file @file{udf_example.cc} that is included in the MySQL source distribution. This file contains the following functions: @itemize @bullet @item @code{metaphon()} returns a metaphon string of the string argument. This is something like a soundex string, but it's more tuned for English. @item @code{myfunc_double()} returns the sum of the ASCII values of the characters in its arguments, divided by the sum of the length of its arguments. @item @code{myfunc_int()} returns the sum of the length of its arguments. @item @code{sequence([const int])} returns an sequence starting from the given number or 1 if no number has been given. @item @code{lookup()} returns the IP number for a hostname. @item @code{reverse_lookup()} returns the hostname for an IP number. The function may be called with a string @code{"xxx.xxx.xxx.xxx"} or four numbers. @end itemize A dynamically loadable file should be compiled as a sharable object file, using a command something like this: @example shell> gcc -shared -o udf_example.so myfunc.cc @end example You can easily find out the correct compiler options for your system by running this command in the @file{sql} directory of your MySQL source tree: @example shell> make udf_example.o @end example You should run a compile command similar to the one that @code{make} displays, except that you should remove the @code{-c} option near the end of the line and add @code{-o udf_example.so} to the end of the line. (On some systems, you may need to leave the @code{-c} on the command.) Once you compile a shared object containing UDFs, you must install it and tell MySQL about it. Compiling a shared object from @file{udf_example.cc} produces a file named something like @file{udf_example.so} (the exact name may vary from platform to platform). Copy this file to some directory searched by @code{ld}, such as @file{/usr/lib}. On many systems, you can set the @code{LD_LIBRARY} or @code{LD_LIBRARY_PATH} environment variable to point at the directory where you have your UDF function files. The @code{dlopen} manual page tells you which variable you should use on your system. You should set this in @code{mysql.server} or @code{safe_mysqld} and restart @code{mysqld}. After the library is installed, notify @code{mysqld} about the new functions with these commands: @example mysql> CREATE FUNCTION metaphon RETURNS STRING SONAME "udf_example.so"; mysql> CREATE FUNCTION myfunc_double RETURNS REAL SONAME "udf_example.so"; mysql> CREATE FUNCTION myfunc_int RETURNS INTEGER SONAME "udf_example.so"; mysql> CREATE FUNCTION lookup RETURNS STRING SONAME "udf_example.so"; mysql> CREATE FUNCTION reverse_lookup RETURNS STRING SONAME "udf_example.so"; @end example Functions can be deleted using @code{DROP FUNCTION}: @example mysql> DROP FUNCTION metaphon; mysql> DROP FUNCTION myfunc_double; mysql> DROP FUNCTION myfunc_int; mysql> DROP FUNCTION lookup; mysql> DROP FUNCTION reverse_lookup; @end example The @code{CREATE FUNCTION} and @code{DROP FUNCTION} statements update the system table @code{func} in the @code{mysql} database. The function's name, type and shared library name are saved in the table. You must have the @strong{insert} and @strong{delete} privileges for the @code{mysql} database to create and drop functions. You should not use @code{CREATE FUNCTION} to add a function that has already been created. If you need to reinstall a function, you should remove it with @code{DROP FUNCTION} and then reinstall it with @code{CREATE FUNCTION}. You would need to do this, for example, if you recompile a new version of your function, so that @code{mysqld} gets the new version. Otherwise the server will continue to use the old version. Active functions are reloaded each time the server starts, unless you start @code{mysqld} with the @code{--skip-grant-tables} option. In this case, UDF initialization is skipped and UDFs are unavailable. (An active function is one that has been loaded with @code{CREATE FUNCTION} and not removed with @code{DROP FUNCTION}.) @node Adding native function, , Adding UDF, Adding functions @subsection Adding a New Native Function @cindex adding, native functions @cindex native functions, adding @cindex functions, native, adding The procedure for adding a new native function is described below. Note that you cannot add native functions to a binary distribution because the procedure involves modifying MySQL source code. You must compile MySQL yourself from a source distribution. Also note that if you migrate to another version of MySQL (for example, when a new version is released), you will need to repeat the procedure with the new version. To add a new native MySQL function, follow these steps: @enumerate @item Add one line to @file{lex.h} that defines the function name in the @code{sql_functions[]} array. @item If the function prototype is simple (just takes zero, one, two or three arguments), you should in lex.h specify SYM(FUNC_ARG#) (where # is the number of arguments) as the second argument in the @code{sql_functions[]} array and add a function that creates a function object in @file{item_create.cc}. Take a look at @code{"ABS"} and @code{create_funcs_abs()} for an example of this. If the function prototype is complicated (for example takes a variable number of arguments), you should add two lines to @file{sql_yacc.yy}. One indicates the preprocessor symbol that @code{yacc} should define (this should be added at the beginning of the file). Then define the function parameters and add an ``item'' with these parameters to the @code{simple_expr} parsing rule. For an example, check all occurrences of @code{ATAN} in @file{sql_yacc.yy} to see how this is done. @item In @file{item_func.h}, declare a class inheriting from @code{Item_num_func} or @code{Item_str_func}, depending on whether your function returns a number or a string. @item In @file{item_func.cc}, add one of the following declarations, depending on whether you are defining a numeric or string function: @example double Item_func_newname::val() longlong Item_func_newname::val_int() String *Item_func_newname::Str(String *str) @end example If you inherit your object from any of the standard items (like @code{Item_num_func} you probably only have to define one of the above functions and let the parent object take care of the other functions. For example, the @code{Item_str_func} class defines a @code{val()} function that executes @code{atof()} on the value returned by @code{::str()}. @item You should probably also define the following object function: @example void Item_func_newname::fix_length_and_dec() @end example This function should at least calculate @code{max_length} based on the given arguments. @code{max_length} is the maximum number of characters the function may return. This function should also set @code{maybe_null = 0} if the main function can't return a @code{NULL} value. The function can check if any of the function arguments can return @code{NULL} by checking the arguments @code{maybe_null} variable. You can take a look at @code{Item_func_mod::fix_length_and_dec} for a typical example of how to do this. @end enumerate All functions must be thread safe (In other words, don't use any global or static variables in the functions without protecting them with mutexes). If you want to return @code{NULL}, from @code{::val()}, @code{::val_int()} or @code{::str()} you should set @code{null_value} to 1 and return 0. For @code{::str()} object functions, there are some additional considerations to be aware of: @itemize @bullet @item The @code{String *str} argument provides a string buffer that may be used to hold the result. (For more information about the @code{String} type, take a look at the @file{sql_string.h} file.) @item The @code{::str()} function should return the string that holds the result or @code{(char*) 0} if the result is @code{NULL}. @item All current string functions try to avoid allocating any memory unless absolutely necessary! @end itemize @node Adding procedures, MySQL internals, Adding functions, Extending MySQL @section Adding New Procedures to MySQL @cindex procedures, adding @cindex adding, procedures @cindex new procedures, adding In MySQL, you can define a procedure in C++ that can access and modify the data in a query before it is sent to the client. The modification can be done on row-by-row or @code{GROUP BY} level. We have created an example procedure in MySQL Version 3.23 to show you what can be done. Additionally we recommend you to take a look at 'mylua', which you can find in the Contrib directory. @xref{Contrib}. Which this you can use the LUA language to load a procedure at runtime into @code{mysqld}. @menu * procedure analyse:: Procedure analyse * Writing a procedure:: Writing a procedure. @end menu @node procedure analyse, Writing a procedure, Adding procedures, Adding procedures @subsection Procedure Analyse @code{analyse([max elements,[max memory]])} This procedure is defined in the @file{sql/sql_analyse.cc}. This examines the result from your query and returns an analysis of the results: @itemize @bullet @item @code{max elements} (default 256) is the maximum number of distinct values @code{analyse} will notice per column. This is used by @code{analyse} to check if the optimal column type should be of type @code{ENUM}. @item @code{max memory} (default 8192) is the maximum memory @code{analyse} should allocate per column while trying to find all distinct values. @end itemize @example SELECT ... FROM ... WHERE ... PROCEDURE ANALYSE([max elements,[max memory]]) @end example @node Writing a procedure, , procedure analyse, Adding procedures @subsection Writing a Procedure For the moment, the only documentation for this is the source. You can find all information about procedures by examining the following files: @itemize @bullet @item @file{sql/sql_analyse.cc} @item @file{sql/procedure.h} @item @file{sql/procedure.cc} @item @file{sql/sql_select.cc} @end itemize @node MySQL internals, , Adding procedures, Extending MySQL @section MySQL Internals @cindex internals @cindex threads This chapter describes a lot of things that you need to know when working on the MySQL code. If you plan to contribute to MySQL development, want to have access to the bleeding-edge in-between versions code, or just want to keep track of development, follow the instructions in @xref{Installing source tree}. If you are interested in MySQL internals, you should also subscribe to @email{internals@@lists.mysql.com}. This is a relatively low traffic list, in comparison with @email{mysql@@lists.mysql.com}. @menu * MySQL threads:: MySQL threads * MySQL test suite:: MySQL test suite @end menu @node MySQL threads, MySQL test suite, MySQL internals, MySQL internals @subsection MySQL Threads The MySQL server creates the following threads: @itemize @bullet @item The TCP/IP connection thread handles all connection requests and creates a new dedicated thread to handle the authentication and and SQL query processing for each connection. @item On Windows NT there is a named pipe handler thread that does the same work as the TCP/IP connection thread on named pipe connect requests. @item The signal thread handles all signals. This thread also normally handles alarms and calls @code{process_alarm()} to force timeouts on connections that have been idle too long. @item If @code{mysqld} is compiled with @code{-DUSE_ALARM_THREAD}, a dedicated thread that handles alarms is created. This is only used on some systems where there are problems with @code{sigwait()} or if one wants to use the @code{thr_alarm()} code in ones application without a dedicated signal handling thread. @item If one uses the @code{--flush_time=#} option, a dedicated thread is created to flush all tables at the given interval. @item Every connection has its own thread. @item Every different table on which one uses @code{INSERT DELAYED} gets its own thread. @item If you use @code{--master-host}, a slave replication thread will be started to read and apply updates from the master. @end itemize @code{mysqladmin processlist} only shows the connection, @code{INSERT DELAYED}, and replication threads. @node MySQL test suite, , MySQL threads, MySQL internals @subsection MySQL Test Suite @cindex mysqltest, MySQL Test Suite @cindex testing mysqld, mysqltest Until recently, our main full-coverage test suite was based on proprietary customer data and for that reason has not been publicly available. The only publicly available part of our testing process consisted of the @code{crash-me} test, a Perl DBI/DBD benchmark found in the @code{sql-bench} directory, and miscellaneous tests located in @code{tests} directory. The lack of a standardized publicly available test suite has made it difficult for our users, as well developers, to do regression tests on the MySQL code. To address this problem, we have created a new test system that is included in the source and binary distributions starting in Version 3.23.29. The current set of test cases doesn't test everything in MySQL, but it should catch most obvious bugs in the SQL processing code, OS/library issues, and is quite thorough in testing replication. Our eventual goal is to have the tests cover 100% of the code. We welcome contributions to our test suite. You may especially want to contribute tests that examine the functionality critical to your system, as this will ensure that all future MySQL releases will work well with your applications. @menu * running mysqltest:: Running the MySQL Test Suite * extending mysqltest:: Extending the MySQL Test Suite * Reporting mysqltest bugs:: Reporting Bugs in the MySQL Test Suite @end menu @node running mysqltest, extending mysqltest, MySQL test suite, MySQL test suite @subsubsection Running the MySQL Test Suite The test system consist of a test language interpreter (@code{mysqltest}), a shell script to run all tests(@code{mysql-test-run}), the actual test cases written in a special test language, and their expected results. To run the test suite on your system after a build, type @code{make test} or @code{mysql-test/mysql-test-run} from the source root. If you have installed a binary distribution, @code{cd} to the install root (eg. @code{/usr/local/mysql}), and do @code{scripts/mysql-test-run}. All tests should succeed. If not, you should try to find out why and report the problem if this is a bug in MySQL. @xref{Reporting mysqltest bugs}. If you have a copy of @code{mysqld} running on the machine where you want to run the test suite you do not have to stop it, as long as it is not using ports @code{9306} and @code{9307}. If one of those ports is taken, you should edit @code{mysql-test-run} and change the values of the master and/or slave port to one that is available. You can run one individual test case with @code{mysql-test/mysql-test-run test_name}. If one test fails, you should test running @code{mysql-test-run} with the @code{--force} option to check if any other tests fails. @node extending mysqltest, Reporting mysqltest bugs, running mysqltest, MySQL test suite @subsubsection Extending the MySQL Test Suite You can use the @code{mysqltest} language to write your own test cases. Unfortunately, we have not yet written full documentation for it - we plan to do this shortly. You can, however, look at our current test cases and use them as an example. The following points should help you get started: @itemize @bullet @item The tests are located in @code{mysql-test/t/*.test} @item A test case consists of @code{;} terminated statements and is similar to the input of @code{mysql} command line client. A statement by default is a query to be sent to MySQL server, unless it is recognized as internal command ( eg. @code{sleep} ). @item All queries that produce results, e.g. @code{SELECT}, @code{SHOW}, @code{EXPLAIN}, etc., must be preceded with @code{@@/path/to/result/file}. The file must contain the expected results. An easy way to generate the result file is to run @code{mysqltest -r < t/test-case-name.test} from @code{mysql-test} directory, and then edit the generated result files, if needed, to adjust them to the expected output. In that case, be very careful about not adding or deleting any invisible characters - make sure to only change the text and/or delete lines. If you have to insert a line, make sure the fields are separated with a hard tab, and there is a hard tab at the end. You may want to use @code{od -c} to make sure your text editor has not messed anything up during edit. We, of course, hope that you will never have to edit the output of @code{mysqltest -r} as you only have to do it when you find a bug. @item To be consistent with our setup, you should put your result files in @code{mysql-test/r} directory and name them @code{test_name.result}. If the test produces more than one result, you should use @code{test_name.a.result}, @code{test_name.b.result}, etc. @item If a statement returns an error, you should on the line before the statement specify with the @code{--error error-number}. The error number can be a list of possible error numbers separated with @code{','}. @item If you are writing a replication test case, you should on the first line of the test file, put @code{source include/master-slave.inc;}. To switch between master and slave, use @code{connection master;} and @code{connection slave;}. If you need to do something on an alternate connection, you can do @code{connection master1;} for the master, and @code{connection slave1;} for the slave. @item If you need to do something in a loop, you can use something like this: @example let $1=1000; while ($1) @{ # do your queries here dec $1; @} @end example @item To sleep between queries, use the @code{sleep} command. It supports fractions of a second, so you can do @code{sleep 1.3;}, for example, to sleep 1.3 seconds. @item To run the slave with additional options for your test case, put them in the command-line format in @code{mysql-test/t/test_name-slave.opt}. For the master, put them in @code{mysql-test/t/test_name-master.opt}. @item If you have a question about the test suite, or have a test case to contribute, e-mail to @email{internals@@lists.mysql.com}. As the list does not accept attachments, you should ftp all the relevant files to: @url{ftp://support.mysql.com/pub/mysql/Incoming} @end itemize @node Reporting mysqltest bugs, , extending mysqltest, MySQL test suite @subsubsection Reporting Bugs in the MySQL Test Suite If your MySQL version doesn't pass the test suite you should do the following: @itemize @bullet @item Don't send a bug report before you have found out as much as possible of what when wrong! When you do it, please use the @code{mysqlbug} script so that we can get information about your system and @code{MySQL} version. @xref{Bug reports}. @item Make sure to include the output of @code{mysql-test-run}, as well as contents of all @code{.reject} files in @code{mysql-test/r} directory. @item If a test in the test suite fails, check if the test fails also when run by its own: @example cd mysql-test mysql-test-run --local test-name @end example If this fails, then you should configure MySQL with @code{--with-debug} and run @code{mysql-test-run} with the @code{--debug} option. If this also fails send the trace file @file{var/tmp/master.trace} to ftp://support.mysql.com/pub/mysql/secret so that we can examine it. Please remember to also include a full description of your system, the version of the mysqld binary and how you compiled it. @item Try also to run @code{mysql-test-run} with the @code{--force} option to see if there is any other test that fails. @item If you have compiled MySQL yourself, check our manual for how to compile MySQL on your platform or, preferable, use one of the binaries we have compiled for you at @uref{http://www.mysql.com/downloads/}. All our standard binaries should pass the test suite ! @item If you get an error, like @code{Result length mismatch} or @code{Result content mismatch} it means that the output of the test didn't match exactly the expected output. This could be a bug in MySQL or that your mysqld version produces slight different results under some circumstances. Failed test results are put in a file with the same base name as the result file with the @code{.reject} extension. If your test case is failing, you should do a diff on the two files. If you cannot see how they are different, examine both with @code{od -c} and also check their lengths. @item If a test fails totally, you should check the logs file in the @code{mysql-test/var/log} directory for hints of what went wrong. @item If you have compiled MySQL with debugging you can try to debug this by running @code{mysql-test-run} with the @code{--gdb} and/or @code{--debug} options. @xref{Making trace files}. If you have not compiled MySQL for debugging you should probably do that. Just specify the @code{--with-debug} options to @code{configure}! @xref{Installing source}. @end itemize @node Problems, Users, Extending MySQL, Top @appendix Problems and Common Errors @cindex problems, common errors @cindex errors, common @menu * What is crashing:: How to determine what is causing problems * Common errors:: Some common errors when using MySQL * Installation Issues:: Installation Related Issues * Administration Issues:: Administration Related Issues * Query Issues:: Query Related Issues * Table Definition Issues:: Table Definition Related Issues @end menu This chapter lists some common problems and error messages that users have run into. You will learn how to figure out what the problem is, and what to do to solve it. You will also find proper solutions to some common problems. @node What is crashing, Common errors, Problems, Problems @appendixsec How to Determine What Is Causing Problems When you run into problems, the first thing you should do is to find out which program / piece of equipment is causing problems: @itemize @bullet @item If you have one of the following symptoms, then it is probably a hardware (like memory, motherboard, CPU, or hard disk) or kernel problem: @itemize @minus @item The keyboard doesn't work. This can normally be checked by pressing Caps Lock. If the Caps Lock light doesn't change you have to replace your keyboard. (Before doing this, you should try to reboot your computer and check all cables to the keyboard.) @item The mouse pointer doesn't move. @item The machine doesn't answer to a remote machine's pings. @item Different, unrelated programs don't behave correctly. @item If your system rebooted unexpectedly (a faulty user level program should NEVER be able to take down your system). @end itemize In this case you should start by checking all your cables and run some diagnostic tool to check your hardware! You should also check if there are any patches, updates, or service packs for your operating system that could likely solve your problems. Check also that all your libraries (like glibc) are up to date. It's always good to use a machine with ECC memory to discover memory problems early! @item If your keyboard is locked up, you may be able to fix this by logging into your machine from another machine and execute @code{kbd_mode -a} on it. @item Please examine your system log file (/var/log/messages or similar) for reasons for your problems. If you think the problem is in MySQL then you should also examine MySQL's log files. @xref{Update log}. @item If you don't think you have hardware problems, you should try to find out which program is causing problems. Try using @code{top}, @code{ps}, @code{taskmanager}, or some similar program, to check which program is taking all CPU or is locking the machine. @item Check with @code{top}, @code{df}, or a similar program if you are out of memory, disk space, open files, or some other critical resource. @item If the problem is some runaway process, you can always try to kill it. If it doesn't want to die, there is probably a bug in the operating system. @end itemize If after you have examined all other possibilities and you have concluded that it's the MySQL server or a MySQL client that is causing the problem, it's time to do a bug report for our mailing list or our support team. In the bug report, try to give a very detailed description of how the system is behaving and what you think is happening. You should also state why you think it's MySQL that is causing the problems. Take into consideration all the situations in this chapter. State any problems exactly how they appear when you examine your system. Use the 'cut and paste' method for any output and/or error messages from programs and/or log files! Try to describe in detail which program is not working and all symptoms you see! We have in the past received many bug reports that just state "the system doesn't work". This doesn't provide us with any information about what could be the problem. If a program fails, it's always useful to know: @itemize @bullet @item Has the program in question made a segmentation fault (core dumped)? @item Is the program taking the whole CPU? Check with @code{top}. Let the program run for a while, it may be evaluating something heavy. @item If it's the @code{mysqld} server that is causing problems, can you do @code{mysqladmin -u root ping} or @code{mysqladmin -u root processlist}? @item What does a client program say (try with @code{mysql}, for example) when you try to connect to the MySQL server? Does the client jam? Do you get any output from the program? @end itemize When sending a bug report, you should of follow the outlines described in this manual. @xref{Asking questions}. @node Common errors, Installation Issues, What is crashing, Problems @appendixsec Some Common Errors When Using MySQL @cindex errors, list of @menu * Error Access denied:: @code{Access denied} Error * Gone away:: @code{MySQL server has gone away} error * Can not connect to server:: @code{Can't connect to [local] MySQL server} error * Blocked host:: @code{Host '...' is blocked} error * Too many connections:: @code{Too many connections} error * Non-transactional tables:: @code{Some non-transactional changed tables couldn't be rolled back} Error * Out of memory:: @code{Out of memory} error * Packet too large:: @code{Packet too large} error * Communication errors:: Communication errors / Aborted connection * Full table:: @code{The table is full} error * Cannot create:: @code{Can't create/write to file} Error * Commands out of sync:: @code{Commands out of sync} error in client * Ignoring user:: @code{Ignoring user} error * Cannot find table:: @code{Table 'xxx' doesn't exist} error * Cannot initialize character set:: @code{Can@'t initialize character set xxx} error. * Not enough file handles:: File Not Found @end menu This section lists some errors that users frequently get. You will find descriptions of the errors, and how to solve the problem here. @node Error Access denied, Gone away, Common errors, Common errors @appendixsubsec @code{Access denied} Error @cindex errors, access denied @cindex problems, access denied errors @cindex access denied errors @xref{Privileges}, and especially. @xref{Access denied}. @node Gone away, Can not connect to server, Error Access denied, Common errors @appendixsubsec @code{MySQL server has gone away} Error This section also covers the related @code{Lost connection to server during query} error. The most common reason for the @code{MySQL server has gone away} error is that the server timed out and closed the connection. By default, the server closes the connection after 8 hours if nothing has happened. You can change the time limit by setting the @code{wait_timeout} variable when you start @code{mysqld}. Another common reason to receive the @code{MySQL server has gone away} error is because you have issued a ``close'' on your MySQL connection and then tried to run a query on the closed connection. You can check that the MySQL hasn't died by executing @code{mysqladmin version} and examining the uptime. If you have a script, you just have to issue the query again for the client to do an automatic reconnection. You normally can get the following error codes in this case (which one you get is OS-dependent): @multitable @columnfractions .3 .7 @item @code{CR_SERVER_GONE_ERROR} @tab The client couldn't send a question to the server. @item @code{CR_SERVER_LOST} @tab The client didn't get an error when writing to the server, but it didn't get a full answer (or any answer) to the question. @end multitable You can also get these errors if you send a query to the server that is incorrect or too large. If @code{mysqld} gets a packet that is too large or out of order, it assumes that something has gone wrong with the client and closes the connection. If you need big queries (for example, if you are working with big @code{BLOB} columns), you can increase the query limit by starting @code{mysqld} with the @code{-O max_allowed_packet=#} option (default 1M). The extra memory is allocated on demand, so @code{mysqld} will use more memory only when you issue a big query or when @code{mysqld} must return a big result row! @node Can not connect to server, Blocked host, Gone away, Common errors @appendixsubsec @code{Can't connect to [local] MySQL server} error A MySQL client on Unix can connect to the @code{mysqld} server in two different ways: Unix sockets, which connect through a file in the file system (default @file{/tmp/mysqld.sock}) or TCP/IP, which connects through a port number. Unix sockets are faster than TCP/IP but can only be used when connecting to a server on the same computer. Unix sockets are used if you don't specify a hostname or if you specify the special hostname @code{localhost}. On Windows you can connect only with TCP/IP if the @code{mysqld} server is running on Win95/Win98. If mysqld is running on NT and started with @code{enable-named-pipe}, you can also connect with named pipes. The name of the named pipe is MySQL. If you don't give a hostname when connecting to @code{mysqld}, a MySQL client will first try to connect to the named pipe, and if this doesn't work it will connect to the TCP/IP port. You can force the use of named pipes on Windows by using @code{.} as the hostname. The error (2002) @code{Can't connect to ...} normally means that there isn't a MySQL server running on the system or that you are using a wrong socket file or TCP/IP port when trying to connect to the @code{mysqld} server. Start by checking (using @code{ps} or the task manager on Windows) that there is a process running named @code{mysqld} on your server! If there isn't any @code{mysqld} process, you should start one. @xref{Starting server}. If a @code{mysqld} process is running, you can check the server by trying these different connections (the port number and socket pathname might be different in your setup, of course): @example shell> mysqladmin version shell> mysqladmin variables shell> mysqladmin -h `hostname` version variables shell> mysqladmin -h `hostname` --port=3306 version shell> mysqladmin -h 'ip for your host' version shell> mysqladmin --socket=/tmp/mysql.sock version @end example Note the use of backquotes rather than forward quotes with the @code{hostname} command; these cause the output of @code{hostname} (that is, the current hostname) to be substituted into the @code{mysqladmin} command. Here are some reasons the @code{Can't connect to local MySQL server} error might occur: @itemize @bullet @item @code{mysqld} is not running. @item You are running on a system that uses MIT-pthreads. If you are running on a system that doesn't have native threads, @code{mysqld} uses the MIT-pthreads package. @xref{Which OS}. However, all MIT-pthreads versions doesn't support Unix sockets. On a system without sockets support you must always specify the hostname explicitly when connecting to the server. Try using this command to check the connection to the server: @example shell> mysqladmin -h `hostname` version @end example @item Someone has removed the Unix socket that @code{mysqld} uses (default @file{/tmp/mysqld.sock}). You might have a @code{cron} job that removes the MySQL socket (for example, a job that removes old files from the @file{/tmp} directory). You can always run @code{mysqladmin version} and check that the socket @code{mysqladmin} is trying to use really exists. The fix in this case is to change the @code{cron} job to not remove @file{mysqld.sock} or to place the socket somewhere else. @xref{Problems with mysql.sock}. @item You have started the @code{mysqld} server with the @code{--socket=/path/to/socket} option. If you change the socket pathname for the server, you must also notify the MySQL clients about the new path. You can do this by providing the socket path as an argument to the client. @xref{Problems with mysql.sock}. @item You are using Linux and one thread has died (core dumped). In this case you must kill the other @code{mysqld} threads (for example, with the @code{mysql_zap} script before you can start a new MySQL server. @xref{Crashing}. @item You may not have read and write privilege to either the directory that holds the socket file or privilege to the socket file itself. In this case you have to either change the privilege for the directory / file or restart @code{mysqld} so that it uses a directory that you can access. @end itemize If you get the error message @code{Can't connect to MySQL server on some_hostname}, you can try the following things to find out what the problem is : @itemize @bullet @item Check if the server is up by doing @code{telnet your-host-name tcp-ip-port-number} and press @code{RETURN} a couple of times. If there is a MySQL server running on this port you should get a responses that includes the version number of the running MySQL server. If you get an error like @code{telnet: Unable to connect to remote host: Connection refused}, then there is no server running on the given port. @item Try connecting to the @code{mysqld} daemon on the local machine and check the TCP/IP port that @code{mysqld} it's configured to use (variable @code{port}) with @code{mysqladmin variables}. @item Check that your @code{mysqld} server is not started with the @code{--skip-networking} option. @end itemize @node Blocked host, Too many connections, Can not connect to server, Common errors @appendixsubsec @code{Host '...' is blocked} Error If you get an error like this: @example Host 'hostname' is blocked because of many connection errors. Unblock with 'mysqladmin flush-hosts' @end example this means that @code{mysqld} has gotten a lot (@code{max_connect_errors}) of connect requests from the host @code{'hostname'} that have been interrupted in the middle. After @code{max_connect_errors} failed requests, @code{mysqld} assumes that something is wrong (like an attack from a cracker), and blocks the site from further connections until someone executes the command @code{mysqladmin flush-hosts}. By default, @code{mysqld} blocks a host after 10 connection errors. You can easily adjust this by starting the server like this: @example shell> safe_mysqld -O max_connect_errors=10000 & @end example Note that if you get this error message for a given host, you should first check that there isn't anything wrong with TCP/IP connections from that host. If your TCP/IP connections aren't working, it won't do you any good to increase the value of the @code{max_connect_errors} variable! @node Too many connections, Non-transactional tables, Blocked host, Common errors @appendixsubsec @code{Too many connections} Error If you get the error @code{Too many connections} when you try to connect to MySQL, this means that there is already @code{max_connections} clients connected to the @code{mysqld} server. If you need more connections than the default (100), then you should restart @code{mysqld} with a bigger value for the @code{max_connections} variable. Note that @code{mysqld} actually allows (@code{max_connections}+1) clients to connect. The last connection is reserved for a user with the @strong{process} privilege. By not giving this privilege to normal users (they shouldn't need this), an administrator with this privilege can log in and use @code{SHOW PROCESSLIST} to find out what could be wrong. @xref{SHOW}. The maximum number of connects MySQL is depending on how good the thread library is on a given platform. Linux or Solaris should be able to support 500-1000 simultaneous connections, depending on how much RAM you have and what your clients are doing. @node Non-transactional tables, Out of memory, Too many connections, Common errors @appendixsubsec @code{Some non-transactional changed tables couldn't be rolled back} Error @cindex Non-transactional tables If you get the error/warning: @code{Warning: Some non-transactional changed tables couldn't be rolled back} when trying to do a @code{ROLLBACK}, this means that some of the tables you used in the transaction didn't support transactions. These non-transactional tables will not be affected by the @code{ROLLBACK} statement. The most typical case when this happens is when you have tried to create a table of a type that is not supported by your @code{mysqld} binary. If @code{mysqld} doesn't support a table type (or if the table type is disabled by a startup option) , it will instead create the table type with the table type that is most resembles to the one you requested, probably @code{MyISAM}. You can check the table type for a table by doing: @code{SHOW TABLE STATUS LIKE 'table_name'}. @xref{SHOW TABLE STATUS}. You can check the extensions your @code{mysqld} binary supports by doing: @code{show variables like 'have_%'}. @xref{SHOW VARIABLES}. @node Out of memory, Packet too large, Non-transactional tables, Common errors @appendixsubsec @code{Out of memory} Error If you issue a query and get something like the following error: @example mysql: Out of memory at line 42, 'malloc.c' mysql: needed 8136 byte (8k), memory in use: 12481367 bytes (12189k) ERROR 2008: MySQL client ran out of memory @end example note that the error refers to the MySQL client @code{mysql}. The reason for this error is simply that the client does not have enough memory to store the whole result. To remedy the problem, first check that your query is correct. Is it reasonable that it should return so many rows? If so, you can use @code{mysql --quick}, which uses @code{mysql_use_result()} to retrieve the result set. This places less of a load on the client (but more on the server). @node Packet too large, Communication errors, Out of memory, Common errors @appendixsubsec @code{Packet too large} Error When a MySQL client or the @code{mysqld} server gets a packet bigger than @code{max_allowed_packet} bytes, it issues a @code{Packet too large} error and closes the connection. If you are using the @code{mysql} client, you may specify a bigger buffer by starting the client with @code{mysql --set-variable=max_allowed_packet=8M}. If you are using other clients that do not allow you to specify the maximum packet size (such as @code{DBI}), you need to set the packet size when you start the server. You cau use a command-line option to @code{mysqld} to set @code{max_allowed_packet} to a larger size. For example, if you are expecting to store the full length of a @code{BLOB} into a table, you'll need to start the server with the @code{--set-variable=max_allowed_packet=16M} option. You can also get strange problems with large packets if you are using big blobs, but you haven't given @code{mysqld} access to enough memory to handle the query. If you suspect this is the case, try adding @code{ulimit -d 256000} to the beginning of the @code{safe_mysqld} script and restart @code{mysqld}. @node Communication errors, Full table, Packet too large, Common errors @appendixsubsec Communication Errors / Aborted Connection @cindex aborted clients @cindex aborted connection @cindex connection, aborted Starting with @code{MySQL 3.23.40} you only get the @code{Aborted connection} error of you start @code{mysqld} with @code{--warnings}. If you find errors like the following in your error log. @example 010301 14:38:23 Aborted connection 854 to db: 'users' user: 'josh' @end example @xref{Error log}. This means that something of the following has happened: @itemize @bullet @item The client program did not call @code{mysql_close()} before exit. @item The client had been sleeping more than @code{wait_timeout} or @code{interactive_timeout} without doing any requests. @xref{SHOW VARIABLES}. @item The client program ended abruptly in the middle of the transfer. @end itemize When the above happens, the server variable @code{Aborted_clients} is incremented. The server variable @code{Aborted_connects} is incremented when: @itemize @bullet @item When a connection packet doesn't contain the right information. @item When the user didn't have privileges to connect to a database. @item When a user uses a wrong password. @item When it takes more than @code{connect_timeout} seconds to get a connect package. @end itemize Note that the above could indicate that someone is trying to break into your database! @xref{SHOW VARIABLES}. Other reasons for problems with Aborted clients / Aborted connections. @itemize @bullet @item Usage of duplex Ethernet protocol, both half and full with Linux. Many Linux Ethernet drivers have this bug. You should test for this bug by transferring a huge file via ftp between these two machines. If a transfer goes in burst-pause-burst-pause ... mode then you are experiencing a Linux duplex syndrome. The only solution to this problem is switching of both half and full duplexing on hubs and switches. @item Some problem with the thread library that causes interrupts on reads. @item Badly configured TCP/IP. @item Faulty Ethernets or hubs or switches, cables ... This can be diagnosed properly only by replacing hardware. @item @code{max_allowed_packet} is too small or queries require more memory than you have alloacated for @code{mysqld}. @xref{Packet too large}. @end itemize @node Full table, Cannot create, Communication errors, Common errors @appendixsubsec @code{The table is full} Error @cindex table is full This error occurs in older MySQL versions when an in-memory temporary table becomes larger than @code{tmp_table_size} bytes. To avoid this problem, you can use the @code{-O tmp_table_size=#} option to @code{mysqld} to increase the temporary table size or use the SQL option @code{SQL_BIG_TABLES} before you issue the problematic query. @xref{SET OPTION, , @code{SET OPTION}}. You can also start @code{mysqld} with the @code{--big-tables} option. This is exactly the same as using @code{SQL_BIG_TABLES} for all queries. In MySQL Version 3.23, in-memory temporary tables will automatically be converted to a disk-based @code{MyISAM} table after the table size gets bigger than @code{tmp_table_size}. @node Cannot create, Commands out of sync, Full table, Common errors @appendixsubsec @code{Can't create/write to file} Error @cindex can't create/write to file If you get an error for some queries of type: @example Can't create/write to file '\\sqla3fe_0.ism'. @end example this means that MySQL can't create a temporary file for the result set in the given temporary directory. (The above error is a typical error message on Windows, and the Unix error message is similar.) The fix is to start @code{mysqld} with @code{--tmpdir=path} or to add to your option file: @example [mysqld] tmpdir=C:/temp @end example assuming that the @file{c:\\temp} directory exists. @xref{Option files}. Check also the error code that you get with @code{perror}. One reason may also be a disk full error; @example shell> perror 28 Error code 28: No space left on device @end example @node Commands out of sync, Ignoring user, Cannot create, Common errors @appendixsubsec @code{Commands out of sync} Error in Client @cindex commands out of sync If you get @code{Commands out of sync; You can't run this command now} in your client code, you are calling client functions in the wrong order! This can happen, for example, if you are using @code{mysql_use_result()} and try to execute a new query before you have called @code{mysql_free_result()}. It can also happen if you try to execute two queries that return data without a @code{mysql_use_result()} or @code{mysql_store_result()} in between. @node Ignoring user, Cannot find table, Commands out of sync, Common errors @appendixsubsec @code{Ignoring user} Error If you get the following error: @code{Found wrong password for user: 'some_user@@some_host'; Ignoring user} this means that when @code{mysqld} was started or when it reloaded the permissions tables, it found an entry in the @code{user} table with an invalid password. As a result, the entry is simply ignored by the permission system. Possible causes of and fixes for this problem: @itemize @bullet @item You may be running a new version of @code{mysqld} with an old @code{user} table. You can check this by executing @code{mysqlshow mysql user} to see if the password field is shorter than 16 characters. If so, you can correct this condition by running the @code{scripts/add_long_password} script. @item The user has an old password (8 characters long) and you didn't start @code{mysqld} with the @code{--old-protocol} option. Update the user in the @code{user} table with a new password or restart @code{mysqld} with @code{--old-protocol}. @item @findex PASSWORD() You have specified a password in the @code{user} table without using the @code{PASSWORD()} function. Use @code{mysql} to update the user in the @code{user} table with a new password. Make sure to use the @code{PASSWORD()} function: @example mysql> update user set password=PASSWORD('your password') where user='XXX'; @end example @end itemize @node Cannot find table, Cannot initialize character set, Ignoring user, Common errors @appendixsubsec @code{Table 'xxx' doesn't exist} Error If you get the error @code{Table 'xxx' doesn't exist} or @code{Can't find file: 'xxx' (errno: 2)}, this means that no table exists in the current database with the name @code{xxx}. Note that as MySQL uses directories and files to store databases and tables, the database and table names are @strong{case sensitive}! (On Windows the databases and tables names are not case sensitive, but all references to a given table within a query must use the same case!) You can check which tables you have in the current database with @code{SHOW TABLES}. @xref{SHOW, , @code{SHOW}}. @node Cannot initialize character set, Not enough file handles, Cannot find table, Common errors @appendixsubsec @code{Can@'t initialize character set xxx} error. @cindex multibyte character sets If you get an error like: @example MySQL Connection Failed: Can't initialize character set xxx @end example This means one of the following things: @itemize @bullet @item The character set is a multi-byte character set and you have not support for the character set in the client. In this case you need to recompile the client with @code{--with-charset=xxx} or with @code{--with-extra-charsets=xxx}. @xref{configure options}. All standard MySQL binaries are compiled with @code{--with-extra-character-sets=complex} which will enable support for all multi-byte character sets. @xref{Character sets}. @item The character set is a simple character set which is not compiled into @code{mysqld} and the character set definition files is not in the place where the client expect to find them. In this case you need to: @itemize @bullet @item Recompile the client with support for the character set. @xref{configure options}. @item Specify to the client where the character set definition files are. For many client you can do this with the @code{--character-sets-dir=path-to-charset-dir} option. @item Copy the character definition files to the path where the client expect them to be. @end itemize @end itemize @node Not enough file handles, , Cannot initialize character set, Common errors @appendixsubsec File Not Found If you get @code{ERROR '...' not found (errno: 23)}, @code{Can't open file: ... (errno: 24)}, or any other error with @code{errno 23} or @code{errno 24} from MySQL, it means that you haven't allocated enough file descriptors for MySQL. You can use the @code{perror} utility to get a description of what the error number means: @example shell> perror 23 File table overflow shell> perror 24 Too many open files shell> perror 11 Resource temporarily unavailable @end example The problem here is that @code{mysqld} is trying to keep open too many files simultaneously. You can either tell @code{mysqld} not to open so many files at once or increase the number of file descriptors available to @code{mysqld}. To tell @code{mysqld} to keep open fewer files at a time, you can make the table cache smaller by using the @code{-O table_cache=32} option to @code{safe_mysqld} (the default value is 64). Reducing the value of @code{max_connections} will also reduce the number of open files (the default value is 90). @tindex ulimit To change the number of file descriptors available to @code{mysqld}, you can use the option @code{--open-files-limit=#} to @code{safe_mysqld} or @code{-O open-files-limit=#} to @code{mysqld}. @xref{SHOW VARIABLES}. The easiest way to do that is to add the option to your option file. @xref{Option files}. If you have an old @code{mysqld} version that doesn't support this, you can edit the @code{safe_mysqld} script. There is a commented-out line @code{ulimit -n 256} in the script. You can remove the @code{'#'} character to uncomment this line, and change the number 256 to affect the number of file descriptors available to @code{mysqld}. @code{ulimit} (and @code{open-files-limit}) can increase the number of file descriptors, but only up to the limit imposed by the operating system. There is also a 'hard' limit that can only be overrided if you start @code{safe_mysqld} or @code{mysqld} as root (Just remember that you need to also use the @code{--user=..} option in this case). If you need to increase the OS limit on the number of file descriptors available to each process, consult the documentation for your operating system. Note that if you run the @code{tcsh} shell, @code{ulimit} will not work! @code{tcsh} will also report incorrect values when you ask for the current limits! In this case you should start @code{safe_mysqld} with @code{sh}! @node Installation Issues, Administration Issues, Common errors, Problems @appendixsec Installation Related Issues @menu * Link errors:: Problems When Linking with the MySQL Client Library * Changing MySQL user:: How to Run MySQL As a Normal User * File permissions :: Problems with File Permissions @end menu @node Link errors, Changing MySQL user, Installation Issues, Installation Issues @appendixsubsec Problems When Linking with the MySQL Client Library @cindex linking, errors @cindex errors, linking @cindex problems, linking If you are linking your program and you get errors for unreferenced symbols that start with @code{mysql_}, like the following: @example /tmp/ccFKsdPa.o: In function `main': /tmp/ccFKsdPa.o(.text+0xb): undefined reference to `mysql_init' /tmp/ccFKsdPa.o(.text+0x31): undefined reference to `mysql_real_connect' /tmp/ccFKsdPa.o(.text+0x57): undefined reference to `mysql_real_connect' /tmp/ccFKsdPa.o(.text+0x69): undefined reference to `mysql_error' /tmp/ccFKsdPa.o(.text+0x9a): undefined reference to `mysql_close' @end example you should be able to solve this by adding @code{-Lpath-to-the-mysql-library -lmysqlclient} @strong{LAST} on your link line. If you get @code{undefined reference} errors for the @code{uncompress} or @code{compress} function, add @code{-lz} @strong{LAST} on your link line and try again! If you get @code{undefined reference} errors for functions that should exist on your system, like @code{connect}, check the man page for the function in question, for which libraries you should add to the link line! If you get @code{undefined reference} errors for functions that don't exist on your system, like the following: @example mf_format.o(.text+0x201): undefined reference to `__lxstat' @end example it usually means that your library is compiled on a system that is not 100 % compatible with yours. In this case you should download the latest MySQL source distribution and compile this yourself. @xref{Installing source}. If you are trying to run a program and you then get errors for unreferenced symbols that start with @code{mysql_} or that the @code{mysqlclient} library can't be found, this means that your system can't find the share @code{libmysqlclient.so} library. The fix for this is to tell your system to search after shared libraries where the library is located by one of the following methods: @itemize @bullet @item Add the path to the directory where you have @code{libmysqlclient.so} the @code{LD_LIBRARY_PATH} environment variable. @item Add the path to the directory where you have @code{libmysqlclient.so} the @code{LD_LIBRARY} environment variable. @item Copy @code{libmysqlclient.so} to some place that is searched by your system, like @file{/lib}, and update the shared library information by executing @code{ldconfig}. @end itemize Another way to solve this problem is to link your program statically, with @code{-static}, or by removing the dynamic MySQL libraries before linking your code. In the second case you should be sure that no other programs are using the dynamic libraries! @node Changing MySQL user, File permissions , Link errors, Installation Issues @appendixsubsec How to Run MySQL As a Normal User @cindex starting, @code{mysqld} @cindex @code{mysqld}, starting The MySQL server @code{mysqld} can be started and run by any user. In order to change @code{mysqld} to run as a Unix user @code{user_name}, you must do the following: @enumerate @item Stop the server if it's running (use @code{mysqladmin shutdown}). @item Change the database directories and files so that @code{user_name} has privileges to read and write files in them (you may need to do this as the Unix @code{root} user): @example shell> chown -R user_name /path/to/mysql/datadir @end example If directories or files within the MySQL data directory are symlinks, you'll also need to follow those links and change the directories and files they point to. @code{chown -R} may not follow symlinks for you. @item Start the server as user @code{user_name}, or, if you are using MySQL Version 3.22 or later, start @code{mysqld} as the Unix @code{root} user and use the @code{--user=user_name} option. @code{mysqld} will switch to run as the Unix user @code{user_name} before accepting any connections. @item To start the server as the given user name automatically at system startup time, add a @code{user} line that specifies the user name to the @code{[mysqld]} group of the @file{/etc/my.cnf} option file or the @file{my.cnf} option file in the server's data directory. For example: @example [mysqld] user=user_name @end example @end enumerate At this point, your @code{mysqld} process should be running fine and dandy as the Unix user @code{user_name}. One thing hasn't changed, though: the contents of the permissions tables. By default (right after running the permissions table install script @code{mysql_install_db}), the MySQL user @code{root} is the only user with permission to access the @code{mysql} database or to create or drop databases. Unless you have changed those permissions, they still hold. This shouldn't stop you from accessing MySQL as the MySQL @code{root} user when you're logged in as a Unix user other than @code{root}; just specify the @code{-u root} option to the client program. Note that accessing MySQL as @code{root}, by supplying @code{-u root} on the command line, has @emph{nothing} to do with MySQL running as the Unix @code{root} user, or, indeed, as another Unix user. The access permissions and user names of MySQL are completely separate from Unix user names. The only connection with Unix user names is that if you don't provide a @code{-u} option when you invoke a client program, the client will try to connect using your Unix login name as your MySQL user name. If your Unix box itself isn't secured, you should probably at least put a password on the MySQL @code{root} users in the access tables. Otherwise, any user with an account on that machine can run @code{mysql -u root db_name} and do whatever he likes. @node File permissions , , Changing MySQL user, Installation Issues @appendixsubsec Problems with File Permissions @cindex files, permissions @cindex error mesaages, can't find file @cindex files, not found message If you have problems with file permissions, for example, if @code{mysql} issues the following error message when you create a table: @example ERROR: Can't find file: 'path/with/filename.frm' (Errcode: 13) @end example @tindex UMASK environment variable @tindex Environment variable, UMASK then the environment variable @code{UMASK} might be set incorrectly when @code{mysqld} starts up. The default umask value is @code{0660}. You can change this behavior by starting @code{safe_mysqld} as follows: @example shell> UMASK=384 # = 600 in octal shell> export UMASK shell> /path/to/safe_mysqld & @end example @tindex UMASK_DIR environment variable @tindex Environment variable, UMASK_DIR By default MySQL will create database and @code{RAID} directories with permission type 0700. You can modify this behavior by setting the @code{UMASK_DIR} variable. If you set this, new directories are created with the combined @code{UMASK} and @code{UMASK_DIR}. For example, if you want to give group access to all new directories, you can do: @example shell> UMASK_DIR=504 # = 770 in octal shell> export UMASK_DIR shell> /path/to/safe_mysqld & @end example In MySQL Version 3.23.25 and above, MySQL assumes that the value for @code{UMASK} and @code{UMASK_DIR} is in octal if it starts with a zero. @xref{Environment variables}. @node Administration Issues, Query Issues, Installation Issues, Problems @appendixsec Administration Related Issues @menu * Crashing:: What To Do If MySQL Keeps Crashing * Resetting permissions:: How to Reset a Forgotten Password * Full disk:: How MySQL Handles a Full Disk * Temporary files:: Where MySQL Stores Temporary Files * Problems with mysql.sock:: How to Protect @file{/tmp/mysql.sock} from Being Deleted * Timezone problems:: Time Zone Problems @end menu @node Crashing, Resetting permissions, Administration Issues, Administration Issues @appendixsubsec What To Do If MySQL Keeps Crashing @cindex crash, repeated All MySQL versions are tested on many platforms before they are released. This doesn't mean that there aren't any bugs in MySQL, but it means if there are bugs, they are very few and can be hard to find. If you have a problem, it will always help if you try to find out exactly what crashes your system, as you will have a much better chance of getting this fixed quickly. First, you should try to find out whether the problem is that the @code{mysqld} daemon dies or whether your problem has to do with your client. You can check how long your @code{mysqld} server has been up by executing @code{mysqladmin version}. If @code{mysqld} has died, you may find the reason for this in the file @file{mysql-data-directory/`hostname`.err}. @xref{Error log}. Many crashes of MySQL are caused by corrupted index / data files. MySQL will update the data on disk, with the @code{write()} system call, after every SQL statement and before the client is notified about the result. (This is not true if you are running with @code{delayed_key_writes}, in which case only the data is written.) This means that the data is safe even if @code{mysqld} crashes, as the OS will ensure that the not flushed data is written to disk. You can force MySQL to sync everything to disk after every SQL command by starting @code{mysqld} with @code{--flush}. The above means that normally you shouldn't get corrupted tables unless: @itemize @bullet @item Someone/something killed @code{mysqld} or the machine in the middle of an update. @item You have found a bug in @code{mysqld} that caused it to die in the middle of an update. @item Someone is manipulating the data/index files outside of @strong{mysqld} without locking the table properly. @item If you are running many @code{mysqld} servers on the same data on a system that doesn't support good file system locks (normally handled by the @code{lockd} daemon ) or if you are running multiple servers with @code{--skip-locking} @item You have a crashed index/data file that contains very wrong data that got @code{mysqld} confused. @item You have found a bug in the data storage code. This isn't that likely, but it's at least possible. In this case you can try to change the file type to another database handler by using @code{ALTER TABLE} on a repaired copy of the table! @end itemize Because it is very difficult to know why something is crashing, first try to check whether or not things that work for others crash for you. Please try the following things: @itemize @bullet @item Take down the @code{mysqld} daemon with @code{mysqladmin shutdown}, run @code{myisamchk --silent --force */*.MYI} on all tables, and restart the @code{mysqld} daemon. This will ensure that you are running from a clean state. @xref{MySQL Database Administration}. @item Use @code{mysqld --log} and try to determine from the information in the log whether or not some specific query kills the server. About 95% of all bugs are related to a particular query! Normally this is one of the last queries in the log file just before MySQL restarted. @xref{Query log}. If you can repeatadly kill MySQL with one of the queries, even when you have checked all tables just before doing the query, then you have been able to locate the bug and should do a bug report for this! @xref{Bug reports}. @item Try to make a test case that we can use to reproduce the problem. @xref{Reproduceable test case}. @item Try running the included mysql-test test and the MySQL benchmarks. @xref{MySQL test suite}. They should test MySQL rather well. You can also add code that to the benchmarks to simulates your application! The benchmarks can be found in the @file{bench} directory in the source distribution or, for a binary distribution, in the @file{sql-bench} directory under your MySQL installation directory. @item Try @code{fork_test.pl} and @code{fork2_test.pl}. @item If you configure MySQL for debugging, it will be much easier to gather information about possible errors if something goes wrong. Reconfigure MySQL with the @code{--with-debug} option or @code{--with-debug=full} to @code{configure} and then recompile. @xref{Debugging server}. @item Configuring MySQL for debugging causes a safe memory allocator to be included that can find some errors. It also provides a lot of output about what is happening. @item Have you applied the latest patches for your operating system? @item Use the @code{--skip-locking} option to @code{mysqld}. On some systems, the @code{lockd} lock manager does not work properly; the @code{--skip-locking} option tells @code{mysqld} not to use external locking. (This means that you cannot run 2 @code{mysqld} servers on the same data and that you must be careful if you use @code{myisamchk}, but it may be instructive to try the option as a test.) @item Have you tried @code{mysqladmin -u root processlist} when @code{mysqld} appears to be running but not responding? Sometimes @code{mysqld} is not comatose even though you might think so. The problem may be that all connections are in use, or there may be some internal lock problem. @code{mysqladmin processlist} will usually be able to make a connection even in these cases, and can provide useful information about the current number of connections and their status. @item Run the command @code{mysqladmin -i 5 status} or @code{mysqladmin -i 5 -r status} or in a separate window to produce statistics while you run your other queries. @item Try the following: @enumerate @item Start @code{mysqld} from @code{gdb} (or in another debugger). @xref{Using gdb on mysqld}. @item Run your test scripts. @item Print the backtrace and the local variables at the 3 lowest levels. In gdb you can do this with the following commands when @code{mysqld} has crashed inside gdb: @example backtrace info local up info local up info local @end example With gdb you can also examine which threads exist with @code{info threads} and switch to a specific thread with @code{thread #}, where @code{#} is the thread id. @end enumerate @item Try to simulate your application with a Perl script to force MySQL to crash or misbehave. @item Send a normal bug report. @xref{Bug reports}. Be even more detailed than usual. Because MySQL works for many people, it may be that the crash results from something that exists only on your computer (for example, an error that is related to your particular system libraries). @item If you have a problem with tables with dynamic-length rows and you are not using @code{BLOB/TEXT} columns (but only @code{VARCHAR} columns), you can try to change all @code{VARCHAR} to @code{CHAR} with @code{ALTER TABLE}. This will force MySQL to use fixed-size rows. Fixed-size rows take a little extra space, but are much more tolerant to corruption! The current dynamic row code has been in use at MySQL AB for at least 3 years without any problems, but by nature dynamic-length rows are more prone to errors, so it may be a good idea to try the above to see if it helps! @end itemize @node Resetting permissions, Full disk, Crashing, Administration Issues @appendixsubsec How to Reset a Forgotten Password @cindex passwords, forgotten @cindex passwords, resetting @cindex root user, password resetting If you have forgotten the @code{root} user password for MySQL, you can restore it with the following procedure: @enumerate @item Take down the @code{mysqld} server by sending a @code{kill} (not @code{kill -9}) to the @code{mysqld} server. The pid is stored in a @code{.pid} file, which is normally in the MySQL database directory: @example kill `cat /mysql-data-directory/hostname.pid` @end example You must be either the Unix @code{root} user or the same user the server runs as to do this. @item Restart @code{mysqld} with the @code{--skip-grant-tables} option. @item Connect to the @code{mysqld} server with @code{mysql -h hostname mysql} and change the password with a @code{GRANT} command. @xref{GRANT,,@code{GRANT}}. You can also do this with @code{mysqladmin -h hostname -u user password 'new password'} @item Load the privilege tables with: @code{mysqladmin -h hostname flush-privileges} or with the SQL command @code{FLUSH PRIVILEGES}. @end enumerate Note that after you started @code{mysqld} with @code{--skip-grant-tables}, any usage of @code{GRANT} commands will give you an @code{Unknown command} error until you have executed @code{FLUSH PRIVILEGES}. @node Full disk, Temporary files, Resetting permissions, Administration Issues @appendixsubsec How MySQL Handles a Full Disk @cindex full disk @cindex disk full @noindent When a disk-full condition occurs, MySQL does the following: @itemize @bullet @item It checks once every minute to see whether or not there is enough space to write the current row. If there is enough space, it continues as if nothing had happened. @item Every 6 minutes it writes an entry to the log file warning about the disk full condition. @end itemize @noindent To alleviate the problem, you can take the following actions: @itemize @bullet @item To continue, you only have to free enough disk space to insert all records. @item To abort the thread, you must send a @code{mysqladmin kill} to the thread. The thread will be aborted the next time it checks the disk (in 1 minute). @item Note that other threads may be waiting for the table that caused the disk full condition. If you have several ``locked'' threads, killing the one thread that is waiting on the disk-full condition will allow the other threads to continue. @end itemize Exceptions to the above behaveour is when you use @code{REPAIR} or @code{OPTIMIZE} or when the indexes are created in a batch after an @code{LOAD DATA INFILE} or after an @code{ALTER TABLE} statement. All of the above commands may use big temporary files that left to themself would cause big problems for the rest of the system. If MySQL gets disk full while doing any of the above operations, it will remove the big temporary files and mark the table as crashed (except for @code{ALTER TABLE}, in which the old table will be left unchanged). @node Temporary files, Problems with mysql.sock, Full disk, Administration Issues @appendixsubsec Where MySQL Stores Temporary Files MySQL uses the value of the @code{TMPDIR} environment variable as the pathname of the directory in which to store temporary files. If you don't have @code{TMPDIR} set, MySQL uses the system default, which is normally @file{/tmp} or @file{/usr/tmp}. If the file system containing your temporary file directory is too small, you should edit @code{safe_mysqld} to set @code{TMPDIR} to point to a directory in a file system where you have enough space! You can also set the temporary directory using the @code{--tmpdir} option to @code{mysqld}. MySQL creates all temporary files as hidden files. This ensures that the temporary files will be removed if @code{mysqld} is terminated. The disadvantage of using hidden files is that you will not see a big temporary file that fills up the file system in which the temporary file directory is located. When sorting (@code{ORDER BY} or @code{GROUP BY}), MySQL normally uses one or two temporary files. The maximum disk-space needed is: @example (length of what is sorted + sizeof(database pointer)) * number of matched rows * 2 @end example @code{sizeof(database pointer)} is usually 4, but may grow in the future for really big tables. For some @code{SELECT} queries, MySQL also creates temporary SQL tables. These are not hidden and have names of the form @file{SQL_*}. @code{ALTER TABLE} creates a temporary table in the same directory as the original table. @node Problems with mysql.sock, Timezone problems, Temporary files, Administration Issues @appendixsubsec How to Protect or change the MySQL socket file @file{/tmp/mysql.sock} @cindex @code{mysql.sock}, protection @cindex deletion, @code{mysql.sock} If you have problems with the fact that anyone can delete the MySQL communication socket @file{/tmp/mysql.sock}, you can, on most versions of Unix, protect your @file{/tmp} file system by setting the @code{sticky} bit on it. Log in as @code{root} and do the following: @example shell> chmod +t /tmp @end example This will protect your @file{/tmp} file system so that files can be deleted only by their owners or the superuser (@code{root}). You can check if the @code{sticky} bit is set by executing @code{ls -ld /tmp}. If the last permission bit is @code{t}, the bit is set. @cindex changing socket location You can change the place where MySQL uses / puts the socket file the following ways: @itemize @bullet @item Specify the path in a global or local option file. For example, put in @code{/etc/my.cnf}: @example [client] socket=path-for-socket-file [mysqld] socket=path-for-socket-file @end example @xref{Option files}. @item Specifying this on the command line to @code{safe_mysqld} and most clients with the @code{--socket=path-for-socket-file} option. @item Specify the path to the socket in the @code{MYSQL_UNIX_PORT} environment variable. @item Defining the path with the @code{configure} option @code{--with-unix-socket-path=path-for-socket-file}. @xref{configure options}. @end itemize You can test that the socket works with this command: @example shell> mysqladmin --socket=/path/to/socket version @end example @node Timezone problems, , Problems with mysql.sock, Administration Issues @appendixsubsec Time Zone Problems @cindex timezone problems @cindex problems, timezone @tindex TZ environment variable @tindex Environment variable, TZ If you have a problem with @code{SELECT NOW()} returning values in GMT and not your local time, you have to set the @code{TZ} environment variable to your current time zone. This should be done for the environment in which the server runs, for example, in @code{safe_mysqld} or @code{mysql.server}. @xref{Environment variables}. @node Query Issues, Table Definition Issues, Administration Issues, Problems @appendixsec Query Related Issues @menu * Case sensitivity:: Case Sensitivity in Searches * Using DATE:: Problems Using @code{DATE} Columns * Problems with NULL:: Problems with @code{NULL} Values * Problems with alias:: Problems with @code{alias} * Deleting from related tables:: Deleting Rows from Related Tables * No matching rows:: Solving Problems with No Matching Rows @end menu @node Case sensitivity, Using DATE, Query Issues, Query Issues @appendixsubsec Case Sensitivity in Searches @cindex case sensitivity, in searches @cindex searching, and case-sensitivity @cindex Chinese @cindex Big5 Chinese character encoding By default, MySQL searches are case-insensitive (although there are some character sets that are never case insensitive, such as @code{czech}). That means that if you search with @code{col_name LIKE 'a%'}, you will get all column values that start with @code{A} or @code{a}. If you want to make this search case-sensitive, use something like @code{INSTR(col_name, "A")=1} to check a prefix. Or use @code{STRCMP(col_name, "A") = 0} if the column value must be exactly @code{"A"}. Simple comparison operations (@code{>=, >, = , < , <=}, sorting and grouping) are based on each character's ``sort value''. Characters with the same sort value (like E, e and �) are treated as the same character! In older MySQL versions @code{LIKE} comparisons where done on the uppercase value of each character (E == e but E <> �). In newer MySQL versions @code{LIKE} works just like the other comparison operators. If you want a column always to be treated in case-sensitive fashion, declare it as @code{BINARY}. @xref{CREATE TABLE, , @code{CREATE TABLE}}. If you are using Chinese data in the so-called big5 encoding, you want to make all character columns @code{BINARY}. This works because the sorting order of big5 encoding characters is based on the order of ASCII codes. @node Using DATE, Problems with NULL, Case sensitivity, Query Issues @appendixsubsec Problems Using @code{DATE} Columns @findex DATE @cindex DATE columns, problems @cindex problems, @code{DATE} columns The format of a @code{DATE} value is @code{'YYYY-MM-DD'}. According to ANSI SQL, no other format is allowed. You should use this format in @code{UPDATE} expressions and in the WHERE clause of @code{SELECT} statements. For example: @example mysql> SELECT * FROM tbl_name WHERE date >= '1997-05-05'; @end example As a convenience, MySQL automatically converts a date to a number if the date is used in a numeric context (and vice versa). It is also smart enough to allow a ``relaxed'' string form when updating and in a @code{WHERE} clause that compares a date to a @code{TIMESTAMP}, @code{DATE}, or a @code{DATETIME} column. (Relaxed form means that any punctuation character may be used as the separator between parts. For example, @code{'1998-08-15'} and @code{'1998#08#15'} are equivalent.) MySQL can also convert a string containing no separators (such as @code{'19980815'}), provided it makes sense as a date. The special date @code{'0000-00-00'} can be stored and retrieved as @code{'0000-00-00'.} When using a @code{'0000-00-00'} date through @strong{MyODBC}, it will automatically be converted to @code{NULL} in @strong{MyODBC} Version 2.50.12 and above, because ODBC can't handle this kind of date. Because MySQL performs the conversions described above, the following statements work: @example mysql> INSERT INTO tbl_name (idate) VALUES (19970505); mysql> INSERT INTO tbl_name (idate) VALUES ('19970505'); mysql> INSERT INTO tbl_name (idate) VALUES ('97-05-05'); mysql> INSERT INTO tbl_name (idate) VALUES ('1997.05.05'); mysql> INSERT INTO tbl_name (idate) VALUES ('1997 05 05'); mysql> INSERT INTO tbl_name (idate) VALUES ('0000-00-00'); mysql> SELECT idate FROM tbl_name WHERE idate >= '1997-05-05'; mysql> SELECT idate FROM tbl_name WHERE idate >= 19970505; mysql> SELECT mod(idate,100) FROM tbl_name WHERE idate >= 19970505; mysql> SELECT idate FROM tbl_name WHERE idate >= '19970505'; @end example @noindent However, the following will not work: @example mysql> SELECT idate FROM tbl_name WHERE STRCMP(idate,'19970505')=0; @end example @code{STRCMP()} is a string function, so it converts @code{idate} to a string and performs a string comparison. It does not convert @code{'19970505'} to a date and perform a date comparison. Note that MySQL does no checking whether or not the date is correct. If you store an incorrect date, such as @code{'1998-2-31'}, the wrong date will be stored. If the date cannot be converted to any reasonable value, a @code{0} is stored in the @code{DATE} field. This is mainly a speed issue and we think it is up to the application to check the dates, and not the server. @node Problems with NULL, Problems with alias, Using DATE, Query Issues @appendixsubsec Problems with @code{NULL} Values @cindex @code{NULL} values, vs. empty values @tindex NULL The concept of the @code{NULL} value is a common source of confusion for newcomers to SQL, who often think that @code{NULL} is the same thing as an empty string @code{''}. This is not the case! For example, the following statements are completely different: @example mysql> INSERT INTO my_table (phone) VALUES (NULL); mysql> INSERT INTO my_table (phone) VALUES (""); @end example Both statements insert a value into the @code{phone} column, but the first inserts a @code{NULL} value and the second inserts an empty string. The meaning of the first can be regarded as ``phone number is not known'' and the meaning of the second can be regarded as ``she has no phone''. In SQL, the @code{NULL} value is always false in comparison to any other value, even @code{NULL}. An expression that contains @code{NULL} always produces a @code{NULL} value unless otherwise indicated in the documentation for the operators and functions involved in the expression. All columns in the following example return @code{NULL}: @example mysql> SELECT NULL,1+NULL,CONCAT('Invisible',NULL); @end example If you want to search for column values that are @code{NULL}, you cannot use the @code{=NULL} test. The following statement returns no rows, because @code{expr = NULL} is FALSE, for any expression: @example mysql> SELECT * FROM my_table WHERE phone = NULL; @end example To look for @code{NULL} values, you must use the @code{IS NULL} test. The following shows how to find the @code{NULL} phone number and the empty phone number: @example mysql> SELECT * FROM my_table WHERE phone IS NULL; mysql> SELECT * FROM my_table WHERE phone = ""; @end example In MySQL, as in many other SQL servers, you can't index columns that can have @code{NULL} values. You must declare such columns @code{NOT NULL}. Conversely, you cannot insert @code{NULL} into an indexed column. @findex LOAD DATA INFILE When reading data with @code{LOAD DATA INFILE}, empty columns are updated with @code{''}. If you want a @code{NULL} value in a column, you should use @code{\N} in the text file. The literal word @code{'NULL'} may also be used under some circumstances. @xref{LOAD DATA, , @code{LOAD DATA}}. When using @code{ORDER BY}, @code{NULL} values are presented first. If you sort in descending order using @code{DESC}, @code{NULL} values are presented last. When using @code{GROUP BY}, all @code{NULL} values are regarded as equal. To help with @code{NULL} handling, you can use the @code{IS NULL} and @code{IS NOT NULL} operators and the @code{IFNULL()} function. @cindex @code{TIMESTAMP}, and @code{NULL} values @cindex @code{AUTO_INCREMENT}, and @code{NULL} values @cindex @code{NULL} values, and @code{TIMESTAMP} columns @cindex @code{NULL} values, and @code{AUTO_INCREMENT} columns For some column types, @code{NULL} values are handled specially. If you insert @code{NULL} into the first @code{TIMESTAMP} column of a table, the current date and time is inserted. If you insert @code{NULL} into an @code{AUTO_INCREMENT} column, the next number in the sequence is inserted. @node Problems with alias, Deleting from related tables, Problems with NULL, Query Issues @appendixsubsec Problems with @code{alias} @tindex alias You can use an alias to refer to a column in the @code{GROUP BY}, @code{ORDER BY}, or in the @code{HAVING} part. Aliases can also be used to give columns better names: @example SELECT SQRT(a*b) as rt FROM table_name GROUP BY rt HAVING rt > 0; SELECT id,COUNT(*) AS cnt FROM table_name GROUP BY id HAVING cnt > 0; SELECT id AS "Customer identity" FROM table_name; @end example Note that ANSI SQL doesn't allow you to refer to an alias in a @code{WHERE} clause. This is because when the @code{WHERE} code is executed the column value may not yet be determined. For example, the following query is @strong{illegal}: @example SELECT id,COUNT(*) AS cnt FROM table_name WHERE cnt > 0 GROUP BY id; @end example The @code{WHERE} statement is executed to determine which rows should be included in the @code{GROUP BY} part while @code{HAVING} is used to decide which rows from the result set should be used. @node Deleting from related tables, No matching rows, Problems with alias, Query Issues @appendixsubsec Deleting Rows from Related Tables @cindex deleting, rows @cindex rows, deleting @cindex tables, deleting rows As MySQL doesn't support sub-selects or use of more than one table in the @code{DELETE} statement, you should use the following approach to delete rows from 2 related tables: @enumerate @item @code{SELECT} the rows based on some @code{WHERE} condition in the main table. @item @code{DELETE} the rows in the main table based on the same condition. @item @code{DELETE FROM related_table WHERE related_column IN (selected_rows)}. @end enumerate If the total number of characters in the query with @code{related_column} is more than 1,048,576 (the default value of @code{max_allowed_packet}, you should split it into smaller parts and execute multiple @code{DELETE} statements. You will probably get the fastest @code{DELETE} by only deleting 100-1000 @code{related_column} id's per query if the @code{related_column} is an index. If the @code{related_column} isn't an index, the speed is independent of the number of arguments in the @code{IN} clause. @node No matching rows, , Deleting from related tables, Query Issues @appendixsubsec Solving Problems with No Matching Rows @cindex no matching rows @cindex rows, matching problems If you have a complicated query that has many tables and that doesn't return any rows, you should use the following procedure to find out what is wrong with your query: @enumerate @item Test the query with @code{EXPLAIN} and check if you can find something that is obviously wrong. @xref{EXPLAIN, , @code{EXPLAIN}}. @item Select only those fields that are used in the @code{WHERE} clause. @item Remove one table at a time from the query until it returns some rows. If the tables are big, it's a good idea to use @code{LIMIT 10} with the query. @item Do a @code{SELECT} for the column that should have matched a row against the table that was last removed from the query. @item If you are comparing @code{FLOAT} or @code{DOUBLE} columns with numbers that have decimals, you can't use @code{=}! This problem is common in most computer languages because floating-point values are not exact values: @example mysql> SELECT * FROM table_name WHERE float_column=3.5; -> mysql> SELECT * FROM table_name WHERE float_column between 3.45 and 3.55; @end example In most cases, changing the @code{FLOAT} to a @code{DOUBLE} will fix this! @item If you still can't figure out what's wrong, create a minimal test that can be run with @code{mysql test < query.sql} that shows your problems. You can create a test file with @code{mysqldump --quick database tables > query.sql}. Open the file in an editor, remove some insert lines (if there are too many of these), and add your select statement at the end of the file. Test that you still have your problem by doing: @example shell> mysqladmin create test2 shell> mysql test2 < query.sql @end example Post the test file using @code{mysqlbug} to @email{mysql@@lists.mysql.com}. @end enumerate @node Table Definition Issues, , Query Issues, Problems @appendixsec Table Definition Related Issues @menu * ALTER TABLE problems:: Problems with @code{ALTER TABLE}. * Change column order:: How To Change the Order of Columns in a Table * Temporary table problems:: TEMPORARY TABLE problems @end menu @node ALTER TABLE problems, Change column order, Table Definition Issues, Table Definition Issues @appendixsubsec Problems with @code{ALTER TABLE}. @tindex ALTER TABLE @code{ALTER TABLE} changes a table to the current character set. If you during @code{ALTER TABLE} get a duplicate key error, then the cause is either that the new character sets maps to keys to the same value or that the table is corrupted, in which case you should run @code{REPAIR TABLE} on the table. If @code{ALTER TABLE} dies with an error like this: @example Error on rename of './database/name.frm' to './database/B-a.frm' (Errcode: 17) @end example the problem may be that MySQL has crashed in a previous @code{ALTER TABLE} and there is an old table named @file{A-something} or @file{B-something} lying around. In this case, go to the MySQL data directory and delete all files that have names starting with @code{A-} or @code{B-}. (You may want to move them elsewhere instead of deleting them.) @code{ALTER TABLE} works the following way: @itemize @bullet @item Create a new table named @file{A-xxx} with the requested changes. @item All rows from the old table are copied to @file{A-xxx}. @item The old table is renamed @file{B-xxx}. @item @file{A-xxx} is renamed to your old table name. @item @file{B-xxx} is deleted. @end itemize If something goes wrong with the renaming operation, MySQL tries to undo the changes. If something goes seriously wrong (this shouldn't happen, of course), MySQL may leave the old table as @file{B-xxx}, but a simple rename on the system level should get your data back. @node Change column order, Temporary table problems, ALTER TABLE problems, Table Definition Issues @appendixsubsec How To Change the Order of Columns in a Table @cindex reordering, columns @cindex columns, changing @cindex changing, column order @cindex tables, changing column order The whole point of SQL is to abstract the application from the data storage format. You should always specify the order in which you wish to retrieve your data. For example: @example SELECT col_name1, col_name2, col_name3 FROM tbl_name; @end example will return columns in the order @code{col_name1}, @code{col_name2}, @code{col_name3}, whereas: @example SELECT col_name1, col_name3, col_name2 FROM tbl_name; @end example will return columns in the order @code{col_name1}, @code{col_name3}, @code{col_name2}. You should @strong{NEVER}, in an application, use @code{SELECT *} and retrieve the columns based on their position, because the order in which columns are returned @strong{CANNOT} be guaranteed over time. A simple change to your database may cause your application to fail rather dramatically. If you want to change the order of columns anyway, you can do it as follows: @enumerate @item Create a new table with the columns in the right order. @item Execute @code{INSERT INTO new_table SELECT fields-in-new_table-order FROM old_table}. @item Drop or rename @code{old_table}. @item @code{ALTER TABLE new_table RENAME old_table}. @end enumerate @node Temporary table problems, , Change column order, Table Definition Issues @appendixsubsec TEMPORARY TABLE problems @cindex temporary tables, problems The following are a list of the limitations with @code{TEMPORARY TABLES}. @itemize @bullet @item A temporary table can only be of type @code{HEAP}, @code{ISAM} or @code{MyISAM}. @item You can't use temporary tables more than once in the same query. For example, the following doesn't work. @example select * from temporary_table, temporary_table as t2; @end example We plan to fix the above in 4.0. @item You can't use @code{RENAME} on a @code{TEMPORARY} table. Note that @code{ALTER TABLE org_name RENAME new_name} works! We plan to fix the above in 4.0. @end itemize @node Users, MySQL customer usage, Problems, Top @appendix Some MySQL Users @cindex users, of MySQL @cindex news sites This appendix lists some users of MySQL that have given us permission to list them in our documentation. It is by far not a complete list, but should give you a general idea of who uses MySQL and what it can be used for. @appendixsec General News Sites @itemize @bullet @item @uref{http://www.yahoo.com/, Yahoo!} @item @uref{http://slashdot.org/, Slashdot: A pro-Linux/tech news and comment/discussion site} @item @uref{http://www.linux.com/, All about Linux} @item @uref{http://www.linuxtoday.com/, Linuxtoday} @item @uref{http://www.32bitsonline.com/, 32Bits Online: because there's more than one way to compute} @item @uref{http://www.freshmeat.net/, Freshmeat: News about new versions of computer-related stuff} @end itemize @cindex search engines, web @cindex web search engines @appendixsec Some Web Search Engines @itemize @bullet @item @uref{http://www.aaa.com.au, AAA Matilda Web Search} @item @uref{http://www.whatsnu.com/, What's New} @item @uref{http://www.aladin.de/, Aladin} @item @uref{http://www.columbus-finder.de/, Columbus Finder} @item @uref{http://www.spider.de/, Spider} @item @uref{http://www.blitzsuche.de/, Blitzsuche} @item @uref{http://www.indoseek.co.id, Indoseek Indonesia} @item @uref{http://www.yaboo.dk/, Yaboo - Yet Another BOOkmarker} @item @uref{http://www.ozsearch.com.au, OzSearch Internet Guide} @item @uref{http://www.splatsearch.com/, Splat! Search} @item @uref{http://osdls.library.arizona.edu/, The Open Source Digital Library System Project} @end itemize @appendixsec Some Information Search Engines Concentrated on Some Area @itemize @bullet @item @uref{http://www.spylog.ru/, SpyLOG ; A very popular Web counter site} @item @uref{http://www.tucows.com/, TuCows Network; Free Software archive} @item @uref{http://www.jobvertise.com,Jobvertise: Post and search for jobs} @item @uref{http://www.musicdatabase.com, The Music Database} @item @uref{http://www.soccersearch.com, Football (Soccer) search page} @item @uref{http://www.headrush.net/takedown, TAKEDOWN - wrestling} @item @uref{http://www.lyrics.net, The International Lyrics Network} @item @uref{http://TheMatrix.com/~matrix/band_search.phtml, Musicians looking for other musicians (Free Service)} @item @uref{http://www.addall.com/AddBooks/Stores.html,AddALL books searching and price comparison} @item @uref{http://www.herbaria.harvard.edu/Data/Gray/gray.html,Harvard's Gray Herbarium Index of Plant Names} @item @uref{http://www.game-developer.com/,The Game Development Search Engine} @item @uref{www.theinnkeeper.com, The Innkeeper Vacation Guides} @item @uref{http://www.macgamedatabase.com/, The Mac Game Database uses PHP and MySQL} @c From: Marc Antony Vose <suzerain@suzerain.com> @item @uref{http://www.csse.monash.edu.au/publications/, Research Publications at Monash University in Australia} @item @uref{http://www.ipielle.emr.it/bts/index.html, Occupational Health & Safety Web site database (a project for the ECC)} @c c.presutti@ipielle.emr.it @item @uref{http://data.mch.mcgill.ca/, Bioinformatics databases at the Montreal Children's Hospital using MySQL} @c saeed@www.debelle.mcgill.ca @end itemize @cindex online magazines @cindex magazines, online @appendixsec Online Magazines @itemize @bullet @item @uref{http://www.spoiler.com, Spoiler Webzine}. An online magazine featuring music, literature, arts, and design content. @item @uref{http://www.linux-magazin.de/newsflash/, Daily news about Linux in German language} @item @uref{http://www.betazine.com,Betazine - The Ultimate Online Beta Tester's Magazine} @item @uref{http://www.currents.net/ccinfo/aboutcc.html,Computer Currents Magazine} @end itemize @cindex web sites @appendixsec Web Sites that Use MySQL as a Backend @itemize @bullet @item @uref{http://liftoff.msfc.nasa.gov, NASA} @item @uref{http://kids.msfc.nasa.gov, NASA KIDS} @item @uref{http://science.nasa.gov, Sience@@NASA} @item @uref{http://www.handy.de/, handy.de} @item @uref{http://lindev.jmc.tju.edu/qwor, Qt Widget and Object Repository} @item @uref{http://www.samba-choro.com.br, Brazilian samba site (in Portuguese)} @item @uref{http://pgss.iss.uw.edu.pl/en_index.ISS, Polish General Social Survey} @item @uref{http://www.expo2000.com, Expo2000} World-wide distribution of tickets for this event is implemented using MySQL and tcl/tk. More than 5000 travel agencies all over the world have access to it. @item @uref{http://www.freevote.com/, FreeVote.com is a free voting service with millions of users.} @item @uref{http://f1.tauzero.se, Forza Motorsport} @item @uref{http://www.dreamhost.com/, DreamHost Web Hosting} @end itemize @cindex services @appendixsec Some Domain/Internet/Web and Related Services @itemize @bullet @item @uref{http://www.wix.com/mysql-hosting, Registry of Web providers that support MySQL} @item @uref{http://www.yi.org/, Dynamic DNS Services} @item @uref{http://www.dynodns.net/, Dynamic domain name service} @item @uref{http://www.ods.org/, Open DNS Project; free dynamic DNS service} @c @item @uref{http://dynodns.net, Free dynamic DNS implementation} @c EMAIL: A Moore <amoore@mooresystems.com> @item @uref{http://www.hn.org/, Hammernode; Public DNS Servers} @item @uref{http://www.fdns.net/, Free 3rd level domains} @item @uref{http://worldcommunity.com/, Online Database} @item @uref{http://www.bigbiz.com, BigBiz Internet Services} @item @uref{http://virt.circle.net, The Virt Gazette} @item @uref{http://www.california.com, Global InfoNet Inc} @item @uref{http://www.webhosters.com, WebHosters - A Guide to WWW Providers} @item @uref{http://online.dn.ru, Internet information server} @item @uref{http://www.stopbit.com, A technology news site} @item @uref{http://www.worldnetla.net, WorldNet Communications - An Internet Services Provider} @item @uref{http://www.netizen.com.au/, Netizen: Australian-based Web consultancy} @item @uref{http://www.trainingpages.co.uk, Search site for training courses in the UK} @item @uref{http://chat.nitco.com, Gannon Chat (GPL). Written in Perl and Javascript} @item @uref{http://www.addurls.com/,A general links directory} @item @uref{http://www.bookmarktracker.com, A Web-based bookmark management service} @item @uref{http://www.cdrom.com,Walnut Creek CDROM} @item @uref{http://www.wwwthreads.org/, WWWThreads; Interactive discussion Forums} @item @uref{http://pvmon.portici.enea.it/Meteo, In Italian; Storage data from meteo station} @item @uref{http://www.buysell.net/, Online "Person To Person" Auction} @item @uref{http://tips.pair.com,Tips on Web development} @item @uref{http://www.mailfriends.com, Mailfriends.com is a FREE service for everybody who wants to find friends over the internet.} @item @uref{http://www.uninova.com/cgi-bin/wctelnets?list, Web Page Telnet BBS List} @item @uref{http://www.uninova.com/cnc.html,UniNova Digital Postcards} @c @item @uref{http://cabinboy.powersurfr.com, An Internet RFC search engine} @item @uref{http://www.dslreports.com, DSL-provider search with reviews}. Made with MySQL and Modperl, all pages are generated dynamically out of the MySQL database @end itemize @cindex PHP, web sites @appendixsec Web Sites that Use @code{PHP} and MySQL @itemize @bullet @c @item @uref{http://www.wh200th.com, White House 200th Anniversary site} @item @uref{http://support.jgaa.com/, Jgaa's Internet - Official Support Site} @item @uref{http://io.incluso.com, Ionline - online publication:} MySQL, PHP, Java, Web programming, DB development @item @uref{http://www.baboo.com, BaBoo(Browse and bookmark). Free Web-based bookmark manager and Calendar} @item @uref{http://www.courses.pjc.cc.fl.us/Schedule/index.php, Course Schedule System at Pensacola Junior College} @item @uref{http://www.fccj.org, Florida Community College at Jacksonville} @item @uref{http://www.32bit.com/, 32bit.com; An extensive shareware / freeware archive} @item @uref{http://www.jokes2000.com/, Jokes 2000} @c Added 990604; EMAIL: ah@dybdahl.dk @item @uref{http://www.burken.nu/ , Burken.NU} Burken is a webhotel that provides scripts, among other things, for remote users, like counters, guestbooks etc. @c Added 990608; EMAIL: spacedmp@SpaceDump.Burken.NU (Anders Olausson) @item @uref{http://tips.pair.com, tips.pair.com} Contains tips on html, javascript, 2d/3d graphics, and PHP3/MySQL. All pages are generated from a database. @c Added 990614; EMAIL: downey@image.dk (Rune Madsen) @item @uref{http://www.softwarezrus.com/, Softwarezrus.com} Ecommerce site that is selling computers. @end itemize @cindex consultants, list of @appendixsec Some MySQL Consultants @itemize @bullet @item @uref{http://www.ayni.com, Ayni AG} @item @uref{http://worldcommunity.com/, Online Database} @item @uref{http://www2.dataguard.no/,DataGuard (Uses MySQL and PHP)} @item @uref{http://wwits.net/programs/mysql.phtml, WWITS (Uses MySQL and PHP)} @item @uref{http://www.worldcommunity.com/, WCN - The World Community Network} @item @uref{http://www.chipcastle.com, Chip Castle Dot Com Inc} @c Added 990603 EMAIL: chip@chipcastle.com (Chip Castle) @item @uref{http://www.cyber.com.au/, Cybersource Pty. Ltd} @item @uref{http://www.spring.de, Spring infotainment gmbh & co. kg} @c added 990905 "Oliver Pischke" <opischke@spring.de> @item @uref{http://www.wamdesign.com/, Develops Web sites using MySQL} @c Added 990905; max@wamdesign.com @item @uref{http://www.berkeleyconsultants.com, Berkeley Consultants Group} @item @uref{http://www.jammconsulting.com/, JAMM Consulting Inc.} @end itemize @appendixsec Programming @cindex web pages, miscellaneous @appendixsec Uncategorized Pages @itemize @bullet @item @uref{http://www.feature-showcase.com/htmls/demo_mysql.sql, AZC.COM's Feature Showcase} @item @uref{http://www.teach.org.uk/subjects/trainingcourse/g.html, Course Search} @item @uref{http://www.northerbys.com, Northerbys Online Auctions} @item @uref{http://www.schiphol.nl/flights/home.htm, Amsterdam Airport Schiphol} @item @uref{http://TheMatrix.com/seventhsin/query.phtml, CD database} @item @uref{http://TheMatrix.com/~flmm/GEAR.html, Used Audio Gear Database} @item @uref{http://www.kiss.de/musik-mueller, Musical note-sheets} @item @uref{http://www.bagism.com, Bagism - A John Lennon fan page} @item @uref{http://www.selftaught.com/, US Folk art broker} @item @uref{http://organizer.net/, Mail reading on the Web} @item @uref{http://www.mypage.org/, Free home pages on www.somecoolname.mypage.org} @item @uref{http://www.schulweb.de/, Der Server f@"ur Schulen im Web (In German)} @item @uref{http://www.ald.net/, Auldhaefen Online Services} @item @uref{http://www.cary.net/, CaryNET Information Center} @item @uref{http://www.dataden.com/, Dataden Computer Systems} @item @uref{http://andree.grm.se/, Andr@'emuseet (In Swedish)} @item @uref{http://www.him.net/, HOMESITE Internet Marketing} @item @uref{http://www.jade-v.com/techinfo.html, Jade-V Network Services } @item @uref{http://ww2010.atmos.uiuc.edu/(Gl)/abt/aknw/tech.rxml, Weather World 2010 Technical Credits} @* @item @uref{http://gimp.foebud.org/registry/doc/, About The Gimp plugin registry} @item @uref{http://www.fast-inc.com/Products/Archiver/database.html, Java tool Archiver technical detail (Slightly optimistic about MySQL ANSI-92 compliance)} @item @uref{http://www.gamesdomain.com/cheats/usrcheat.phtml, Games Domain Cheats Database} @item @uref{http://www.kcilink.com/poweredby/, The "Powered By" Page (Kcilink)} @item @uref{http://www.netcasting.net/index.whtml, Netcasting} @item @uref{http://homepages.tig.com.au/~mjj/nbltips, NBL (Australian National Basketball League) tipping} @item @uref{http://www.cgishop.com/, CGI shop} @item @uref{http://www.whirlycott.com/, Whirlycott: Website Design} @item @uref{http://www.mtp.dk, Museum Tusculanum Press} @item @uref{http://csdgi.historie.ku.dk/biblio, Centro Siciliano di Documentazione} @item @uref{http://caribou.dyn.ml.org:8000, Quake statistics database} @item @uref{http://www.astroforum.ch, Astroforum: Astrologie and related things (in German)} @item @uref{http://www.opendebate.com, OpenDebate - Interactive Polls & Open Discussion} @item @uref{http://vermeer.organik.uni-erlangen.de/dissertationen/, Online chemical dissertation server} @item @uref{http://www.freschinfo.com, FreSch! The Free Scholarship Search Service} @item @uref{http://www.nada.kth.se/~staffanu/pinball, Stockholm Pinball Locator} @item @uref{http://www.hek.com, HEK A construction company} @item @uref{http://www.nbi.nl, Elsevier Bussines Information} @item @uref{http://vaccination.medicallink.se/, Medical Links (Using ColdFusion and MySQL)} @item @uref{http://www.joblink-usa.com, Search for jobs & people at JobLink-USA} @item @uref{http://www.skydive.net/competfs, Competition Formation Skydiving} @item @uref{http://www.galaxy-net.net/Galaxy-NET Telecommunications, E-commerce and internal accounting} @item @uref{http://www.borsen.dk/, Denmark's leading business daily newspaper B@o{}rsen} @item @uref{http://tmmm.simplenet.com/indb/, The Internet NES Database} @item @uref{http://www.russia.cz, Travel agency in Prague in 3 languages} @item @uref{http://www.linkstation.de, Linkstation} @item @uref{http://www.peoplestaff.com, Searchable online database at Peoplestaff} @item @uref{http://www.dreamhorse.com, A searchable database system for horse classified ads} @item @uref{http://pootpoot.com/,The Poot site} @item @uref{http://grateful.net/hw_html/,"Playin' in the LAN"; a network monitoring suite} @c Update from Christopher Milton <cmilton@bwn.net> 1999-12-21 @item @uref{http://www.usapa.army.mil,U.S. Army Publishing Agency} @item @uref{http://www.nekretnine.co.yu/,Realestate handling in Yugoslavia} @item @uref{http://demo.cpsoft.com/pims/devFAQ.html, PIMS; a Patient Information Management System} @item @uref{http://cpsoft.com,Pilkington Software Inc} @item @uref{http://www.no-quarter.org/,A Vietnam Veteran's Memorial (The Wall) database} @item @uref{http://www.gamers-union.com/,Gamer's Union specializes in auctions of used & out-of-print gaming material} @item @uref{http://www.montereyhigh.com/office/dbul.php3, A daily bulletin at Monterey High school} @item @uref{http://www.myEastside.com,Community-owned site serving Lake Washington's Eastside residents and businesses} @item @uref{http://bowling-france.net/,French bowling site} @end itemize Send any additions to this list to @email{webmaster@@mysql.com}. @page @node MySQL customer usage, Contrib, Users, Top @appendix MySQL customer usage @cindex MySQL usage The section 'Some MySQL Users' contains a lot of different links to MySQL users but doesn't provide that much information about how they are using MySQL. @xref{Users}. This manual section is to give you an idea of how other MySQL users are using MySQL to solve their problems. This manual section is very new and we plan to add more stories here shortly. If you are interested in contributing of how you use MySQL in a unique environment or have success store about how you use MySQL, you can write to @code{docs@@mysql.com} with subject @code{Success:}. Note that as we are very busy it may take some time before you get some feedback for your story. @itemize @bullet @item @strong{Peter Zaitsev of Spylog.ru} writes: I think you might be interested in my database size. The whole database is currently on 15 servers and I think it's about 60.000 of tables containing about 5.000.000.000 of rows. My mostly loaded server currently holds about 10.000 of tables with 1.000.000.000 of rows in it. Hugest tables have about 50.000.000 of rows, and this value will raise as soon as I'll move to 2.4 kernel with large files. Currently I have to delete much of logs for large sites to hold table sizes in 2Gb. @item @strong{Texas Instruments} is using MySQL for handling tables that contain up to 2,000 million rows in a validation regression database. @end itemize @page @node Contrib, Credits, MySQL customer usage, Top @appendix Contributed Programs @cindex contributed programs @cindex programs, contributed Many users of MySQL have contributed @emph{very} useful support tools and add-ons. @ifclear web A list of what is available at @uref{http://www.mysql.com/Downloads/Contrib/} (or any mirror) is shown below. If you want to build MySQL support for the Perl @code{DBI}/@code{DBD} interface, you should fetch the @code{Data-Dumper}, @code{DBI}, and @code{Msql-Mysql-modules} files and install them. @xref{Perl support}. @end ifclear @appendixsec APIs @cindex Perl, modules @itemize @bullet @item Perl Modules @itemize @minus @item @uref{http://www.mysql.com/Downloads/Contrib/Data-Dumper-2.101.tar.gz, Data-Dumper-2.101.tar.gz} Perl @code{Data-Dumper} module. Useful with @code{DBI}/@code{DBD} support for older Perl installations. @item @uref{http://www.mysql.com/Downloads/Contrib/DBI-1.15.tar.gz, DBI-1.15.tar.gz} Perl @code{DBI} module. @item @uref{http://www.mysql.com/Downloads/Contrib/KAMXbase1.2.tar.gz,KAMXbase1.2.tar.gz} Convert between @file{.dbf} files and MySQL tables. Perl module written by Pratap Pereira @email{pereira@@ee.eng.ohio-state.edu}, extended by Kevin A. McGrail @email{kmcgrail@@digital1.peregrinehw.com}. This converter can handle MEMO fields. @item @uref{http://www.mysql.com/Downloads/Contrib/Msql-Mysql-modules-1.2216.tar.gz, Msql-Mysql-modules-1.2216.tar.gz} Perl @code{DBD} module to access mSQL and MySQL databases. @item @uref{http://www.mysql.com/Downloads/Contrib/Data-ShowTable-3.3.tar.gz, Data-ShowTable-3.3.tar.gz} Perl @code{Data-ShowTable} module. Useful with @code{DBI}/@code{DBD} support. @item @uref{http://www.mysql.com/Downloads/Contrib/HandySQL-1.1.tar.gz, HandySQL-1.1.tar.gz} HandySQL is a MySQL access module. It offers a C interface embedded in Perl and is approximately 20% faster than regular DBI. @end itemize @cindex JDBC @item JDBC @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/mm.mysql.jdbc-1.2c.tar.gz, mm.mysql.jdbc-1.2c.tar.gz} The mm JDBC driver for MySQL. This is a production release and is actively developed. By Mark Matthews (@email{mmatthew@@ecn.purdue.edu}). @item @uref{http://www.mysql.com/Downloads/Contrib/mm.mysql.jdbc-2.0pre5.tar.gz, mm.mysql.jdbc-2.0pre5.tar.gz} The mm JDBC driver for MySQL. This is a pre-release beta version and is actively developed. By Mark Matthews (@email{mmatthew@@ecn.purdue.edu}). The two drivers above have an LGPL license. Please check @uref{http://www.worldserver.com/mm.mysql/} for the latest drivers (and other JDBC information) because these drivers may be out of date. @item @uref{http://www.caucho.com/projects/jdbc-mysql/index.xtp} The Resin commercial JDBC driver, which is released under open source. It claims to be faster than the mm driver, but we haven't received that much information about this yet. @item @uref{http://www.mysql.com/Downloads/Contrib/twz1jdbcForMysql-1.0.4-GA.tar.gz, twz1jdbcForMysql-1.0.4-GA.tar.gz} The twz driver: A type 4 JDBC driver by Terrence W. Zellers @email{zellert@@voicenet.com}. This is commercial but is free for private and educational use. (Not supported anymore.) @c no answer from server 990830 @c You can always find the latest driver at @uref{http://www.voicenet.com/~zellert/tjFM/}. @item @uref{http://www.mysql.com/Downloads/Contrib/pmdamysql.tgz,pmdamysql.tgz} A MySQL PMDA. Provides MySQL server status and configuration variables. @end itemize @cindex OLEDB @item OLEDB @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Win32/MyOLEDB.exe, MyOLEDB.exe} OLEDB handler for MySQL. By SWsoft. @item @uref{http://www.mysql.com/Downloads/Win32/MySamples.zip, MySamples.zip} Examples and documentation for MyOLEDB. By SWsoft. @item @uref{http://www.mysql.com/Downloads/Win32/Myoledb.zip, Myoledb.zip} Source for MyOLEDB. By SWsoft. @item @uref{http://www.mysql.com/Downloads/Win32/MyOLEDB.chm, MyOLEDB.chm} Help files for MyOLEDB. @item @uref{http://www.mysql.com/Downloads/Win32/libmyodbc.zip, libmyodbc.zip} Static MyODBC library used for build MyOLEDB. Based on MyODBC code. @end itemize @cindex C++ @item C++ @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-c++-0.02.tar.gz, mysql-c++-0.02.tar.gz} MySQL C++ wrapper library. By Roland Haenel, @email{rh@@ginster.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/MyDAO.tar.gz, MyDAO} MySQL C++ API. By Satish @email{spitfire@@pn3.vsnl.net.in}. Inspired by Roland Haenel's C++ API and Ed Carp's MyC library. @item @uref{http://www.mysql.com/download_mysql++.html, mysql++} MySQL C++ API (More than just a wrapper library.) Originally by @email{kevina@@clark.net}. Nowadays maintained by Sinisa at MySQL AB. @item @uref{http://nelsonjr.homepage.com/NJrAPI,NJrAPI} A C++ database independent library that supports MySQL. @end itemize @cindex Delphi @item Delphi @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/delphi-interface.gz, delphi-interface.gz} Delphi interface to @code{libmysql.dll}, by Blestan Tabakov, @email{root@@tdg.bis.bg}. @item @uref{http://www.mysql.com/Downloads/Contrib/DelphiMySQL2.zip, DelphiMySQL2.zip} Delphi interface to @code{libmysql.dll}, by @email{bsilva@@umesd.k12.or.us}. @item @uref{http://www.mysql.com/Downloads/Contrib/Udmysel.pas, Udmysql.pas} A wrapper for libmysql.dll for usage in Delphi. By Reiner Sombrowsky. @item @uref{http://www.fichtner.net/delphi/mysql.delphi.phtml, A Delphi interface to MySQL.} With source code. By Matthias Fichtner. @item @uref{http://www.productivity.org/projects/tmysql/, @strong{TmySQL} A library to use MySQL with Delphi}. @item @uref{http://www.geocities.com/CapeCanaveral/2064/mysql.html, Delphi TDataset-component}. @item @item @uref{http://www.mysql.com/Downloads/Contrib/Win32/SBMySQL50Share.exe, Delphi 5 Shareware MySQL Dataset Components} @end itemize @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-ruby-2.2.0.tar.gz, mysql-ruby-2.2.0.tar.gz} MySQL Ruby module. By TOMITA Masahiro @email{tommy@@tmtm.org} @uref{http://www.netlab.co.jp/ruby/. Ruby} is an Object-Oriented Interpreter Language. @item @uref{http://www.mysql.com/Downloads/Contrib/JdmMysqlDriver-0.1.0.tar.gz,JdmMysqlDriver-0.1.0.tar.gz} A VisualWorks 3.0 Smalltalk driver for MySQL. By @email{joshmiller@@earthlink.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/Db.py, Db.py} Python module with caching. By @email{gandalf@@rosmail.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQLmodule-1.4.tar.gz, MySQLmodule-1.4.tar.gz} Python interface for MySQL. By Joseph Skinner @email{joe@@earthlight.co.nz}. Modified by Joerg Senekowitsch @email{senekow@@ibm.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQL-python-0.3.0.tar.gz, MySQL-python-0.3.0.tar.gz} MySQLdb Python is an DB-API v2.0-compliant interface to MySQL. Transactions are supported if the server and tables support them. It is thread-safe, and contains a compatibility module for older code written for the no-longer-maintained MySQLmodule interface. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_mex_12.tar.gz, mysql_mex_1_12.tar.gz} An interface program for the Matlab program by MathWorks. The interface is done by Kimmo Uutela and John Fisher (not by Mathworks). Check @uref{http://boojum.hut.fi/~kuutela/mysqlmex.html,mysqlmex.html} for more information. @item @uref{http://www.mysql.com/Downloads/Contrib/mysqltcl-1.53.tar.gz, mysqltcl-1.53.tar.gz} Tcl interface for MySQL. Based on @file{msqltcl-1.50.tar.gz}. Updated by Tobias Ritzau, @email{tobri@@ida.liu.se}. @item @uref{http://www.mysql.com/Downloads/Contrib/MyC-0.1.tar.gz, MyC-0.1.tar.gz} A Visual Basic-like API, by Ed Carp. @item @uref{http://www.mysql.com/Downloads/Contrib/sqlscreens-1.0.1.tar.gz, sqlscreens-1.0.1.tar.gz} Tcl/Tk code to generate database screens. By Jean-Francois Dockes. @item @uref{http://www.mysql.com/Downloads/Contrib/Vdb-dflts-2.1.tar.gz, Vdb-dflts-2.1.tar.gz} This is a new version of a set of library utilities intended to provide a generic interface to SQL database engines such that your application becomes a 3-tiered application. The advantage is that you can easily switch between and move to other database engines by implementing one file for the new backend without making any changes to your applications. By @email{damian@@cablenet.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/DbFramework-1.10.tar.gz, DbFramework-1.10.tar.gz} DbFramework is a collection of classes for manipulating MySQL databases. The classes are loosely based on the CDIF Data Model Subject Area. By Paul Sharpe @email{paul@@miraclefish.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/pike-mysql-1.4.tar.gz, pike-mysql-1.4.tar.gz} MySQL module for pike. For use with the Roxen web server. @item @uref{http://www.mysql.com/Downloads/Contrib/squile.tar.gz, squile.tar.gz} Module for @code{guile} that allows @code{guile} to interact with SQL databases. By Hal Roberts. @item @uref{http://www.mysql.com/Downloads/Contrib/stk-mysql.tar.gz, stk-mysql.tar.gz} Interface for Stk. Stk is the Tk widgets with Scheme underneath instead of Tcl. By Terry Jones. @item @uref{http://www.mysql.com/Downloads/Contrib/eiffel-wrapper-1.0.tar.gz,eiffel-wrapper-1.0.tar.gz} Eiffel wrapper by Michael Ravits. @item @uref{http://www.mysql.com/Downloads/Contrib/SQLmy0.06.tgz,SQLmy0.06.tgz} FlagShip Replaceable Database Driver (RDD) for MySQL. By Alejandro Fernandez Herrero. @uref{http://www.fship.com/rdds.html, Flagship RDD home page} @item @uref{http://www.mysql.com/Downloads/Contrib/mydsn-1.0.zip,mydsn-1.0.zip} Binary and source for @code{mydsn.dll}. mydsn should be used to build and remove the DSN registry file for the MyODBC driver in Coldfusion applications. By Miguel Angel Sol�rzano. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQL-ADA95_API.zip, MySQL-ADA95_API.zip} An ADA95 interface to the MySQL API. By Francois Fabien. @item @uref{http://www.mysql.com/Downloads/Contrib/MyTool-DLL_for_VB_and_MySQL.zip, MyTool-DLL_for_VB_and_MySQL.zip} A DLL with MySQL C API for Visual Basic. By Ken Menzel @email{kenm@@icarz.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/MYSQLX.EXE, MYSQL.EXE} MySQL ActiveX Object for directly accessing your MySQL servers from IIS/ASP, VB, VC++ skipping the slower ODBC methods. Fully updatable, multithreaded with full support for all MySQL fieldtypes (version 2001.1.1). By SciBit @uref{http://www.scibit.com/}. @item @uref{http://www.fastflow.it/mylua/, MyLUA home page} How to use the LUA language to write MySQL @code{PROCEDURE} that can be loaded runtime. @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/lua-4.0.tar.gz, Lua 4.0} LUA 4.0 @item @uref{http://www.mysql.com/Downloads/Contrib/mylua-3.23.32.1.tar.gz, mylua-3.23.32.1.tar.gz} Patch for MySQL 3.23.32 to use LUA 4.0. By Cristian Giussani. @end itemize @item @uref{http://www.mysql.com/Downloads/Contrib/patched_myodbc.zip, patched_myodbc.zip} Patch (for Omniform 4.0 support) to the MyODBC driver. By Thomas Thaele @email{tthaele@@papenmeier.de} @end itemize @appendixsec Clients @itemize @bullet @item Graphical clients @itemize @minus @item @uref{http://www.ideit.com/products/dbvis/, DbVisualizer}. Freeware JDBC client to graphically visualize the data and structure of several databases simultaneously. By Innovative-IT Development AB. @item @uref{http://www.mysql.com/downloads/gui-clients.html, MySQLGUI} The MySQL GUI client homepage. By Sinisa at MySQL AB. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_navigator_0.9.0.tar.gz, MySQL navigator 0.9} MySQL Navigator is a MySQL database server GUI client program. The purpose of MySQL Navigator is to provide a useful client interface to MySQL database servers, whilst supporting multiple operating systems and languages. You can currently import/export database, enter queries, get result sets, edit scripts, run scripts, add, alter, and delete users, and retrieve client and server information. Uses QT 2.2. GPL @uref{http://sql.kldp.org/mysql, Home page for MySQL Navigator}. @item @uref{http://www.mysql.com/Downloads/Win32/secman.zip, MySQL Security GUI} A user and security management GUI for MySQL on Windows. By Martin Jeremic. @uref{http://jsoft.webjump.com/, Home page for MySQL Security GUI}. @item @uref{http://www.mysql.com/Downloads/Contrib/kmysqladmin-0.4.1.tar.gz, kmysqladmin-0.4.1.tar.gz}. @item @uref{http://www.mysql.com/Downloads/Contrib/kmysqladmin-0.4.1-1.src.rpm, kmysqladmin-0.4.1-1.src.rpm}. @item @uref{http://www.mysql.com/Downloads/Contrib/kmysqladmin-0.4.1-1.i386.rpm, kmysqladmin-0.4.1-1.i386.rpm} An administration tool for the MySQL server using QT / KDE. Tested only on Linux. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-admin-using-java+swing.tar.gz, Java client using Swing} By Fredy Fischer, @email{se-afs@@dial.eunet.ch}. You can always find the latest version @uref{http://www.trash.net/~ffischer/admin/index.html, here}. @item @uref{http://www.mysql.com/Downloads/Win32/MySQL-Maker-1.0.zip,MySQL-Maker 1.0}. Shareware MySQL client for Windows. It's a WYSIWYG tool which allows you to create, change and delete databases and tables. You can change field - structure and add, change and delete data in these tables directly without ODBC-driver. @uref{http://www.presult.de/presult/frames/fs_mysqlmaker.html, MySQL Maker homepage} @item @uref{http://www.mysql.com/Downloads/Contrib/mysqlwinadmn.zip, mysqlwinadmn.zip} Windows GUI (binary only) to administrate a database, by David B. Mansel, @email{david@@zhadum.org}. @item @uref{http://home.online.no/~runeberg/myqa, MyQA} is a Linux-based query client for the MySQL database server. MyQA lets you enter SQL queries, execute them, and view the results, all in a graphical user interface. The GUI is roughly similar to that of the 'Query Analyzer' client that comes with MS SQL Server. @item @uref{http://members.xoom.com/_opex_/mysqlmanager/index.html, MySQL Manager} a graphical MySQL server manager for MySQL server written in Java, for Windows @item @uref{http://www.mysql.com/Downloads/Win32/netadmin.zip, netadmin.zip} An administrator tool for MySQL on Windows 95/98 and Windows NT 4.0. Only tested with MySQL Versions 3.23.5 - 3.23.7. Written using the Tmysql components. You can write queries and show tables, indexes, table syntax, and administrate user, host, and database and so on. This is beta and still has some bugs. You can test the program with all features. Please send bugs and hints to Marco Suess @email{ms@@it-netservice.de}. Original URL @url{http://www.it-netservice.de/pages/software/index.html}. @item @uref{http://www.mysql.com/Downloads/Win32/netadmin2.zip, netadmin2.zip} New version of netadmin. See above for details. @item @uref{http://www.mysql.com/Downloads/Win32/ARTADMIN203.EXE,Atronic's MySQL client for Windows 2.0.3.0}. Home page for this can be found at: @uref{http://www.artronic.hr}. @item @uref{http://www.mysql.com/Downloads/Win32/mysqlfront.zip, mysqlfront} Home page: @uref{http://www.mysqlfront.de/}. Win32-Client for accessing and managing dbs, tables, table-data, indexes, import-/export-files. (Freeware). By Ansgar Becker. @item @uref{http://www.mysql.com/Downloads/Win32/W9xstop.zip,Utility from Artronic to stop MySQL on win9x}. @item @uref{http://bardo.hyperlink.cz/mysqlmon,a light weight GUI client for Windows}. @item @uref{http://dbtools.vila.bol.com.br/, Dbtools} A tool to manage MySQL databases. Currently only for Windows. Some features: @itemize @bullet @item Manage servers, databases, tables, columns, indexes, and users @item Import wizard to import structure and data from MS Access, MS Excel, Dbase, FoxPro, Paradox, and ODBC Databases. @item @uref{http://www.mysql.com/Downloads/Contrib/KMYENG113.zip,KMYENG113.zip} An administrator GUI for MySQL. Works only on windows, no source. Available in English and Japanese. By Mitunobu Kaneko. Home page: @uref{http://sql.jnts.ne.jp/} @end itemize @item @uref{http://www.mysql.com/Downloads/Contrib/xmysqladmin-1.0.tar.gz, xmysqladmin-1.0.tar.gz} An X-based front end to the MySQL database engine. It allows reloads, status check, process control, myisamchk, grant/revoke privileges, creating databases, dropping databases, create, alter, browse, and drop tables. Originally by Gilbert Therrien, @email{gilbert@@ican.net} but now in public domain and supported by MySQL AB. @item @uref{http://www.mysql.com/Downloads/Contrib/xmysql-1.9.tar.gz, xmysql-1.9.tar.gz}. @item @uref{http://web.wt.net/~dblhack, xmysql home page} A front end to the MySQL database engine. It allows for simple queries and table maintenance, as well as batch queries. By Rick Mehalick, @email{dblhack@@wt.net}. Requires @uref{http://bragg.phys.uwm.edu/xforms,xforms 0.88} to work. @item @uref{http://www.tamos.net/sw/dbMetrix,dbMetrix} An open source client for exploring databases and executing SQL. Supports MySQL, Oracle, PostgreSQL, and mSQL. @item @uref{http://www.multimania.com/bbrox/GtkSQL,GtkSQL} A query tool for MySQL and PostgreSQL. @item @uref{http://dbman.linux.cz/,dbMan} A query tool written in Perl. Uses DBI and Tk. @item @uref{http://www.mysql.com/Downloads/Win32/Msc201.EXE, Mascon 202} @item @uref{http://www.mysql.com/Downloads/Win32/FrMsc202.EXE, Free Mascon 202} Mascon is a powerful Win32 GUI for the administering MySQL server databases. Mascon's features include visual table design, connections to multiple servers, data and blob editing of tables, security setting, SQL color coding, dump functionality and much more. @uref{http://www.scibit.com/Products/Software/Utils/Mascon.asp,Mascon home page}. @item @uref{http://www.virtualbeer.net/dbui/,DBUI} DBUI is a Gtk graphical database editor. @item @uref{http://www.rtlabs.com/, MacSQL} GUI for MySQL, ODBC, and JDBC databases for the Mac OS. @item @uref{http://www.caleb.com.au/, JRetriever} JRetriever is a generic database front-end tool for JDBC compliant databases written with Java 2. JRetriever displays database tables/views in a Windows explorer-like front end. Users can retrieve data either by clicking on the table folder or by composing their own SQL statements with our built-in SQL editor. The tool has been tested with Oracle 8 and MySQL as the back-end databases. It requires JDK 1.3 from JavaSoft. @item @uref{http://www.jetools.com/products/databrowser/, DataBrowser} The DataBrowser is a cross-database, cross-platform data access tool. It is more user friendly than tools like SQL Plus, psql (command line based tools). It is more flexible than TOAD, ISQL, PGAccess which are GUI's that are limitied to a single platform or database. @item @uref{http://www.intrex.net/amit/software/, SQLC} The SQL Console is a standalone java application that allows you to connect to a SQL database system and issue SQL queries and updates. It has an easy-to use graphical user interface. The SQL Console uses JDBC to connect to the database systems and, therefore, with proper JDBC drivers, you can use this utility to connect to some of the most popular database systems. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_mmc.zip, MySQL MMC} MySQL MMC is a GUI Management Tool developed using kdevelop with a very good interface completely like Microsoft Enterprise Tool (for SQL Server) or Sybase Central. We can use it to manage server, database, table, index, users and to edit table data in grid or execute Sql by Query Analysis. @end itemize @cindex Web clients @item Web Clients @itemize @minus @item @uref{http://www.mysql.com/Downloads/Contrib/mysqladmin-atif-1.0.tar.gz, mysqladmin-atif-1.0.tar.gz} WWW MySQL administrator for the @code{user,} @code{db} and @code{host} tables. By Tim Sailer, modified by Atif Ghaffar @email{aghaffar@@artemedia.ch}. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-webadmin-1.0a8-rz.tar.gz, mysql-webadmin-1.0a8-rz.tar.gz} A tool written in PHP-FI to administrate MySQL databases remotely over the web within a Web-Browser. By Peter Kuppelwieser, @email{peter.kuppelwieser@@kantea.it}. Updated by Wim Bonis, @email{bonis@@kiss.de}. Not maintained anymore! @item @uref{http://www.mysql.com/Downloads/Contrib/mysqladm.tar.gz, mysqladm.tar.gz} MySQL Web Database Administration written in Perl. By Tim Sailer. @item @uref{http://www.mysql.com/Downloads/Contrib/mysqladm-2.tar.gz, mysqladm-2.tar.gz} Updated version of @file{mysqladm.tar.gz}, by High Tide. @item @uref{http://www.mysql.com/Downloads/Contrib/billowmysql.zip, billowmysql.zip} Updated version of @file{mysqladm.tar.gz}, by Ying Gao. You can get the newest version from @uref{http://civeng.com/sqldemo/, the home site}. @item @uref{http://www.mysql.com/Downloads/Contrib/myadmin-0.4.tar.gz, myadmin-0.4.tar.gz}. @item @uref{http://myadmin.cheapnet.net/, MyAdmin home page} A Web-based MySQL administrator by Mike Machado. @item @uref{http://www.mysql.com/Downloads/Contrib/phpMyAdmin_2.0.1.tar.gz,phpMyAdmin_2.0.1.tar.gz} A set of PHP3-scripts to adminstrate MySQL over the WWW. @item @uref{http://www.phpwizard.net/projects/phpMyAdmin/, phpMyAdmin home page} A PHP3 tool in the spirit of mysql-webadmin, by Tobias Ratschiller, tobias@@dnet.it. @item @uref{http://www.mysql.com/Downloads/Contrib/useradm.tar.gz, useradm.tar.gz} MySQL administrator in PHP. By Ofni Thomas @email{othomas@@vaidsystems.com}. @item @uref{http://gossamer-threads.com/perl/mysqlman/mysql.cgi, MySQLMan} Similar functionality as phpmyadmin, but written with Perl and using html templates. By Alex Krohn. @end itemize @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-editor.tar.gz,mysql-editor.tar.gz} This cgi scripts in Perl enables you to edit content of Mysql database. By Tomas Zeman. @item @uref{http://worldcommunity.com/opensource/futuresql, FutureSQL Web Database Administration Tool}. FutureSQL by Peter F. Brown, is a free, open source rapid application development Web database administration tool, written in Perl, using MySQL. It uses @code{DBI:DBD} and @code{CGI.pm}. FutureSQL allows one to easily set up config files to view, edit, delete, and otherwise process records from a MySQL database. It uses a data dictionary, configuration files and templates, and allows "pre-processing" and "post-processing" on both fields, records, and operations. @end itemize @cindex web tools @cindex tools,, web @appendixsec Web Tools @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/mod_mysql_include_1.0.tar.gz, mod_mysql_include_1.0.tar.gz} Apache module to include HTML from MySQL queries into your pages, and run update queries. Originally written to implement a simple fast low-overhead banner-rotation system. By Sasha Pachev. @item @uref{http://htcheck.sourceforge.net, htCheck} - URL checker with MySQL backend. Spidered URLs can later be queried using SQL to retrieve various kinds of information, eg. broken links. Written by Gabriele Bartolini. @item @uref{http://www.odbsoft.com/cook/sources.htm} This package has various functions for generating html code from a SQL table structure and for generating SQL statements (Select, Insert, Update, Delete) from an html form. You can build a complete forms interface to a SQL database (query, add, update, delete) without any programming! By Marc Beneteau, @email{marc@@odbsoft.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/sqlhtml.tar.gz, sqlhtml.tar.gz} SQL/HTML is an HTML database manager for MySQL using @code{DBI} 1.06. @item @uref{http://www.mysql.com/Downloads/Contrib/udmsearch-3.0.23.tar.gz, UdmSearch 3.0.23 (stable version)}. @item @uref{http://www.mysql.com/Downloads/Contrib/mnogosearch-3.1.12.tar.gz, mnogosearch 3.1.12 (development but recommended version)}. @item @uref{http://search.mnoGo.ru, UdmSearch home page} A SQL-based search engine for Internet. By Alexander I. Barkov @email{bar@@izhcom.ru}. @item @uref{http://www.mysql.com/Downloads/Contrib/wmtcl.doc, wmtcl.doc}. @item @uref{http://www.mysql.com/Downloads/Contrib/wmtcl.lex, wmtcl.lex} With this you can write HTML files with inclusions of Tcl code. By @email{vvs@@scil.npi.msu.su}. @item @uref{http://www.mysql.com/Downloads/Contrib/www-sql-0.5.7.lsm, www-sql-0.5.7.lsm}. @item @uref{http://www.mysql.com/Downloads/Contrib/www-sql-0.5.7.tar.gz, www-sql-0.5.7.tar.gz} A CGI program that parses an HTML file containing special tags, parses them, and inserts data from a MySQL database. @item @uref{http://www.mysql.com/Downloads/Contrib/genquery.zip, genquery.zip} Perl SQL database interface package for html. @item @uref{http://www.mysql.com/Downloads/Contrib/cgi++-0.8.tar.gz, cgi++-0.8.tar.gz} A macro-processor to simply writing CGI/Database programs in C++ by Sasha Pachev. @item @uref{http://www.mysql.com/Downloads/Contrib/webboard-1.0.zip, WebBoard 1.0} EU-Industries Internet-Message-Board. @item @uref{http://www.mysql.com/Downloads/Contrib/DBIx-TextIndex-0.02.tar.gz, DBIx-TextIndex-0.02.tar.gz} Full-text searching with Perl on @code{BLOB}/@code{TEXT} columns by Daniel Koch. @end itemize @cindex tools, benchmarking @cindex benchmarking, tools @appendixsec Performance Benchmarking Tools @itemize @bullet @item @uref{http://www.mysql.com/Downloads/super-smack/super-smack-1.0.tar.gz, super-smack} Multi-threaded benchmarking tool for MySQL and @strong{PostgreSQL}. Written in C++. Easy to extend to support other databases that have C/C++ client libraries. By Sasha Pachev. @end itemize @cindex tools, authentication @cindex authentication tools @appendixsec Authentication Tools @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/ascend-radius-mysql-0.7.2.patch.gz,ascend-radius-mysql-0.7.2.patch.gz} This is an authentication and logging patch using MySQL for Ascend-Radius. By @email{takeshi@@SoftAgency.co.jp}. @item @uref{http://www.mysql.com/Downloads/Contrib/icradius-0.10.tar.gz, icradius 0.10} @uref{http://www.mysql.com/Downloads/Contrib/icradius.README, icradius readme file}. @item @uref{http://www.mysql.com/Downloads/Contrib/checkpassword-0.81-mysql-0.6.6.patch.gz, checkpassword-0.81-mysql-0.6.6.patch.gz} MySQL authentication patch for QMAIL and checkpassword. These are useful for management user (mail, pop account) by MySQL. By @email{takeshi@@SoftAgency.co.jp}. @item @uref{http://www.mysql.com/Downloads/Contrib/jradius-diff.gz, jradius-diff.gz} MySQL support for Livingston's Radius 2.01. Authentication and Accounting. By Jose de Leon, @email{jdl@@thevision.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/mod_auth_mysql-2.20.tar.gz, mod_auth_mysql-2.20.tar.gz} Apache authentication module for MySQL. By Zeev Suraski, @email{bourbon@@netvision.net.il}. @c @strong{Please} register this module at: @c @url{http://bourbon.netvision.net.il/mysql/mod_auth_mysql/register.html}. The @c registering information is only used for statistical purposes and will @c encourage further development of this module! @item @uref{http://www.mysql.com/Downloads/Contrib/mod_log_mysql-1.05.tar.gz, mod_log_mysql-1.05.tar.gz} MySQL logging module for Apache. By Zeev Suraski, @email{bourbon@@netvision.net.il}. @item @uref{http://www.mysql.com/Downloads/Contrib/mypasswd-2.0.tar.gz, mypasswd-2.0.tar.gz} Extra for @code{mod_auth_mysql}. This is a little tool that allows you to add/change user records storing group and/or password entries in MySQL tables. By Harry Brueckner, @email{brueckner@@respublica.de}. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-passwd.README, mysql-passwd.README}. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql-passwd-1.2.tar.gz, mysql-passwd-1.2.tar.gz} Extra for @code{mod_auth_mysql}. This is a two-part system for use with @code{mod_auth_mysql}. @item @uref{http://www.mysql.com/Downloads/Contrib/pam_mysql.tar.gz, pam_mysql.tar.gz} This module authenticates users via @code{pam}, using MySQL. @item @uref{http://www.mysql.com/Downloads/Contrib/nsapi_auth_mysql.tar, nsapi_auth_mysql.tar} Netscape Web Server API (NSAPI) functions to authenticate (BASIC) users against MySQL tables. By Yuan John Jiang. @item @uref{http://www.mysql.com/Downloads/Contrib/qmail-1.03-mysql-0.6.6.patch.gz,qmail-1.03-mysql-0.6.6.patch.gz} Patch for qmail to authenticate users from a MySQL table. By @email{takeshi@@SoftAgency.co.jp}. @item @uref{http://www.mysql.com/Downloads/Contrib/proftpd-1.2.0rc2-fix-mysql.patch, proftpd-1.2.0rc2-fix-mysql.patch} Patch for proftpd1.2.0rc2. By @email{takeshi@@SoftAgency.co.jp}. @item @uref{http://www.mysql.com/Downloads/Contrib/pwcheck_mysql-0.1.tar.gz,pwcheck_mysql-0.1.tar.gz} An authentication module for the Cyrus IMAP server. By Aaron Newsome. @end itemize @cindex converters @appendixsec Converters @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/mssql2mysql.txt, mssql2mysql.txt} Converter from MS-SQL to MySQL. By Michael Kofler. @uref{http://www.kofler.cc/mysql/mssql2mysql.html, mssql2mysql home page}. @item @uref{http://www.mysql.com/Downloads/Contrib/dbf2mysql-1.14.tar.gz, dbf2mysql-1.14.tar.gz} Convert between @file{.dbf} files and MySQL tables. By Maarten Boekhold (@email{boekhold@@cindy.et.tudelft.nl}), William Volkman, and Michael Widenius. This converter includes rudimentary read-only support for MEMO fields. @item @uref{http://www.mysql.com/Downloads/Contrib/dbf2mysql-1.13.tgz, dbf2mysql-1.13.tgz} Convert between @file{.dbf} files and MySQL tables. By Maarten Boekhold, @email{boekhold@@cindy.et.tudelft.nl}, and Michael Widenius. This converter can't handle MEMO fields. @item @uref{http://www.mysql.com/Downloads/Contrib/dbf2mysql.zip, dbf2mysql.zip} Convert between FoxPro @file{.dbf} files and MySQL tables on Windows. By Alexander Eltsyn, @email{ae@@nica.ru} or @email{ae@@usa.net}. @item @uref{http://www.mysql.com/Downloads/Contrib/dbf2sql.zip, dbf2sql.zip} Short and simple prg that can help you transport your data from foxpro table into MySQL table. By Danko Josic. @item @uref{http://www.mysql.com/Downloads/Contrib/dump2h-1.20.gz, dump2h-1.20.gz} Convert from @code{mysqldump} output to a C header file. By Harry Brueckner, @email{brueckner@@mail.respublica.de}. @item @uref{http://www.mysql.com/Downloads/Contrib/exportsql.txt, exportsql.txt} A script that is similar to @code{access_to_mysql.txt}, except that this one is fully configurable, has better type conversion (including detection of @code{TIMESTAMP} fields), provides warnings and suggestions while converting, quotes @strong{all} special characters in text and binary data, and so on. It will also convert to @code{mSQL} v1 and v2, and is free of charge for anyone. See @uref{http://www.cynergi.net/exportsql/} for the latest version. By Pedro Freire, @email{support@@cynergi.net}. NOTE: Doesn't work with Access2! @item @uref{http://www.mysql.com/Downloads/Contrib/access_to_mysql.txt, access_to_mysql.txt} Paste this function into an Access module of a database that has the tables you want to export. See also @code{exportsql}. By Brian Andrews. NOTE: Doesn't work with Access2! @item @uref{http://www.mysql.com/Downloads/Contrib/importsql.txt, importsql.txt} A script that does the exact reverse of @code{exportsql.txt}. That is, it imports data from MySQL into an Access database via ODBC. This is very handy when combined with exportsql, because it lets you use Access for all DB design and administration, and synchronize with your actual MySQL server either way. Free of charge. See @uref{http://www.netdive.com/freebies/importsql/} for any updates. Created by Laurent Bossavit of NetDIVE. @strong{NOTE:} Doesn't work with Access2! @item @uref{http://www.mysql.com/Downloads/Contrib/mdb2sql.bas, mdb2sql.bas} Converter from Access97 to MySQL by Moshe Gurvich. @item @uref{http://www.mysql.com/Downloads/Contrib/msql2mysqlWrapper-1.0.tgz, msql2mysqlWrapper 1.0} A C wrapper from @code{mSQL} to MySQL. By @email{alfred@@sb.net} @item @uref{http://www.mysql.com/Downloads/Contrib/sqlconv.pl, sqlconv.pl} A simple script that can be used to copy fields from one MySQL table to another in bulk. Basically, you can run @code{mysqldump} and pipe it to the @code{sqlconv.pl} script. The script will parse through the @code{mysqldump} output and will rearrange the fields so they can be inserted into a new table. An example is when you want to create a new table for a different site you are working on, but the table is just a bit different (that is - fields in different order, etc.). By Steve Shreeve. @item @uref{http://www.mysql.com/Downloads/Contrib/oracledump, oracledump} Perl program to convert Oracle databases to MySQL. Has same output format as mysqldump. By Johan Andersson. @item @uref{http://www.mysql.com/Downloads/Contrib/excel2mysql.pl, excel2mysql.pl} Perl program to import Excel spreadsheets into a MySQL database. By Stephen Hurd @email{shurd@@sk.sympatico.ca} @item @uref{http://www.mysql.com/Downloads/Contrib/T2S_100.ZIP, T2S_100.ZIP}. Windows program to convert text files to MySQL databases. By Asaf Azulay. @end itemize @appendixsec Using MySQL with Other Products @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/emacs-sql-mode.tar.gz, emacs-sql-mode.tar.gz} Raw port of a SQL mode for XEmacs. Supports completion. Original by Peter D. Pezaris @email{pez@@atlantic2.sbi.com} and partial MySQL port by David Axmark. @item @uref{http://www.mysql.com/Downloads/Win32/myaccess97_1_4.zip, MyAccess97 1.4}. @item @uref{http://www.mysql.com/Downloads/Win32/myaccess2000_1_4.zip, MyAccess2000 1.4}. MyAccess is an AddIn for MS Access 97/2000 that allows you to manage MySQL databases from within Access. Main functions are: @itemize @minus @item Create/Modify Tables @item Execute Queries against MySQL @item Extract ''Create Table-Scripts'' from MySQL @item Import/Export tables from Access to MySQL and vice versa @item Log Changes @item Show a "Database Definition Report @end itemize Written by Hubertus Hiden. @uref{http://www.accessmysql.com, MyAccess homepage}. @item @uref{http://www.mysql.com/Downloads/Contrib/radius-0.3.tar.gz, radius-0.3.tar.gz} Patches for @code{radiusd} to make it support MySQL. By Wim Bonis, @email{bonis@@kiss.de}. @end itemize @cindex tools, useful @appendixsec Useful Tools @itemize @bullet @item @uref{http://worldcommunity.com/opensource/utilities/mysql_backup.html, MySQL Backup}. A backup script for MySQL. By Peter F. Brown. @item @uref{http://www.mysql.com/Downloads/Contrib/mytop, mytop} @item @uref{http://public.yahoo.com/~jzawodn/mytop/, mytop home page} mytop is a Perl program that allows you to monitor MySQL servers by viewing active threads, queries, and overall server performance numbers. By Jeremy D. Zawodny. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_watchdog.pl, mysql_watchdog.pl} Monitor the MySQL daemon for possible lockups. By Yermo Lamers, @email{yml@@yml.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/mysqltop.tar.gz, mysqltop.tar.gz} Sends a query in a fixed time interval to the server and shows the resulting table. By Thomas Wana. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_structure_dumper.tar.gz,mysql_structure_dumper.tar.gz} Prints out the structure of the all tables in a database. By Thomas Wana. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_structure_dumper.tgz, structure_dumper.tgz} Prints the structure of every table in a database. By Thomas Wana. @item @uref{http://www.mysql.com/Downloads/Contrib/mysqlsync, mysqlsync-1.0-alpha.tar.gz}. A Perl script to keep remote copies of a MySQL database in sync with a central master copy. By Mark Jeftovic. @email{markjr@@easydns.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQLTutor-0.2.tar.gz, MySQLTutor}. MySQLTutor. A MySQL tutorial for beginners. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQLDB.zip, MySQLDB.zip} A COM library for MySQL by Alok Singh. @item @uref{http://www.mysql.com/Downloads/Contrib/MySQLDB-readme.html, MySQLDB-readme.html}. @item @uref{http://www.mysql.com/Downloads/Contrib/mysql_replicate.pl, mysql_replicate.pl} Perl program that handles replication. By @email{elble@@icculus.nsg.nwu.edu} @item @uref{http://www.mysql.com/Downloads/Contrib/DBIx-TextIndex-0.02.tar.gz, DBIx-TextIndex-0.02.tar.gz} Perl script that uses reverse indexing to handle text searching. By Daniel Koch. @item @uref{http://www.mysql.com/Downloads/Contrib/dbcheck, dbcheck} Perl script that takes a backup of tables before running isamchk on them. By Elizabeth. @item @uref{http://www.mysql.com/Downloads/Contrib/mybackup}. @item @uref{http://www.mswanson.com/mybackup, mybackup home page} Wrapper for mysqldump to backup all databases. By Marc Swanson. @item @uref{http://www.mysql.com/Downloads/Contrib/mdu.pl.gz,mdu.pl.gz} Prints the storage usage of a MySQL database. @end itemize @cindex RPMs, for common tools @cindex tools, RPMs for @appendixsec RPMs for Common Tools (Most Are for RedHat 6.1) @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/perl-Data-ShowTable-3.3-2.i386.rpm,perl-Data-ShowTable-3.3-2.i386.rpm} @item @uref{http://www.mysql.com/Downloads/Contrib/perl-Msql-Mysql-modules-1.2210-2.i386.rpm,perl-Msql-Mysql-modules-1.2210-2.i386.rpm} @item @uref{http://www.mysql.com/Downloads/Contrib/php-pg-3.0.13-1.i386.rpm,php-pg-3.0.13-1.i386.rpm} @item @uref{http://www.mysql.com/Downloads/Contrib/php-pg-manual-3.0.13-1.i386.rpm,php-pg-manual-3.0.13-1.i386.rpm} @item @uref{http://www.mysql.com/Downloads/Contrib/php-pg-mysql-3.0.13-1.i386.rpm,php-pg-mysql-3.0.13-1.i386.rpm} @item @uref{http://www.mysql.com/Downloads/Contrib/phpMyAdmin-2.0.5-1.noarch.rpm,phpMyAdmin-2.0.5-1.noarch.rpm} @end itemize @cindex functions, useful @appendixsec Useful Functions @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/mysnprintf.c,mysnprintf.c} sprintf() function for SQL queries that can escape blobs. By Chunhua Liu. @end itemize @appendixsec Windows programs @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/LaunchMySQL.zip, LaunchMySQL.zip} The program launches the MySQL server, shuts it down, and display status information. By Bill Thompson @end itemize @appendixsec Uncategorized @itemize @bullet @item @uref{http://www.mysql.com/Downloads/Contrib/findres.pl, findres.pl} Find reserved words in tables. By Nem W Schlecht. @item @uref{http://www.mysql.com/Downloads/Contrib/handicap.tar.gz, handicap.tar.gz} Performance handicapping system for yachts. Uses PHP. By @email{rhill@@stobyn.ml.org}. @item @uref{http://www.mysql.com/Downloads/Contrib/hylalog-1.0.tar.gz, hylalog-1.0.tar.gz} Store @code{hylafax} outgoing faxes in a MySQL database. By Sinisa Milivojevic, @email{sinisa@@mysql.com}. @item @uref{http://www.mysql.com/Downloads/Contrib/mrtg-mysql-1.0.tar.gz, mrtg-mysql-1.0.tar.gz} MySQL status plotting with MRTG, by Luuk de Boer, @email{luuk@@wxs.nl}. @item @uref{http://www.mysql.com/Downloads/Contrib/wuftpd-2.4.2.18-mysql_support.2.tar.gz, wuftpd-2.4.2.18-mysql_support.2.tar.gz} Patches to add logging to MySQL for WU-ftpd. By Zeev Suraski, @email{bourbon@@netvision.net.il}. @item @uref{http://www.mysql.com/Downloads/Contrib/wu-ftpd-2.6.0-mysql.4.tar.gz,wu-ftpd-2.6.0-mysql.4.tar.gz} Patches to add logging to MySQL for WU-ftpd 2.6.0. By @email{takeshi@@SoftAgency.co.jp}, based on Zeev Suraski wuftpd patches. @item @uref{http://www.mysql.com/Downloads/Contrib/Old-Versions, Old-Versions} Previous versions of things found here that you probably won't be interested in. @end itemize @page @node Credits, News, Contrib, Top @appendix Credits @cindex developers, list of This appendix lists the developers, contributors, and supporters that have helped to make MySQL what it is today. @menu * Developers:: Developers at MySQL AB * Contributors:: Contributors to MySQL * Supporters:: Supporters to MySQL @end menu @node Developers, Contributors, Credits, Credits @appendixsec Developers at MySQL AB These are the developers that are or have been employed by MySQL AB to work on MySQL, roughly in the order they started to work with us. Following each developer is a small list of the tasks that the developer is responsible for, or the accomplishments they have made. @table @asis @item Michael (Monty) Widenius Has written the following parts of MySQL: @itemize @bullet @item All the main code in @code{mysqld}. @item New functions for the string library. @item Most of the @code{mysys} library. @item The @code{ISAM} and @code{MyISAM} libraries (B-tree index file handlers with index compression and different record formats). @item The @code{HEAP} library. A memory table system with our superior full dynamic hashing. In use since 1981 and published around 1984. @item The @code{replace} program (look into it, it's COOL!). @item @strong{MyODBC}, the ODBC driver for Windows95. @item Fixing bugs in MIT-pthreads to get it to work for MySQL. And also Unireg, a curses-based application tool with many utilities. @item Porting of @code{mSQL} tools like @code{msqlperl}, @code{DBD}/@code{DBI}, and @code{DB2mysql}. @item Most of crash-me and the foundation for the MySQL benchmarks. @end itemize @item David Axmark @itemize @bullet @item Coordinator and initial main writer of the @strong{Reference Manual}, including enhancements to @code{texi2html}. @item Automatic Web site updating from the manual. @item Initial Autoconf, Automake, and Libtool support. @item The licensing stuff. @item Parts of all the text files. (Nowadays only the @file{README} is left. The rest ended up in the manual.) @item Lots of testing of new features. @item Our in-house ``free'' software lawyer. @item Mailing list maintainer (who never has the time to do it right...) @item Our original portability code (more than 10 years old now). Nowadays only some parts of @code{mysys} are left. @item Someone for Monty to call in the middle of the night when he just got that new feature to work. @end itemize @item Jani Tolonen @itemize @bullet @item @code{mysqlimport} @item A lot of extensions to the @code{mysql} client. @item @code{procedure analyse()} @end itemize @item Sinisa Milivojevic @itemize @bullet @item Compression (with @code{zlib}) in the client/server protocol. @item Perfect hashing for the lexical analyzer phase. @item The MySQLGUI client. @item Maintainer of mysql++. @end itemize @item Tonu Samuel @itemize @bullet @item Our security expert. @item Vio interface (The foundation for the encrypted client/server protocol). @item MySQL Filesystem (A way to use MySQL databases as files and directories). @item The CASE Expression. @item The MD5() and COALESCE() functions. @item @code{RAID} support for @code{MyISAM} tables. @end itemize @item Sasha Pachev @itemize @bullet @item Replication. @item @code{SHOW CREATE TABLE}. @item mod_mysql_include @item cgi++ @item mysql-bench @end itemize @item Matt Wagner @itemize @bullet @item MySQL test suite. @item Our webmaster. @end itemize @item Miguel Solorzano @itemize @bullet @item Winmysqladmin. @end itemize @item Timothy Smith @itemize @bullet @item Dynamic character support. @item Responsible for MySQL configure. @end itemize @item Sergei Golubchik @itemize @bullet @item Full-text search. @item Added keys to the @code{MERGE} library. @end itemize @item Jeremy Cole @itemize @bullet @item Proofreading and editing this fine manual. @item @code{ALTER TABLE ... ORDER BY ...}. @item @code{UPDATE ... ORDER BY ...}. @item @code{DELETE ... ORDER BY ...}. @end itemize @item John Dean @itemize @bullet @item The new MySQL GUI client. @end itemize @item Indrek Siitan @itemize @bullet @item Designer/programmer of our web interface. @end itemize @end table The following non-developers are also working in/with MySQL AB: @table @asis @item Hans Kierkegaard - Responsible for MySQL license handling. @item Antti Halonen - Sales manager. @item Jonas Norrman - Handles licensing questions sent to @email{info@@mysql.com}. @item Erik Granberg - Handles MySQL partners (and a lot of other stuff). @item Allan Larsson (The BOSS for TCX DataKonsult AB). @end table @node Contributors, Supporters, Developers, Credits @appendixsec Contributors to MySQL @cindex contributors, list of Contributors to the MySQL distribution are listed here, in somewhat random order: @table @asis @item Paul DuBois Help with making the Reference Manual correct and understandable. That includes rewriting Monty's and David's attempts at English into English as other people know it. @item Gianmassimo Vigazzola @email{qwerg@@mbox.vol.it} or @email{qwerg@@tin.it} The initial port to Win32/NT. @item Kim Aldale Helped to rewrite Monty's and David's early attempts at English into English. @item Per Eric Olsson For more or less constructive criticism and real testing of the dynamic record format. @item Irena Pancirov @email{irena@@mail.yacc.it} Win32 port with Borland compiler. @code{mysqlshutdown.exe} and @code{mysqlwatch.exe} @item David J. Hughes For the effort to make a shareware SQL database. We at TcX started with @code{mSQL}, but found that it couldn't satisfy our purposes so instead we wrote a SQL interface to our application builder Unireg. @code{mysqladmin} and @code{mysql} are programs that were largely influenced by their @code{mSQL} counterparts. We have put a lot of effort into making the MySQL syntax a superset of @code{mSQL}. Many of the API's ideas are borrowed from @code{mSQL} to make it easy to port free @code{mSQL} programs to MySQL. MySQL doesn't contain any code from @code{mSQL}. Two files in the distribution (@file{client/insert_test.c} and @file{client/select_test.c}) are based on the corresponding (non-copyrighted) files in the @code{mSQL} distribution, but are modified as examples showing the changes necessary to convert code from @code{mSQL} to MySQL. (@code{mSQL} is copyrighted David J. Hughes.) @item Fred Fish For his excellent C debugging and trace library. Monty has made a number of smaller improvements to the library (speed and additional options). @item Richard A. O'Keefe For his public domain string library. @item Henry Spencer For his regex library, used in @code{WHERE column REGEXP regexp}. @item Free Software Foundation From whom we got an excellent compiler (@code{gcc}), the @code{libc} library (from which we have borrowed @file{strto.c} to get some code working in Linux), and the @code{readline} library (for the @code{mysql} client). @item Free Software Foundation & The XEmacs development team For a really great editor/environment used by almost everybody at TcX/MySQL AB/detron. @item Patrick Lynch For helping us acquire @code{http://www.mysql.com/}. @item Fred Lindberg For setting up qmail to handle the MySQL mailing list and for the incredible help we got in managing the MySQL mailing lists. @item Igor Romanenko @email{igor@@frog.kiev.ua} @code{mysqldump} (previously @code{msqldump}, but ported and enhanced by Monty). @item Yuri Dario For keeping up and extending the MySQL OS/2 port. @item Tim Bunce, Alligator Descartes For the @code{DBD} (Perl) interface. @item Tim Bunce Author of @code{mysqlhotcopy}. @item Andreas Koenig @email{a.koenig@@mind.de} For the Perl interface to MySQL. @item Eugene Chan @email{eugene@@acenet.com.sg} For porting PHP to MySQL. @item Michael J. Miller Jr. @email{mke@@terrapin.turbolift.com} For the first MySQL manual. And a lot of spelling/language fixes for the FAQ (that turned into the MySQL manual a long time ago). @item Yan Cailin First translator of the MySQL Reference Manual into simplified chinese in early 2000 on which the Big5 and HK coded (@uref{http://mysql.hitstar.com, mysql.hitstar.com}) versions were based. @uref{http://linuxdb.yeah.net, Personal home page at linuxdb.yeah.net}. @item Giovanni Maruzzelli @email{maruzz@@matrice.it} For porting iODBC (Unix ODBC). @item Chris Provenzano Portable user level pthreads. From the copyright: This product includes software developed by Chris Provenzano, the University of California, Berkeley, and contributors. We are currently using version 1_60_beta6 patched by Monty (see @file{mit-pthreads/Changes-mysql}). @item Xavier Leroy @email{Xavier.Leroy@@inria.fr} The author of LinuxThreads (used by MySQL on Linux). @item Zarko Mocnik @email{zarko.mocnik@@dem.si} Sorting for Slovenian language and the @file{cset.tar.gz} module that makes it easier to add other character sets. @item "TAMITO" @email{tommy@@valley.ne.jp} The @code{_MB} character set macros and the ujis and sjis character sets. @item Joshua Chamas @email{joshua@@chamas.com} Base for concurrent insert, extended date syntax, debugging on NT, and answering on the MySQL mailing list. @item Yves Carlier @email{Yves.Carlier@@rug.ac.be} @code{mysqlaccess}, a program to show the access rights for a user. @item Rhys Jones @email{rhys@@wales.com} (And GWE Technologies Limited) For the JDBC, a module to extract data from MySQL with a Java client. @item Dr Xiaokun Kelvin ZHU @email{X.Zhu@@brad.ac.uk} Further development of the JDBC driver and other MySQL-related Java tools. @item James Cooper @email{pixel@@organic.com} For setting up a searchable mailing list archive at his site. @item Rick Mehalick @email{Rick_Mehalick@@i-o.com} For @code{xmysql}, a graphical X client for MySQL. @item Doug Sisk @email{sisk@@wix.com} For providing RPM packages of MySQL for RedHat Linux. @item Diemand Alexander V. @email{axeld@@vial.ethz.ch} For providing RPM packages of MySQL for RedHat Linux-Alpha. @item Antoni Pamies Olive @email{toni@@readysoft.es} For providing RPM versions of a lot of MySQL clients for Intel and SPARC. @item Jay Bloodworth @email{jay@@pathways.sde.state.sc.us} For providing RPM versions for MySQL Version 3.21. @item Jochen Wiedmann @email{wiedmann@@neckar-alb.de} For maintaining the Perl @code{DBD::mysql} module. @item Therrien Gilbert @email{gilbert@@ican.net}, Jean-Marc Pouyot @email{jmp@@scalaire.fr} French error messages. @item Petr snajdr, @email{snajdr@@pvt.net} Czech error messages. @item Jaroslaw Lewandowski @email{jotel@@itnet.com.pl} Polish error messages. @item Miguel Angel Fernandez Roiz Spanish error messages. @item Roy-Magne Mo @email{rmo@@www.hivolda.no} Norwegian error messages and testing of Version 3.21.#. @item Timur I. Bakeyev @email{root@@timur.tatarstan.ru} Russian error messages. @item @email{brenno@@dewinter.com} && Filippo Grassilli @email{phil@@hyppo.com} Italian error messages. @item Dirk Munzinger @email{dirk@@trinity.saar.de} German error messages. @item Billik Stefan @email{billik@@sun.uniag.sk} Slovak error messages. @item Stefan Saroiu @email{tzoompy@@cs.washington.edu} Romanian error messages. @item Peter Feher Hungarian error messages. @item Roberto M. Serqueira Portugise error messages. @item Carsten H. Pedersen Danish error messages @item David Sacerdote @email{davids@@secnet.com} Ideas for secure checking of DNS hostnames. @item Wei-Jou Chen @email{jou@@nematic.ieo.nctu.edu.tw} Some support for Chinese(BIG5) characters. @item Wei He @email{hewei@@mail.ied.ac.cn} A lot of functionality for the Chinese(GBK) character set. @item Zeev Suraski @email{bourbon@@netvision.net.il} @code{FROM_UNIXTIME()} time formatting, @code{ENCRYPT()} functions, and @code{bison} advisor. Active mailing list member. @item Luuk de Boer @email{luuk@@wxs.nl} Ported (and extended) the benchmark suite to @code{DBI}/@code{DBD}. Have been of great help with @code{crash-me} and running benchmarks. Some new date functions. The mysql_setpermissions script. @item Jay Flaherty @email{fty@@mediapulse.com} Big parts of the Perl @code{DBI}/@code{DBD} section in the manual. @item Paul Southworth @email{pauls@@etext.org}, Ray Loyzaga @email{yar@@cs.su.oz.au} Proof-reading of the Reference Manual. @item Alexis Mikhailov @email{root@@medinf.chuvashia.su} User-definable functions (UDFs); @code{CREATE FUNCTION} and @code{DROP FUNCTION}. @item Andreas F. Bobak @email{bobak@@relog.ch} The @code{AGGREGATE} extension to UDF functions. @item Ross Wakelin @email{R.Wakelin@@march.co.uk} Help to set up InstallShield for MySQL-Win32. @item Jethro Wright III @email{jetman@@li.net} The @file{libmysql.dll} library. @item James Pereria @email{jpereira@@iafrica.com} Mysqlmanager, a Win32 GUI tool for administrating MySQL. @item Curt Sampson @email{cjs@@portal.ca} Porting of MIT-pthreads to NetBSD/Alpha and NetBSD 1.3/i386. @item Antony T. Curtis @email{antony.curtis@@olcs.net} Porting of MySQL to OS/2. @item Martin Ramsch @email{m.ramsch@@computer.org} Examples in the MySQL Tutorial. @item Steve Harvey For making @code{mysqlaccess} more secure. @item Konark IA-64 Centre of Persistent Systems Private Limited @uref{http://www.pspl.co.in/konark/}. Help with the Win64 port of the MySQL server. @item Albert Chin-A-Young. Configure updates for Tru64, large file support and better TCP wrappers support. @item John Birrell Emulation of pthread_mutex() for OS/2. @end table Other contributors, bugfinders, and testers: James H. Thompson, Maurizio Menghini, Wojciech Tryc, Luca Berra, Zarko Mocnik, Wim Bonis, Elmar Haneke, @email{jehamby@@lightside}, @email{psmith@@BayNetworks.com}, @email{duane@@connect.com.au}, Ted Deppner @email{ted@@psyber.com}, Mike Simons, Jaakko Hyvatti. And lots of bug report/patches from the folks on the mailing list. A big tribute goes to those that help us answer questions on the @code{mysql@@lists.mysql.com} mailing list: @table @asis @item Daniel Koch @email{dkoch@@amcity.com} Irix setup. @item Luuk de Boer @email{luuk@@wxs.nl} Benchmark questions. @item Tim Sailer @email{tps@@users.buoy.com} @code{DBD-mysql} questions. @item Boyd Lynn Gerber @email{gerberb@@zenez.com} SCO-related questions. @item Richard Mehalick @email{RM186061@@shellus.com} @code{xmysql}-related questions and basic installation questions. @item Zeev Suraski @email{bourbon@@netvision.net.il} Apache module configuration questions (log & auth), PHP-related questions, SQL syntax-related questions and other general questions. @item Francesc Guasch @email{frankie@@citel.upc.es} General questions. @item Jonathan J Smith @email{jsmith@@wtp.net} Questions pertaining to OS-specifics with Linux, SQL syntax, and other things that might need some work. @item David Sklar @email{sklar@@student.net} Using MySQL from PHP and Perl. @item Alistair MacDonald @email{A.MacDonald@@uel.ac.uk} Not yet specified, but is flexible and can handle Linux and maybe HP-UX. Will try to get user to use @code{mysqlbug}. @item John Lyon @email{jlyon@@imag.net} Questions about installing MySQL on Linux systems, using either @file{.rpm} files or compiling from source. @item Lorvid Ltd. @email{lorvid@@WOLFENET.com} Simple billing/license/support/copyright issues. @item Patrick Sherrill @email{patrick@@coconet.com} ODBC and VisualC++ interface questions. @item Randy Harmon @email{rjharmon@@uptimecomputers.com} @code{DBD}, Linux, some SQL syntax questions. @end table @node Supporters, , Contributors, Credits @appendixsec Supporters to MySQL @cindex contributing companies, list of The following companies has helped us finance development of MySQL by either paying us for developing a new feature, developed a MySQL feature themselves or by giving us hardware for MySQL development. @table @asis @item VA Linux / Andover.net Funded replication. @item NuSphere Editing of the MySQL manual. @item Stork Design studio The MySQL web site in use between 1998-2000. @item Intel Contributed to development on Windows and Linux platforms. @item Compaq Contributed to Development on Linux/Alpha. @item SWSoft Development on the embedded @code{mysqld} version. @item FutureQuest @code{--skip-show-variables} @end table @node News, Porting, Credits, Top @appendix MySQL change history @cindex ChangeLog @cindex changes, log @cindex log, changes This appendix lists the changes from version to version in the MySQL source code. Note that we tend to update the manual at the same time we make changes to MySQL. If you find a version listed below that you can't find on the @uref{http://www.mysql.com/downloads/,MySQL download page}, this means that the version has not yet been released! @menu * News-4.0.x:: Changes in release 4.0.x (Development; Alpha) * News-3.23.x:: Changes in release 3.23.x (Stable) * News-3.22.x:: Changes in release 3.22.x (Older; Still supported) * News-3.21.x:: Changes in release 3.21.x * News-3.20.x:: Changes in release 3.20.x * News-3.19.x:: Changes in release 3.19.x @end menu @node News-4.0.x, News-3.23.x, News, News @appendixsec Changes in release 4.0.x (Development; Alpha) We have now started to work on MySQL 4.0. We will update this section as we add new features, so that others can follow our development. Our TODO section contains what we plan to have in 4.0. @xref{TODO MySQL 4.0}. @menu * News-4.0.0:: Changes in release 4.0.0 @end menu @node News-4.0.0, , News-4.0.x, News-4.0.x @appendixsubsec Changes in release 4.0.0 @cindex changes, version 4.0 @itemize @bullet @item Multi-table @code{DELETE}. @item Don't support old client protocols prior to MySQL 3.21 any more. @item Don't include the old C API functions @code{mysql_drop_db}, @code{mysql_create_db} and @code{mysql_connect}, unless compiled with @code{USE_OLD_FUNCTIONS}. @item Renamed @code{safe_mysqld} to @code{mysqld_safe}. @item Allow @code{IN} as a synonym for @code{FROM} in @code{SHOW} commands. @item @code{SHOW INDEXES} is now a synonym for @code{SHOW INDEX}. @item Added support for symbolic links to @code{MyISAM} tables. Symlink handling is now enabled by default for Windows. @item @code{LOAD DATA FROM MASTER} ``auto-magically'' sets up a slave. @item A new @code{HANDLER} interface to @code{MyISAM} tables. @item @code{COUNT(DISTINCT)} is about 30% faster. @item @code{FULLTEXT} index creation now is much faster. @item Searching on packed (@code{CHAR}/@code{VARCHAR}) keys now is much faster. @item Added @code{SQL_CALC_FOUND_ROWS} and @code{FOUND_ROWS()}. This makes it possible to know how many rows a query would have returned without a @code{LIMIT} clause. @item Changed output format of @code{SHOW OPEN TABLES}. @item Allow @code{SELECT expression LIMIT ...}. @item Added @code{IDENTITY} as a synonym for @code{AUTO_INCREMENT} (like Sybase). @item Added @code{ORDER BY} syntax to @code{UPDATE} and @code{DELETE}. @item Optimized queries of type: @code{SELECT DISTINCT * from table_name ORDER by key_part1 LIMIT #} @end itemize @node News-3.23.x, News-3.22.x, News-4.0.x, News @appendixsec Changes in release 3.23.x (Stable) @cindex changes, version 3.23 The 3.23 release has several major features that are not present in previous versions. We have added three new table types: @table @asis @item @strong{MyISAM} A new ISAM library which is tuned for SQL and supports large files. @item @strong{BerkeleyDB} or @strong{BDB} Uses the Berkeley DB library from Sleepycat Software to implement transaction-safe tables. @item @strong{InnoDB} A transaction-safe table handler that supports row level locking, and many Oracle-like features. @end table Note that only MyISAM is available in the standard binary distribution. The 3.23 release also includes support for database replication between a master and many slaves, full-text indexing, and much more. All new features are being developed in the 4.0 version. Only bug fixes and minor enhancements to existing features will be added to 3.23. The replication code and BerkeleyDB code is still not as tested and as the rest of the code, so we will probably need to do a couple of future releases of 3.23 with small fixes for this part of the code. As long as you don't use these features, you should be quite safe with MySQL 3.23! Note that the above doesn't mean that replication or Berkeley DB don't work; We have done a lot of testing of all code, including replication and BDB without finding any problems. It only means that not as many users use this code as the rest of the code and because of this we are not yet 100% confident in this code. @menu * News-3.23.52:: Changes in release 3.23.52 * News-3.23.51:: Changes in release 3.23.51 * News-3.23.50:: Changes in release 3.23.50 * News-3.23.49:: Changes in release 3.23.49 * News-3.23.48:: Changes in release 3.23.48 * News-3.23.47:: Changes in release 3.23.47 * News-3.23.46:: Changes in release 3.23.46 * News-3.23.45:: Changes in release 3.23.45 * News-3.23.44:: Changes in release 3.23.44 * News-3.23.43:: Changes in release 3.23.43 * News-3.23.42:: Changes in release 3.23.42 * News-3.23.41:: Changes in release 3.23.41 * News-3.23.40:: Changes in release 3.23.40 * News-3.23.39:: Changes in release 3.23.39 * News-3.23.38:: Changes in release 3.23.38 * News-3.23.37:: Changes in release 3.23.37 * News-3.23.36:: Changes in release 3.23.36 * News-3.23.35:: Changes in release 3.23.35 * News-3.23.34a:: Changes in release 3.23.34a * News-3.23.34:: Changes in release 3.23.34 * News-3.23.33:: Changes in release 3.23.33 * News-3.23.32:: Changes in release 3.23.32 * News-3.23.31:: Changes in release 3.23.31 * News-3.23.30:: Changes in release 3.23.30 * News-3.23.29:: Changes in release 3.23.29 * News-3.23.28:: Changes in release 3.23.28 * News-3.23.27:: Changes in release 3.23.27 * News-3.23.26:: Changes in release 3.23.26 * News-3.23.25:: Changes in release 3.23.25 * News-3.23.24:: Changes in release 3.23.24 * News-3.23.23:: Changes in release 3.23.23 * News-3.23.22:: Changes in release 3.23.22 * News-3.23.21:: Changes in release 3.23.21 * News-3.23.20:: Changes in release 3.23.20 * News-3.23.19:: Changes in release 3.23.19 * News-3.23.18:: Changes in release 3.23.18 * News-3.23.17:: Changes in release 3.23.17 * News-3.23.16:: Changes in release 3.23.16 * News-3.23.15:: Changes in release 3.23.15 * News-3.23.14:: Changes in release 3.23.14 * News-3.23.13:: Changes in release 3.23.13 * News-3.23.12:: Changes in release 3.23.12 * News-3.23.11:: Changes in release 3.23.11 * News-3.23.10:: Changes in release 3.23.10 * News-3.23.9:: Changes in release 3.23.9 * News-3.23.8:: Changes in release 3.23.8 * News-3.23.7:: Changes in release 3.23.7 * News-3.23.6:: Changes in release 3.23.6 * News-3.23.5:: Changes in release 3.23.5 * News-3.23.4:: Changes in release 3.23.4 * News-3.23.3:: Changes in release 3.23.3 * News-3.23.2:: Changes in release 3.23.2 * News-3.23.1:: Changes in release 3.23.1 * News-3.23.0:: Changes in release 3.23.0 @end menu @node News-3.23.52, News-3.23.51, News-3.23.x, News-3.23.x @appendixsubsec Changes in release 3.23.52 @itemize @bullet @item Fixed bug in ALTERing TABLE of BDB type. @item Fixed bug when logging @code{LOAD DATA INFILE} to binary log with no active database. @item Fixed a bug in range optimiser (causing crashes). @item Fixed possible problem in replication when doing @code{DROP DATABASE} on a database with InnoDB tables. @item Fixed that @code{mysql_info()} returns 0 for 'Duplicates' when using @code{INSERT DELAYED IGNORE}. @end itemize @node News-3.23.51, News-3.23.50, News-3.23.52, News-3.23.x @appendixsubsec Changes in release 3.23.51 @itemize @bullet @item Cleaned up @code{NULL} handling for default values in @code{DESCRIBE table_name}. @item Fixed @code{truncate()} to round up negative values to the nearest integer. @item Remove end space from @code{enum} values. (This fixed a problem with @code{SHOW CREATE TABLE}). @item Fixed bug in @code{CONCAT_WS()} that cut the result. @item Changed name of variables @code{Com_show_master_stat} to @code{Com_show_master_status} and @code{Com_show_slave_stat} to @code{Com_show_slave_status}. @item Changed handling of @code{gethostbyname()} to make the client library threadsafe even if @code{gethostbyname_r} doesn't exists. @item Fixed core-dump problem when giving a wrong password string to @code{GRANT}. @item Fixed bug in @code{DROP DATABASE} with symlinked directory. @item Fixed optimization problem with @code{DATETIME} and value outside @code{DATETIME} range. @item Removed BDB documentation. @item Fixed mit-pthreads to compile with glibc 2.2 (needed for @code{make dist}). @item Fixed the @code{FLOAT(X+1,X)} is not converted to @code{FLOAT(X+2,X)}. (This also affected @code{DECIMAL}, @code{DOUBLE} and @code{REAL} types) @item Fixed the result from @code{IF()} is case in-sensitive if the 2 and third arguments are case sensitive. @item Fixed core dump problem on OSF in @code{gethostbyname_r}. @item Fixed that underflowed decimal fields is not zero filled. @item If we get an overflow when inserting @code{'+11111'} for @code{decimal(5,0) unsigned} columns, we will just drop the sign. @item Fixed optimization bug with @code{ISNULL(expression_which_cannot_be_null)} and @code{ISNULL(constant_expression)}. @item Fixed host lookup bug in the glibc library that we used with the 3.23.50 Linux-x86 binaries. @end itemize @node News-3.23.50, News-3.23.49, News-3.23.51, News-3.23.x @appendixsubsec Changes in release 3.23.50 @itemize @bullet @item Fixed problem with @code{crash-me} and gcc 3.0.4. @item Fixed that @code{@@@@unknown_variable} doesn't hang server. @item Added @code{@@@@VERSION} as a synonym for @code{VERSION()}. @item @code{SHOW VARIABLES LIKE 'xxx'} is now case insensitive. @item Fixed timeout for @code{GET_LOCK()} on HPUX with DCE threads. @item Fixed memory allocation bug in the glibc library used to build Linux binaries, which caused mysqld to die in 'free()'. @item Fixed @code{SIGINT} and @code{SIGQUIT} problems in @code{mysql}. @item Fixed bug in character table converts when used with big ( > 64K) strings. @item @code{InnoDB} now retains foreign key constraints through @code{ALTER TABLE} and @code{CREATE/DROP INDEX}. @item @code{InnoDB} now allows foreign key constraints to be added through the @code{ALTER TABLE} syntax. @item @code{InnoDB} tables can now be set to automatically grow in size (autoextend). @item Fixed some buffer overflow problems when reading startup parameters. @item Because of problems on shutdown we have now disabled named pipes on windows by default. One can enable named pipes by starting mysqld with @code{--enable-named-pipe}. @item Fixed bug when using @code{WHERE key_column = 'J' or key_column='j'}. @item Fixed core-dump bug when using @code{--log-bin} with @code{LOAD DATA INFILE} without an active database. @item Fixed bug in @code{RENAME TABLE} when used with @code{lower_case_table_names=1} (default on Windows). @item Fixed unlikely core-dump bug when using @code{DROP TABLE} on a table that was in use by a thread that also used queries on only temporary tables. @item Fixed problem with @code{SHOW CREATE TABLE} and @code{PRIMARY KEY} when using 32 indexes. @item Fixed that one can use @code{SET PASSWORD} for the anonymous user. @item Fixed core-dump bug when reading client groups from option files using @code{mysql_options()}. @item Memory leak (16 bytes per every @strong{corrupted} table) closed. @item Fixed binary builds to use @code{--enable-local-infile}. @item Update source to work with new @code{bison} version. @item Updated shell scripts to new agree with new POSIX standard. @item Fixed bug where @code{DATE_FORMAT()} returned empty string when used with @code{GROUP BY}. @end itemize @node News-3.23.49, News-3.23.48, News-3.23.50, News-3.23.x @appendixsubsec Changes in release 3.23.49 @itemize @bullet @item Don't give warning for statement that is only a comment; This is needed for @code{mysqldump --disable-keys} to work. @item Fixed unlikely caching bug when doing a join without keys. In this case the last used column for a table always returned @code{NULL}. @item Added options to make @code{LOAD DATA LOCAL INFILE} more secure. @item MySQL binary release 3.23.48 for Linux contained a new glibc library, which has serious problems under high load and RedHat 7.2. The 3.23.49 binary release doesn't have this problem. @item Fixed shutdown problem on NT. @end itemize @node News-3.23.48, News-3.23.47, News-3.23.49, News-3.23.x @appendixsubsec Changes in release 3.23.48 @itemize @bullet @item Changed to use @code{autoconf} 2.52 (from @code{autoconf} 2.13) @item Fixed bug in complicated join with @code{const} tables. @item Added internal safety checks for InnoDB. @item Some InnoDB variables was always shown in @code{SHOW VARIABLES} as @code{OFF} on high-byte-first systems (like sparc). @item Fixed problem with one thread using an InnoDB table and another thread doing an @code{ALTER TABLE} on the same table. Before that, mysqld could crash with an assertion failure in row0row.c, line 474. @item Tuned the InnoDB SQL optimizer to favor more often index searches over table scans. @item Fixed a performance problem with InnoDB tables when several large SELECT queries are run concurrently on a multiprocessor Linux computer. Large CPU-bound SELECT queries will now also generally run faster on all platforms. @item If MySQL binlogging is used, InnoDB now prints after crash recovery the latest MySQL binlog name and the offset InnoDB was able to recover to. This is useful, for example, when resynchronizing a master and a slave database in replication. @item Added better error messages to help in installation problems of InnoDB tables. @item One can now recover also MySQL temporary tables which have become orphaned inside the InnoDB tablespace. @item InnoDB now prevents a @code{FOREIGN KEY} declaration where the signedness is not the same in the referencing and referenced integer columns. @item Calling @code{SHOW CREATE TABLE} or @code{SHOW TABLE STATUS} could cause memory corruption and make mysqld to crash. Especially at risk was @code{mysqldump}, because it calls frequently @code{SHOW CREATE TABLE}. @item If inserts to several tables containing an auto-inc column were wrapped inside one @code{LOCK TABLES}, InnoDB asserted in lock0lock.c. @item In 3.23.47 we allowed several @code{NULLS} in a @code{UNIQUE} secondary index for an InnoDB table. But @code{CHECK TABLE} was not relaxed: it reports the table as corrupt. @code{CHECK TABLE} no longer complains in this situation. @item @code{SHOW GRANTS} now shows @code{REFERENCES} instead of @code{REFERENCE}. @end itemize @node News-3.23.47, News-3.23.46, News-3.23.48, News-3.23.x @appendixsubsec Changes in release 3.23.47 @itemize @bullet @item Fixed in when using the following construct: @code{SELECT ... WHERE key=@@var_name OR $key=@@var_name2} @item Restrict InnoDB keys to 500 bytes. @item InnoDB now supports @code{NULL} in keys. @item Fixed shutdown problem on HPUX. (Introduced in 3.23.46) @item Fixed core-dump bug in replication when using SELECT RELEASE_LOCK(); @item Added new command: @code{DO expression,[expression]} @item Added @code{slave-skip-errors} option @item Added statistics variables for all MySQL commands. (@code{SHOW STATUS} is now much longer). @item Fixed default values for InnoDB tables. @item Fixed that @code{GROUP BY expr DESC} works. @item Fixed bug when using @code{t1 LEFT JOIN t2 ON t2.key=constant}. @item @code{mysql_config} now also work with binary (relocated) distributions. @end itemize @node News-3.23.46, News-3.23.45, News-3.23.47, News-3.23.x @appendixsubsec Changes in release 3.23.46 @itemize @bullet @item Fixed problem with aliased temporary tables replication @item InnoDB and BDB tables will now use index when doing an @code{ORDER BY} on the whole table. @item Fixed bug where one got an empty set instead of a DEADLOCK error when using BDB tables. @item One can now kill @code{ANALYZE},@code{REPAIR} and @code{OPTIMIZE TABLE} when the thread is waiting to get a lock on the table. @item Fixed race condition in @code{ANALYZE TABLE}. @item Fixed bug when joining with caching (unlikely to happen). @item Fixed race condition when using the binary log and @code{INSERT DELAYED} which could cause the binary log to have rows that was not yet written to MyISAM tables. @item Changed caching of binary log to make replication slightly faster. @item Fixed bug in replication on Mac OS X. @end itemize @node News-3.23.45, News-3.23.44, News-3.23.46, News-3.23.x @appendixsubsec Changes in release 3.23.45 @itemize @bullet @item @code{(UPDATE|DELETE) ...WHERE MATCH} bugfix @item shutdown should now work on Darwin (Mac OS X). @item Fixed core-dump when repairing corrupted packed MyISAM files. @item @code{--core-file} now works on Solaris. @item Fix a bug which could cause InnoDB to complain if it cannot find free blocks from the buffer cache during recovery. @item Fixed bug in InnoDB insert buffer B-tree handling that could cause crashes. @item Fixed bug in InnoDB lock timeout handling. @item Fixed core dump bug in @code{ALTER TABLE} on a @code{TEMPORARY} InnoDB table. @item Fixed bug in @code{OPTIMIZE TABLE} that reset index cardinality if it was up to date. @item Fixed problem with @code{t1 LEFT_JOIN t2 ... WHERE t2.date_column IS NULL} when date_column was declared as @code{NOT NULL}. @item Fixed bug with BDB tables and keys on @code{BLOB}'s. @item Fixed bug in @code{MERGE} tables on OS with 32 bit file pointers. @item Fixed bug in @code{TIME_TO_SEC()} when using negative values. @end itemize @node News-3.23.44, News-3.23.43, News-3.23.45, News-3.23.x @appendixsubsec Changes in release 3.23.44 @itemize @bullet @item Fixed @code{Rows_examined} count in slow query log. @item Fixed bug when using a reference to a @code{AVG()} column in @code{HAVING}. @item Fixed that date functions that require correct dates, like @code{DAYOFYEAR(column)} will return @code{NULL} for @code{0000-00-00} dates. @item Fixed bug in const-propagation when comparing columns of different types. (@code{SELECT * FROM date_col="2001-01-01" and date_col=time_col}) @item Fixed bug that caused error message @code{Can't write, because of unique constraint} with some @code{GROUP BY} queries. @item Fixed problem with sjis character strings used within quoted table names. @item Fixed coredump when using @code{CREATE ... FULLTEXT} keys with other table handlers than MyISAM. @item Don't use @code{signal()} on windows because this appears to not be 100 % reliable. @item Fixed bug when doing @code{WHERE column_name=NULL} on an indexed column that had @code{NULL} values. @item Fixed bug when doing @code{LEFT JOIN ... ON (column_name = constant) WHERE column_name = constant}. @item When using replications, aborted queries that contained @code{%} could cause a core dump. @item @code{TCP_NODELAY} was not used on some systems. (Speed problem). @item Applied portability fixes for OS/2 (Patch by Yuri Dario). @end itemize The following changes are for @code{InnoDB} tables: @itemize @bullet @item Add missing @code{InnoDB} variables to @code{SHOW VARIABLES}. @item Foreign keys checking is now done for @code{InnoDB} tables. @item @code{DROP DATABASE} now works also for @code{InnoDB} tables. @item @code{InnoDB} now supports data files and raw disk partitions bigger than 4 GB on those operating systems which have big files. @item @code{InnoDB} calculates better table cardinality estimates for the MySQL optimizer. @item Accent characters in the default character set latin1 are ordered according to the MySQL ordering. NOTE: if you are using latin1 and have inserted characters whose code is > 127 to an indexed CHAR column, you should run CHECK TABLE on your table when you upgrade to 3.23.44, and drop and reimport the table if CHECK TABLE reports an error! @item A new @file{my.cnf} parameter @code{innodb_thread_concurrency} helps in performance tuning in heavily concurrent environments. @item A new @code{my.cnf} parameter @code{innodb_fast_shutdown} speeds up server shutdown. @item A new @code{my.cnf} parameter @code{innodb_force_recovery} helps to save your data in case the disk image of the database becomes corrupt. @item @code{innodb_monitor} has been improved and a new @code{innodb_table_monitor} added. @item Increased maximum key length from 500 to 7000 bytes. @item Fixed a bug in replication of auto-inc columns with multiline inserts. @item Fixed a bug when the case of letters changes in an update of an indexed secondary column. @item Fixed a hang when there are > 24 data files. @item Fixed a crash when @code{MAX(col)} is selected from an empty table, and col is a not the first column in a multi-column index. @item Fixed a bug in purge which could cause crashes. @end itemize @node News-3.23.43, News-3.23.42, News-3.23.44, News-3.23.x @appendixsubsec Changes in release 3.23.43 @itemize @bullet @item Fixed a bug in @code{INSERT DELAYED} and @code{FLUSH TABLES} introduced in 3.23.42. @item Fixed unlikely bug, which returned not matching rows, in SELECT with many tables and multi-column indexes and 'range' type. @item Fixed a unlikely core-dump bug when doing @code{EXPLAIN SELECT} when using many tables and @code{ORDER BY}. @item Fixed bug in @code{LOAD DATA FROM MASTER} when using table with @code{CHECKSUM=1}. @item Added unique error message when one gets a DEADLOCK during a transaction with BDB tables. @item Fixed problem with @code{BDB} tables and @code{UNIQUE} columns defined as @code{NULL}. @item Fixed problem with @code{myisampack} when using pre-space filled CHAR columns. @item Applied patch from Yuri Dario for OS2. @item Fixed bug in @code{--safe-user-create} @end itemize @node News-3.23.42, News-3.23.41, News-3.23.43, News-3.23.x @appendixsubsec Changes in release 3.23.42 @itemize @bullet @item Fixed problem when using @code{LOCK TABLES} and @code{BDB} tables. @item Fixed problem with @code{REPAIR TABLE} on MyISAM tables with row lengths between 65517 - 65520 bytes @item Fixed rare hang when doing @code{mysqladmin shutdown} when there was a lot of activity in other threads. @item Fixed problem with @code{INSERT DELAYED} where delay thread could be hanging on @code{upgrading locks} without any apparent reasons. @item Fixed problem with @code{myisampack} and @code{BLOB}. @item Fixes problem when one edited @code{.MRG} tables by hand. (Patch from Benjamin Pflugmann). @item Enforce that all tables in a @code{MERGE} table come from the same database. @item Fixed bug with @code{LOAD DATA INFILE} and transactional tables. @item Fix bug when using @code{INSERT DELAYED} with wrong column definition. @item Fixed coredump during @code{REPAIR} of some particularly broken tables. @item Fixed bug in @code{InnoDB} and @code{AUTO_INCREMENT} columns. @item Fixed bug in @code{InnoDB} and @code{RENAME TABLE} columns. @item Fixed critical bug in @code{InnoDB} and @code{BLOB} columns. If one has used @code{BLOB} columns larger than 8000 bytes in an @code{InnoDB} table, one must dump the table with @code{mysqldump}, drop it and restore it from the dump. @item Applied large patch for OS/2 from Yuri Dario. @item Fixed problem with @code{InnoDB} when one could get the error @code{Can't execute the given command...} even when one didn't have an active transaction. @item Applied some minor fixes that concern Gemini. @item Use real arithmetic operations even in integer context if not all arguments are integers. (Fixes uncommon bug in some integer contexts). @item Don't force everything to lower cases on Windows. (To fix problem with Windows and @code{ALTER TABLE}). Now @code{--lower_case_names} also works on Unix. @item Fixed that automatic rollback that is done when thread end doesn't lock other threads. @end itemize @node News-3.23.41, News-3.23.40, News-3.23.42, News-3.23.x @appendixsubsec Changes in release 3.23.41 @itemize @bullet @item Added option @code{--sql-mode=option[,option[,option]]}. @xref{Command-line options}. @item Fixed possible problem with @code{shutdown} on Solaris where the @file{.pid} file wasn't deleted. @item InnoDB now supports < 4 GB rows. The former limit was 8000 bytes. @item The @code{doublewrite} file flush method is used in @code{InnoDB}. It reduces the need for Unix fsync calls to a fraction and improves performance on most Unix flavors. @item You can now use the @code{InnoDB} Monitor to print a lot of @code{InnoDB} state information, including locks, to the standard output; useful in performance tuning. @item Several bugs which could cause hangs in @code{InnoDB} have been fixed. @item Split @code{record_buffer} to @code{record_buffer} and @code{record_rnd_buffer}. To make things compatible to previous MySQL versions, if @code{record_rnd_buffer} is not set, then it takes the value of @code{record_buffer}. @item Fixed optimizing bug in @code{ORDER BY} where some @code{ORDER BY} parts where wrongly removed. @item Fixed overflow bug with @code{ALTER TABLE} and @code{MERGE} tables. @item Added prototypes for @code{my_thread_init()} and @code{my_thread_end()} to @file{mysql_com.h} @item Added option @code{--safe-user-create} to @code{mysqld}. @item Fixed bug in @code{SELECT DISTINCT ... HAVING} that casued error message @code{Can't find record in '#...} @end itemize @node News-3.23.40, News-3.23.39, News-3.23.41, News-3.23.x @appendixsubsec Changes in release 3.23.40 @itemize @bullet @item Fixed problem with @code{--low-priority-updates} and @code{INSERT}'s. @item Fixed bug in slave thread when under some rare circumstances it could get 22 bytes ahead on the offset in the master. @item Added @code{slave_wait_timeout} for replication. @item Fixed problem with @code{UPDATE} and @code{BDB} tables. @item Fixed hard bug in @code{BDB} tables when using key parts. @item Fixed problem when using the @code{GRANT FILE ON database.* ...}; Previously we added the @code{DROP} privilege for the database. @item Fixed @code{DELETE FROM table_name ... LIMIT 0} and @code{UPDATE FROM table_name ... LIMIT 0} acted as though the @code{LIMIT} clause was not present (they deleted or updated all selected rows). @item @code{CHECK TABLE} now checks if an @code{AUTO_INCREMENT} column contains the value 0. @item Sending a @code{SIGHUP} to @code{mysqld} will now only flush the logs, not reset the replication. @item Fixed parser to allow floats of type @code{1.0e1} (no sign after @code{e}). @item Option @code{--force} to @code{myisamchk} now also updates states. @item Added option @code{--warnings} to @code{mysqld}. Now @code{mysqld} only prints the error @code{Aborted connection} if this option is used. @item Fixed problem with @code{SHOW CREATE TABLE} when you didn't have a @code{PRIMARY KEY}. @item Properly fixed the rename of @code{innodb_unix_file_flush_method} to @code{innodb_flush_method}. @item Fixed bug when converting @code{UNSIGNED BIGINT} to @code{DOUBLE}. This caused a problem when doing comparisons with @code{BIGINT} values outside of the signed range. @item Fixed bug in @code{BDB} tables when querying empty tables. @item Fixed a bug when using @code{COUNT(DISTINCT)} with @code{LEFT JOIN} and there wasn't any matching rows. @item Removed all documentation referring to the @code{GEMINI} table type. @code{GEMINI} is not released under an Open Source license. @end itemize @node News-3.23.39, News-3.23.38, News-3.23.40, News-3.23.x @appendixsubsec Changes in release 3.23.39 @itemize @bullet @item The @code{AUTO_INCREMENT} sequence wasn't reset when dropping and adding an @code{AUTO_INCREMENT} column. @item @code{CREATE ... SELECT} now creates non-unique indexes delayed. @item Fixed problem where @code{LOCK TABLES table_name READ} followed by @code{FLUSH TABLES} put an exclusive lock on the table. @item @code{REAL} @@variables with was represented with 2 digits when converted to strings. @item Fixed problem that client 'hung' when @code{LOAD TABLE FROM MASTER} failed. @item Running @code{myisamchk --fast --force} will no longer repair tables that only had the open count wrong. @item Added functions to handle symbolic links to make life easier in 4.0. @item We are now using the @code{-lcma} thread library on HP-UX 10.20 so that MySQL will be more stable on HP-UX. @item Fixed problem with @code{IF()} and number of decimals in the result. @item Fixed date-part extraction functions to work with dates where day and/or month is 0. @item Extended argument length in option files from 256 to 512 chars. @item Fixed problem with shutdown when @code{INSERT DELAYED} was waiting for a @code{LOCK TABLE}. @item Fixed coredump bug in @code{InnoDB} when tablespace was full. @item Fixed problem with @code{MERGE} tables and big tables (> 4G) when using @code{ORDER BY}. @end itemize @node News-3.23.38, News-3.23.37, News-3.23.39, News-3.23.x @appendixsubsec Changes in release 3.23.38 @itemize @bullet @item Fixed a bug when @code{SELECT} from @code{MERGE} table sometimes results in incorrectly ordered rows. @item Fixed a bug in @code{REPLACE()} when using the ujis character set. @item Applied Sleepycat BDB patches 3.2.9.1 and 3.2.9.2. @item Added option @code{--skip-stack-trace} to @code{mysqld}. @item @code{CREATE TEMPORARY} now works with @code{InnoDB} tables. @item @code{InnoDB} now promotes sub keys to whole keys. @item Added option @code{CONCURRENT} to @code{LOAD DATA}. @item Better error message when slave @code{max_allowed_packet} is too low to read a very long log event from the master. @item Fixed bug when too many rows where removed when using @code{SELECT DISTINCT ... HAVING}. @item @code{SHOW CREATE TABLE} now returns @code{TEMPORARY} for temporary tables. @item Added @code{Rows_examined} to slow query log. @item Fixed problems with function returning empty string when using together with a group functions and a @code{WHERE} that didn't match any rows. @item New program @code{mysqlcheck}. @item Added database name to output for administrative commands like @code{CHECK}, @code{REPAIR}, @code{OPTIMIZE}. @item Lots of portability fixes for @code{InnoDB}. @item Changed optimizer so that queries like @code{SELECT * FROM table_name,table_name2 ... ORDER BY key_part1 LIMIT #} will use index on @code{key_part1} instead of @code{filesort}. @item Fixed bug when doing @code{LOCK TABLE to_table WRITE,...; INSERT INTO to_table... SELECT ...} when @code{to_table} was empty. @item Fixed bug with @code{LOCK TABLE} and BDB tables. @end itemize @node News-3.23.37, News-3.23.36, News-3.23.38, News-3.23.x @appendixsubsec Changes in release 3.23.37 @itemize @bullet @item Fixed a bug when using @code{MATCH} in @code{HAVING} clause. @item Fixed a bug when using @code{HEAP} tables with @code{LIKE}. @item Added @code{--mysql-version} to @code{safe_mysqld} @item Changed @code{INNOBASE} to @code{InnoDB} (because the @code{INNOBASE} name was already used). All @code{configure} options and @code{mysqld} start options are now using @code{innodb} instead of @code{innobase}. This means that you have to change any configuration files where you have used @code{innobase} options before upgrading to this version! @item Fixed bug when using indexes on @code{CHAR(255) NULL} columns. @item Slave thread will now be started even if @code{master-host} is not set, as long as @code{server-id} is set and valid @code{master.info} is present @item Partial updates (terminated with kill) are now logged with a special error code to the binary log. Slave will refuse to execute them if the error code indicates the update was terminated abnormally, and will have to be recovered with @code{SET SQL_SLAVE_SKIP_COUNTER=1; SLAVE START} after a manual sanity check/correction of data integrity. @item Fixed bug that erroneously logged a drop of internal temporary table on thread termination to the binary log - bug affected replication. @item Fixed a bug in @code{REGEXP()} on 64-bit machines. @item @code{UPDATE} and @code{DELETE} with @code{WHERE unique_key_part IS NULL} didn't update/delete all rows. @item Disabled @code{INSERT DELAYED} for tables that support transactions. @item Fixed bug when using date functions on @code{TEXT}/@code{BLOB} column with wrong date format. @item UDFs now also work on Windows. (Patch by Ralph Mason) @item Fixed bug in @code{ALTER TABLE} and @code{LOAD DATA INFILE} that disabled key-sorting. These commands should now be faster in most cases. @item Fixed performance bug where reopened tables (tables that had been waiting for @code{FLUSH} or @code{REPAIR}) would not use indexes for the next query. @item Fixed problem with @code{ALTER TABLE} to Innobase tables on FreeBSD. @item Added @code{mysqld} variables @code{myisam_max_sort_file_size} and @code{myisam_max_extra_sort_file_size}. @item Initialize signals early to avoid problem with signals in Innobase. @item Applied patch for the @code{tis620} character set to make comparisons case-independent and to fix a bug in @code{LIKE} for this character set. @strong{NOTE}: All tables that uses the @code{tis620} character set must be fixed with @code{myisamchk -r} or @code{REPAIR TABLE} ! @item Added @code{--skip-safemalloc} option to @code{mysqld}. @end itemize @node News-3.23.36, News-3.23.35, News-3.23.37, News-3.23.x @appendixsubsec Changes in release 3.23.36 @itemize @bullet @item Fixed a bug that allowed you to use database names containing a @samp{.} character. This fixes a serious security issue when @code{mysqld} is run as root. @item Fixed bug when thread creation failed (could happen when doing a LOT of connections in a short time). @item Fixed some problems with @code{FLUSH TABLES} and @code{TEMPORARY} tables. (Problem with freeing the key cache and error @code{Can't reopen table...}). @item Fixed a problem in Innobase with other character sets than @code{latin1} and another problem when using many columns. @item Fixed bug that caused a core dump when using a very complex query involving @code{DISTINCT} and summary functions. @item Added @code{SET TRANSACTION ISOLATION LEVEL ...} @item Added @code{SELECT ... FOR UPDATE}. @item Fixed bug where the number of affected rows was not returned when @code{MySQL} was compiled without transaction support. @item Fixed a bug in @code{UPDATE} where keys weren't always used to find the rows to be updated. @item Fixed a bug in @code{CONCAT_WS()} where it returned wrong results. @item Changed @code{CREATE ... INSERT} and @code{INSERT ... SELECT} to not allow concurrent inserts as this could make the binary log hard to repeat. (Concurrent inserts are enabled if you are not using the binary or update log). @item Changed some macros to be able to use fast mutex with glibc 2.2. @end itemize @node News-3.23.35, News-3.23.34a, News-3.23.36, News-3.23.x @appendixsubsec Changes in release 3.23.35 @itemize @bullet @item Fixed newly introduced bug in @code{ORDER BY}. @item Fixed wrong define @code{CLIENT_TRANSACTIONS}. @item Fixed bug in @code{SHOW VARIABLES} when using @code{INNOBASE} tables. @item Setting and using user variables in @code{SELECT DISTINCT} didn't work. @item Tuned @code{SHOW ANALYZE} for small tables. @item Fixed handling of arguments in the benchmark script @file{run-all-tests}. @end itemize @node News-3.23.34a, News-3.23.34, News-3.23.35, News-3.23.x @appendixsubsec Changes in release 3.23.34a @itemize @bullet @item Added extra files to the distribution to allow @code{INNOBASE} support to be compiled. @end itemize @node News-3.23.34, News-3.23.33, News-3.23.34a, News-3.23.x @appendixsubsec Changes in release 3.23.34 @itemize @bullet @item Added the @code{INNOBASE} table handler and the @code{BDB} table handler to the MySQL source distribution. @item Updated the documentation about @code{GEMINI} tables. @item Fixed a bug in @code{INSERT DELAYED} that caused threads to hang when inserting @code{NULL} into an @code{AUTO_INCREMENT} column. @item Fixed a bug in @code{CHECK TABLE} / @code{REPAIR TABLE} that could cause a thread to hang. @item @code{REPLACE} will not replace a row that conflicts with an @code{auto_increment} generated key. @item @code{mysqld} now only sets @code{CLIENT_TRANSACTIONS} in @code{mysql->server_capabilities} if the server supports a transaction-safe handler. @item Fixed @code{LOAD DATA INFILE} to allow numeric values to be read into @code{ENUM} and @code{SET} columns. @item Improved error diagnostic for slave thread exit. @item Fixed bug in @code{ALTER TABLE ... ORDER BY}. @item Added option @code{max_user_connections} to @code{mysqld}. @item Limit query length for replication by @code{max_allowed_packet}, not the arbitrary limit of 4 MB. @item Allow space around @code{=} in argument to @code{--set-variable}. @item Fixed problem in automatic repair that could leave some threads in state @code{Waiting for table}. @item @code{SHOW CREATE TABLE} now dumps the @code{UNION()} for @code{MERGE} tables. @item @code{ALTER TABLE} now remembers the old @code{UNION()} definition. @item Fixed bug when replicating timestamps. @item Fixed bug in bidirectional replication. @item Fixed bug in the @code{BDB} table handler that occurred when using an index on multi-part key where a key part may be @code{NULL}. @item Fixed @code{MAX()} optimization on sub-key for @code{BDB} tables. @item Fixed problem where garbage results were returned when using @code{BDB} tables and @code{BLOB} or @code{TEXT} fields when joining many tables. @item Fixed a problem with @code{BDB} tables and @code{TEXT} columns. @item Fixed bug when using a @code{BLOB} key where a const row wasn't found. @item Fixed that @code{mysqlbinlog} writes the timestamp value for each query. This ensures that one gets same values for date functions like @code{NOW()} when using @code{mysqlbinlog} to pipe the queries to another server. @item Allow one to use @code{--skip-gemini}, @code{--skip-bdb} and @code{--skip-innobase} to @code{mysqld} even if these databases are not compiled in @code{mysqld}. @item One can now do @code{GROUP BY ... DESC}. @item Fixed a deadlock in the @code{SET} code, when one ran @code{SET @@foo=bar}, where @code{bar} is a column reference, an error was not properly generated. @end itemize @node News-3.23.33, News-3.23.32, News-3.23.34, News-3.23.x @appendixsubsec Changes in release 3.23.33 @itemize @bullet @item Fixed that DNS lookups are not using the same mutex as the hostname cache. This will enable known hosts to be quickly resolved even if a DNS lookup takes a long time. @item Added @code{--character-sets-dir} to @code{myisampack}. @item Removed warnings when running @code{REPAIR TABLE ... EXTENDED}. @item Fixed a bug that caused a core dump when using @code{GROUP BY} on an alias, where the alias was the same as an existing column name. @item Added @code{SEQUENCE()} as an example UDF function. @item Changed @code{mysql_install_db} to use @code{BINARY} for @code{CHAR} columns in the privilege tables. @item Changed @code{TRUNCATE table_name} to @code{TRUNCATE TABLE table_name} to use the same syntax as Oracle. Until 4.0 we will also allow @code{TRUNCATE table_name} to not crash old code. @item Fixed 'no found rows' bug in @code{MyISAM} tables when a @code{BLOB} was first part of a multi-part key. @item Fixed bug where @code{CASE} didn't work with @code{GROUP BY}. @item Added option @code{--sort-recover} to @code{myisamchk}. @item @code{myisamchk -S} and @code{OPTIMIZE TABLE} now work on Windows. @item Fixed bug when using @code{DISTINCT} on results from functions that referred to a group function, like: @example SELECT a, DISTINCT SEC_TO_TIME(sum(a)) from table_name GROUP BY a, b; @end example @item Fixed buffer overrun in @code{libmysqlclient} library. Fixed bug in handling @code{STOP} event after @code{ROTATE} event in replication. @item Fixed another buffer overrun in @code{DROP DATABASE}. @item Added @code{Table_locks_immediate} and @code{Table_locks_waited} status variables. @item Fixed bug in replication that broke slave server start with existing @code{master.info}. This fixes a bug introduced in 3.23.32. @item Added @code{SET SQL_SLAVE_SKIP_COUNTER=n} command to recover from replication glitches without a full database copy. @item Added @code{max_binlog_size} variable; the binary log will be rotated automatically when the size crosses the limit. @item Added @code{Last_error}, @code{Last_errno}, and @code{Slave_skip_counter} to @code{SHOW SLAVE STATUS}. @item Fixed bug in @code{MASTER_POS_WAIT()} function. @item Execute coredump handler on @code{SIGILL}, and @code{SIGBUS} in addition to @code{SIGSEGV}. @item On x86 Linux, print the current query and thread (connection) id, if available, in the coredump handler. @item Fixed several timing bugs in the test suite. @item Extended @code{mysqltest} to take care of the timing issues in the test suite. @item @code{ALTER TABLE} can now be used to change the definition for a @code{MERGE} table. @item Fixed creation of @code{MERGE} tables on Windows. @item Portability fixes for OpenBSD and OS2. @item Added @code{--temp-pool} option to @code{mysqld}. Using this option will cause most temporary files created to use a small set of names, rather than a unique name for each new file. This is to work around a problem in the Linux kernel dealing with creating a bunch of new files with different names. With the old behavior, Linux seems to "leak" memory, as it's being allocated to the directory entry cache instead of the disk cache. @end itemize @node News-3.23.32, News-3.23.31, News-3.23.33, News-3.23.x @appendixsubsec Changes in release 3.23.32 @itemize @bullet @item Changed code to get around compiler bug in Compaq C++ on OSF1, that broke @code{BACKUP}, @code{RESTORE}, @code{CHECK}, @code{REPAIR}, and @code{ANALYZE TABLE}. @item Added option @code{FULL} to @code{SHOW COLUMNS}. Now we show the privilege list for the columns only if this option is given. @item Fixed bug in @code{SHOW LOGS} when there weren't any BDB logs. @item Fixed a timing problem in replication that could delay sending an update to the client until a new update was done. @item Don't convert field names when using @code{mysql_list_fields()}. This is to keep this code compatible with @code{SHOW FIELDS}. @item @code{MERGE} tables didn't work on Windows. @item Fixed problem with @code{SET PASSWORD=...} on Windows. @item Added missing @file{my_config.h} to RPM distribution. @item @code{TRIM("foo" from "foo")} didn't return an empty string. @item Added @code{--with-version-suffix} to @code{configure}. @item Fixed coredump when client aborted connection without @code{mysql_close()}. @item Fixed a bug in @code{RESTORE TABLE} when trying to restore from a non-existent directory. @item Fixed a bug which caused a core dump on the slave when replicating @code{SET PASSWORD}. @item Added @code{MASTER_POS_WAIT()}. @end itemize @node News-3.23.31, News-3.23.30, News-3.23.32, News-3.23.x @appendixsubsec Changes in release 3.23.31 @itemize @bullet @item The test suite now tests all reachable BDB interface code. During testing we found and fixed many errors in the interface code. @item Using @code{HAVING} on an empty table could produce one result row when it shouldn't. @item Fixed that the MySQL RPM is not dependent on Perl5 anymore. @item Fixed some problems with @code{HEAP} tables on Windows. @item @code{SHOW TABLE STATUS} didn't show correct average row length for tables larger than 4G. @item @code{CHECK TABLE ... EXTENDED} didn't check row links for fixed size tables. @item Added option @code{MEDIUM} to @code{CHECK TABLE}. @item Fixed problem when using @code{DECIMAL()} keys on negative numbers. @item @code{HOUR()} (and some other @code{TIME} functions) on a @code{CHAR} column always returned @code{NULL}. @item Fixed security bug in something (please upgrade if you are using a earlier MySQL 3.23 version). @item Fixed buffer overflow bug when writing a certain error message. @item Added usage of @code{setrlimit()} on Linux to get @code{-O --open-files-limit=#} to work on Linux. @item Added new @code{mysqld} variable: @code{bdb_version}. @item Fixed bug when using expression of type: @example SELECT ... FROM t1 LEFT JOIN t2 ON (t1.a=t2.a) WHERE t1.a=t2.a @end example In this case the test in the @code{WHERE} clause was wrongly optimized away. @item Fixed bug in @code{MyISAM} when deleting keys with possible @code{NULL} values, but the first key-column was not a prefix-compressed text column. @item Fixed @code{mysql.server} to read the @code{mysql.server} option section instead of @code{mysql_server}. @item Fixed @code{safe_mysqld} and @code{mysql.server} to also read the @code{server} option section. @item Added @code{Threads_created} status variable to @code{mysqld}. @end itemize @node News-3.23.30, News-3.23.29, News-3.23.31, News-3.23.x @appendixsubsec Changes in release 3.23.30 @itemize @bullet @item Added @code{SHOW OPEN TABLES} command. @item Fixed that @code{myisamdump} works against old @code{mysqld} servers. @item Fixed @code{myisamchk -k#} so that it works again. @item Fixed a problem with replication when the binary log file went over 2G on 32-bit systems. @item @code{LOCK TABLES} will now automatically start a new transaction. @item Changed @code{BDB} tables to not use internal subtransactions and reuse open files to get more speed. @item Added option @code{--mysqld=#} to @code{safe_mysqld}. @item Allow hex constants in the @code{--fields-*-by} and @code{--lines-terminated-by} options to @code{mysqldump} and @code{mysqlimport}. By Paul DuBois. @item Added option @code{--safe-show-database} to @code{mysqld}. @item Added @code{have_bdb}, @code{have_gemini}, @code{have_innobase}, @code{have_raid} and @code{have_ssl} to @code{SHOW VARIABLES} to make it easy to test for supported extensions. @item Added option @code{--open-files-limit} to @code{mysqld}. @item Changed option @code{--open-files} to @code{--open-files-limit} in @code{safe_mysqld}. @item Fixed a bug where some rows were not found with @code{HEAP} tables that had many keys. @item Fixed that @code{--bdb-no-sync} works. @item Changed @code{--bdb-recover} to @code{--bdb-no-recover} as recover should be on by default. @item Changed the default number of BDB locks to 10000. @item Fixed a bug from 3.23.29 when allocating the shared structure needed for BDB tables. @item Changed @file{mysqld_multi.sh} to use configure variables. Patch by Christopher McCrory. @item Added fixing of include files for Solaris 2.8. @item Fixed bug with @code{--skip-networking} on Debian Linux. @item Fixed problem that some temporary files where reported as having the name @code{UNOPENED} in error messages. @item Fixed bug when running two simultaneous @code{SHOW LOGS} queries. @end itemize @node News-3.23.29, News-3.23.28, News-3.23.30, News-3.23.x @appendixsubsec Changes in release 3.23.29 @itemize @bullet @item Configure updates for Tru64, large file support, and better TCP wrapper support. By Albert Chin-A-Young. @item Fixed bug in @code{<=>} operator. @item Fixed bug in @code{REPLACE} with BDB tables. @item @code{LPAD()} and @code{RPAD()} will shorten the result string if it's longer than the length argument. @item Added @code{SHOW LOGS} command. @item Remove not used BDB logs on shutdown. @item When creating a table, put @code{PRIMARY} keys first, followed by @code{UNIQUE} keys. @item Fixed a bug in @code{UPDATE} involving multi-part keys where one specified all key parts both in the update and the @code{WHERE} part. In this case MySQL could try to update a record that didn't match the whole @code{WHERE} part. @item Changed drop table to first drop the tables and then the @file{.frm} file. @item Fixed a bug in the hostname cache which caused @code{mysqld} to report the hostname as @code{''} in some error messages. @item Fixed a bug with @code{HEAP} type tables; the variable @code{max_heap_table_size} wasn't used. Now either @code{MAX_ROWS} or @code{max_heap_table_size} can be used to limit the size of a @code{HEAP} type table. @item Changed the default server-id to 1 for masters and 2 for slaves to make it easier to use the binary log. @item Renamed variable @code{bdb_lock_max} to @code{bdb_max_lock}. @item Added support for @code{auto_increment} on sub-fields for BDB tables. @item Added @code{ANALYZE} of BDB tables. @item In BDB tables, we now store the number of rows; This helps to optimize queries when we need an approximation of the number of rows. @item If we get an error in a multi-row statement, we now only rollback the last statement, not the entire transaction. @item If you do a @code{ROLLBACK} when you have updated a non-transactional table you will get an error as a warning. @item Added option @code{--bdb-shared-data} to @code{mysqld}. @item Added status variable @code{Slave_open_temp_tables}. @item Added variables @code{binlog_cache_size} and @code{max_binlog_cache_size} to @code{mysqld}. @item @code{DROP TABLE}, @code{RENAME TABLE}, @code{CREATE INDEX} and @code{DROP INDEX} are now transaction endpoints. @item If you do a @code{DROP DATABASE} on a symbolic linked database, both the link and the original database is deleted. @item Fixed that @code{DROP DATABASE} works on OS/2. @item Fixed bug when doing a @code{SELECT DISTINCT ... table1 LEFT JOIN table2 ...} when table2 was empty. @item Added @code{--abort-slave-event-count} and @code{--disconnect-slave-event-count} options to @code{mysqld} for debugging and testing of replication. @item Fixed replication of temporary tables. Handles everything except slave server restart. @item @code{SHOW KEYS} now shows whether or not key is @code{FULLTEXT}. @item New script @file{mysqld_multi}. @xref{mysqld_multi, , @code{mysqld_multi}}. @item Added new script, @file{mysql-multi.server.sh}. Thanks to Tim Bunce @email{Tim.Bunce@@ig.co.uk} for modifying @file{mysql.server} to easily handle hosts running many @code{mysqld} processes. @item @file{safe_mysqld}, @file{mysql.server}, and @file{mysql_install_db} have been modified to use @code{mysql_print_defaults} instead of various hacks to read the @file{my.cnf} files. In addition, the handling of various paths has been made more consistent with how @code{mysqld} handles them by default. @item Automatically remove Berkeley DB transaction logs that no longer are in use. @item Fixed bug with several @code{FULLTEXT} indexes in one table. @item Added a warning if number of rows changes on @code{REPAIR}/@code{OPTIMIZE}. @item Applied patches for OS/2 by @code{Yuri Dario}. @item @code{FLUSH TABLES table_name} didn't always flush the index tree to disk properly. @item @code{--bootstrap} is now run in a separate thread. This fixes a problem that caused @code{mysql_install_db} to core dump on some Linux machines. @item Changed @code{mi_create()} to use less stack space. @item Fixed bug with optimizer trying to over-optimize @code{MATCH} when used with @code{UNIQUE} key. @item Changed @code{crash-me} and the MySQL benchmarks to also work with FrontBase. @item Allow @code{RESTRICT} and @code{CASCADE} after @code{DROP TABLE} to make porting easier. @item Reset status variable which could cause problem if one used @code{--slow-log}. @item Added variable @code{connect_timeout} to @code{mysql} and @code{mysqladmin}. @item Added @code{connect_timeout} as an alias for @code{timeout} for option files read by @code{mysql_options()}. @end itemize @node News-3.23.28, News-3.23.27, News-3.23.29, News-3.23.x @appendixsubsec Changes in release 3.23.28 @itemize @bullet @item Added new options @code{--pager[=...]}, @code{--no-pager}, @code{--tee=...} and @code{--no-tee} to the @code{mysql} client. The new corresponding interactive commands are @code{pager}, @code{nopager}, @code{tee} and @code{notee}. @xref{mysql, , @code{mysql}}, @code{mysql --help} and the interactive help for more information. @item Fixed crash when automatic repair of @code{MyISAM} table failed. @item Fixed a major performance bug in the table locking code when one constantly had a LOT of @code{SELECT}, @code{UPDATE} and @code{INSERT} statements running. The symptom was that the @code{UPDATE} and @code{INSERT} queries were locked for a long time while new @code{SELECT} statements were executed before the updates. @item When reading @code{options_files} with @code{mysql_options()} the @code{return-found-rows} option was ignored. @item One can now specify @code{interactive-timeout} in the option file that is read by @code{mysql_options()}. This makes it possible to force programs that run for a long time (like @code{mysqlhotcopy}) to use @code{interactive_timeout} instead of @code{wait_timeout}. @item Added to the slow query log the time and the user name for each logged query. If you are using @code{--log-long-format} then also queries that do not use an index are logged, even if the query takes less than @code{long_query_time} seconds. @item Fixed a problem in @code{LEFT JOIN} which caused all columns in a reference table to be @code{NULL}. @item Fixed a problem when using @code{NATURAL JOIN} without keys. @item Fixed a bug when using a multi-part keys where the first part was of type @code{TEXT} or @code{BLOB}. @item @code{DROP} of temporary tables wasn't stored in the update/binary log. @item Fixed a bug where @code{SELECT DISTINCT * ... LIMIT #} only returned one row. @item Fixed a bug in the assembler code in @code{strstr()} for sparc and cleaned up the @file{global.h} header file to avoid a problem with bad aliasing with the compiler submitted with RedHat 7.0. (Reported by Trond Eivind Glomsr�d) @item The option @code{--skip-networking} now works properly on NT. @item Fixed a long outstanding bug in the @code{ISAM} tables when a row with a length of more than 65K was shortened by a single byte. @item Fixed a bug in @code{MyISAM} when running multiple updating processes on the same table. @item Allow one to use @code{FLUSH TABLE tablename}. @item Added @code{--replicate-ignore-table}, @code{--replicate-do-table}, @code{--replicate-wild-ignore-table}, @code{--replicate-wild-do-table}. @item Changed all log files to use our own @code{IO_CACHE} mechanism instead of @code{FILE} to avoid OS problems when there are many files open. @item Added options @code{--open-files} and @code{--timezone} to @code{safe_mysqld}. @item Fixed a fatal bug in @code{CREATE TEMPORARY TABLE ... SELECT ...}. @item Fixed a problem with @code{CREATE TABLE ... SELECT NULL}. @item Added variables @code{large_file_support},@code{net_read_timeout}, @code{net_write_timeout} and @code{query_buffer_size} to @code{SHOW VARIABLES}. @item Added status variables @code{created_tmp_files} and @code{sort_merge_passes} to @code{SHOW STATUS}. @item Fixed a bug where we didn't allow an index name after the @code{FOREIGN KEY} definition. @item Added @code{TRUNCATE table_name} as a synonym for @code{DELETE FROM table_name}. @item Fixed a bug in a BDB key compare function when comparing part keys. @item Added variable @code{bdb_lock_max} to @code{mysqld}. @item Added more tests to the benchmark suite. @item Fixed an overflow bug in the client code when using overly long database names. @item @code{mysql_connect()} now aborts on Linux if the server doesn't answer in @code{timeout} seconds. @item @code{SLAVE START} did not work if you started with @code{--skip-slave-start} and had not explicitly run @code{CHANGE MASTER TO}. @item Fixed the output of @code{SHOW MASTER STATUS} to be consistent with @code{SHOW SLAVE STATUS}. (It now has no directory in the log name.) @item Added @code{PURGE MASTER LOGS TO}. @item Added @code{SHOW MASTER LOGS}. @item Added @code{--safemalloc-mem-limit} option to @code{mysqld} to simulate memory shortage when compiled @code{--with-debug=full}. @item Fixed several coredumps in out-of-memory conditions. @item @code{SHOW SLAVE STATUS} was using an uninitialized mutex if the slave had not been started yet. @item Fixed bug in @code{ELT()} and @code{MAKE_SET()} when the query used a temporary table. @item @code{CHANGE MASTER TO} without specifying @code{MASTER_LOG_POS} would set it to 0 instead of 4 and hit the magic number in the master binlog. @item @code{ALTER TABLE ... ORDER BY ...} syntax added. This will create the new table with the rows in a specific order. @end itemize @node News-3.23.27, News-3.23.26, News-3.23.28, News-3.23.x @appendixsubsec Changes in release 3.23.27 @itemize @bullet @item Fixed a bug where the automatic repair of MyISAM tables sometimes failed when the data file was corrupt. @item Fixed a bug in @code{SHOW CREATE} when using @code{AUTO_INCREMENT} columns. @item Changed BDB tables to use new compare function in Berkeley DB 3.2.3. @item You can now use Unix sockets with @code{mit-pthreads}. @item Added the latin5 (turkish) character set. @item Small portability fixes. @end itemize @node News-3.23.26, News-3.23.25, News-3.23.27, News-3.23.x @appendixsubsec Changes in release 3.23.26 @itemize @bullet @item Fixed @code{<>} to work properly with @code{NULL}. @item Fixed a problem with @code{SUBSTRING_INDEX()} and @code{REPLACE()}. (Patch by Alexander Igonitchev) @item Fix @code{CREATE TEMPORARY TABLE IF NOT EXISTS} not to give an error if the table exists. @item If you don't create a @code{PRIMARY KEY} in a BDB table, a hidden @code{PRIMARY KEY} will be created. @item Added read-only-key optimization to BDB tables. @item @code{LEFT JOIN} in some cases preferred a full table scan when there was no @code{WHERE} clause. @item When using @code{--log-slow-queries}, don't count the time waiting for a lock. @item Fixed bug in lock code on Windows which could cause the key cache to report that the key file was crashed even if it was okay. @item Automatic repair of @code{MyISAM} tables if you start @code{mysqld} with @code{--myisam-recover}. @item Removed the @code{TYPE=} keyword from @code{CHECK} and @code{REPAIR}. Allow @code{CHECK} options to be combined. (You can still use @code{TYPE=} but this usage is deprecated.) @item Fixed mutex bug in the binary replication log - long update queries could be read only in part by the slave if it did it at the wrong time, which was not fatal, but resulted in a performance-degrading reconnect and a scary message in the error log. @item Changed the format of the binary log - added magic number, server version, binlog version. Added server id and query error code for each query event. @item Replication thread from the slave now will kill all the stale threads from the same server. @item Long replication user names were not being handled properly. @item Added @code{--replicate-rewrite-db} option to @code{mysqld}. @item Added @code{--skip-slave-start} option to @code{mysqld}. @item Updates that generated an error code (such as @code{INSERT INTO foo(some_key) values (1),(1)}) erroneously terminated the slave thread. @item Added optimization of queries where @code{DISTINCT} is only used on columns from some of the tables. @item Allow floating-point numbers where there is no sign after the exponent (like 1e1). @item @code{SHOW GRANTS} didn't always show all column grants. @item Added @code{--default-extra-file=#} to all MySQL clients. @item Columns referenced in @code{INSERT} statements now are initialized properly. @item @code{UPDATE} didn't always work when used with a range on a timestamp that was part of the key that was used to find rows. @item Fixed a bug in @code{FULLTEXT} index when inserting a @code{NULL} column. @item Changed to use @code{mkstemp()} instead of @code{tempnam()}. Based on a patch from John Jones. @end itemize @node News-3.23.25, News-3.23.24, News-3.23.26, News-3.23.x @appendixsubsec Changes in release 3.23.25 @itemize @bullet @item Fixed that @code{databasename} works as second argument to @code{mysqlhotcopy}. @item @code{UMASK} and @code{UMASK_DIR} can now be specified in octal. @item Added @code{RIGHT JOIN}. This makes @code{RIGHT} a reserved word. @item Added @code{@@@@IDENTITY} as a synonym for @code{LAST_INSERT_ID()}. (This is for Visual Basic compatibility.) @item Fixed a bug in @code{myisamchk} and @code{REPAIR} when using @code{FULLTEXT} index. @item @code{LOAD DATA INFILE} now works with FIFOs. (Patch by Toni L. Harbaugh-Blackford.) @item @code{FLUSH LOGS} broke replication if you specified a log name with an explicit extension as the value of the @code{log-bin} option. @item Fixed a bug in @code{MyISAM} with packed multi-part keys. @item Fixed crash when using @code{CHECK TABLE} on Windows. @item Fixed a bug where @code{FULLTEXT} index always used the koi8_ukr character set. @item Fixed privilege checking for @code{CHECK TABLE}. @item The @code{MyISAM} repair/reindex code didn't use the @code{--tempdir} option for its temporary files. @item Added @code{BACKUP TABLE/RESTORE TABLE}. @item Fixed coredump on @code{CHANGE MASTER TO} when the slave did not have the master to start with. @item Fixed incorrect @code{Time} in the processlist for @code{Connect} of the slave thread. @item The slave now logs when it connects to the master. @item Fixed a core dump bug when doing @code{FLUSH MASTER} if you didn't specify a filename argument to @code{--log-bin}. @item Added missing @file{ha_berkeley.x} files to the MySQL Windows @item Fixed some mutex bugs in the log code that could cause thread blocks if new log files couldn't be created. @item Added lock time and number of selected processed rows to slow query log. @item Added @code{--memlock} option to @code{mysqld} to lock @code{mysqld} in memory on systems with the @code{mlockall()} call (like in Solaris). @item @code{HEAP} tables didn't use keys properly. (Bug from 3.23.23.) @item Added better support for @code{MERGE} tables (keys, mapping, creation, documentation...). @xref{MERGE}. @item Fixed bug in @code{mysqldump} from 3.23 which caused some @code{CHAR} columns not to be quoted. @item Merged @code{analyze}, @code{check}, @code{optimize} and repair code. @item @code{OPTIMIZE TABLE} is now mapped to @code{REPAIR} with statistics and sorting of the index tree. This means that for the moment it only works on @code{MyISAM} tables. @item Added a pre-alloced block to root_malloc to get fewer mallocs. @item Added a lot of new statistics variables. @item Fixed @code{ORDER BY} bug with BDB tables. @item Removed warning that @code{mysqld} couldn't remove the @file{.pid} file under Windows. @item Changed @code{--log-isam} to log @strong{MyISAM} tables instead of isam tables. @item Fixed @code{CHECK TABLE} to work on Windows. @item Added file mutexes to make @code{pwrite()} safe on Windows. @end itemize @node News-3.23.24, News-3.23.23, News-3.23.25, News-3.23.x @appendixsubsec Changes in release 3.23.24 @itemize @bullet @item Added @code{mysqld} variable @code{created_tmp_disk_tables}. @item To make it possible to reliably dump and restore tables with @code{TIMESTAMP(X)} columns, MySQL now reports columns with @code{X} other than 14 or 8 to be strings. @item Changed sort order for latin1 as it was before MySQL Version 3.23.23. Any table with @code{CHAR} columns that may have characters with ASCII values greater than 128 that was created or modified with 3.23.22 must be repaired! @item Fixed small memory leak introduced from 3.23.22 when creating a temporary table. @item Fixed problem with BDB tables and reading on a unique (not primary) key. @item Restored the win1251 character set (it's now only marked deprecated). @end itemize @node News-3.23.23, News-3.23.22, News-3.23.24, News-3.23.x @appendixsubsec Changes in release 3.23.23 @itemize @bullet @item Changed sort order for 'German'; All tables created with 'German' sortorder must be repaired with @code{REPAIR TABLE} or @code{myisamchk} before use! @item Added option @code{--core-file} to @code{mysqld} to get a core file on Linux if @code{mysqld} dies on the SIGSEGV signal. @item MySQL client @code{mysql} now starts with option @code{--no-named-commands} (@code{-g}) by default. This option can be disabled with @code{--enable-named-commands} (@code{-G}). This may cause incompatibility problems in some cases, for example, in SQL scripts that use named commands without a semicolon, etc. ! Long format commands still work from the first line. @item Fixed a problem when using many pending @code{DROP TABLE} statements at the same time. @item Optimizer didn't use keys properly when using @code{LEFT JOIN} on an empty table. @item Added shorter help text when invoking @code{mysqld} with incorrect options. @item Fixed non-fatal @code{free()} bug in @code{mysqlimport}. @item Fixed bug in @code{MyISAM} index handling of @code{DECIMAL}/@code{NUMERIC} keys. @item Fixed a bug in concurrent insert in @code{MyISAM} tables; In some contexts, usage of @code{MIN(key_part)} or @code{MAX(key_part)} returned an empty set. @item Updated @code{mysqlhotcopy} to use the new @code{FLUSH TABLES table_list} syntax. Only tables which are being backed up are flushed now. @item Changed behavior of @code{--enable-thread-safe-client} so that both non-threaded (@code{-lmysqlclient}) and threaded (@code{-lmysqlclient_r}) libraries are built. Users who linked against a threaded @code{-lmysqlclient} will need to link against @code{libmysqlclient_r} now. @item Added atomic @code{RENAME} command. @item Don't count entries with @code{NULL} in @code{COUNT(DISTINCT ...)}. @item Changed @code{ALTER TABLE}, @code{LOAD DATA INFILE} on empty tables and @code{INSERT ... SELECT ...} on empty tables to create non-unique indexes in a separate batch with sorting. This will make the above calls much faster when you have many indexes. @item @code{ALTER TABLE} now logs the first used insert_id correctly. @item Fixed crash when adding a default value to a @code{BLOB} column. @item Fixed a bug with @code{DATE_ADD/DATE_SUB} where it returned a datetime instead of a date. @item Fixed a problem with the thread cache which made some threads show up as @code{***DEAD***} in @code{SHOW PROCESSLIST}. @item Fixed a lock in our thr_rwlock code, which could make selects that run at the same time as concurrent inserts crash. This only affects systems that don't have the @code{pthread_rwlock_rdlock} code. @item When deleting rows with a non-unique key in a HEAP table, all rows weren't always deleted. @item Fixed bug in range optimizer for HEAP tables for searches on a part index. @item Fixed that @code{SELECT} on part keys works with BDB tables. @item Fixed @code{INSERT INTO bdb_table ... SELECT} to work with BDB tables. @item @code{CHECK TABLE} now updates key statistics for the table. @item @code{ANALYZE TABLE} will now only update tables that have been changed since thee last @code{ANALYZE}. Note that this is a new feature and tables will not be marked to be analyzed until they are updated in any way with 3.23.23 or newer. For older tables, you have to do @code{CHECK TABLE} to update the key distribution. @item Fixed some minor privilege problems with @code{CHECK}, @code{ANALYZE}, @code{REPAIR} and @code{SHOW CREATE} commands. @item Added @code{CHANGE MASTER TO} command. @item Added @code{FAST}, @code{QUICK} @code{EXTENDED} check types to @code{CHECK TABLES}. @item Changed @code{myisamchk} so that @code{--fast} and @code{--check-changed-tables} are also honored with @code{--sort-index} and @code{--analyze}. @item Fixed fatal bug in @code{LOAD TABLE FROM MASTER} that did not lock the table during index re-build. @item @code{LOAD DATA INFILE} broke replication if the database was excluded from replication. @item More variables in @code{SHOW SLAVE STATUS} and @code{SHOW MASTER STATUS}. @item @code{SLAVE STOP} now will not return until the slave thread actually exits. @item Full text search via the @code{MATCH} function and @code{FULLTEXT} index type. (For MyISAM files). This makes @code{FULLTEXT} a reserved word. @end itemize @node News-3.23.22, News-3.23.21, News-3.23.23, News-3.23.x @appendixsubsec Changes in release 3.23.22 @itemize @bullet @item Fixed that @code{lex_hash.h} is created properly for each MySQL distribution. @item Fixed that @code{MASTER} and @code{COLLECTION} are not reserved words. @item The log generated by @code{--slow-query-log} didn't contain the whole queries. @item Fixed that open transactions in BDB tables are rolled back if the connection is closed unexpectedly. @item Added workaround for a bug in @code{gcc} 2.96 (intel) and @code{gcc} 2.9 (Ia64) in @code{gen_lex_hash.c}. @item Fixed memory leak in the client library when using @code{host=} in the @code{my.cnf} file. @item Optimized functions that manipulate the hours/minutes/seconds. @item Fixed bug when comparing the result of @code{DATE_ADD()}/@code{DATE_SUB()} against a number. @item Changed the meaning of @code{-F, --fast} for @code{myisamchk}. Added option @code{-C, --check-only-changed} to @code{myisamchk}. @item Added @code{ANALYZE table_name} to update key statistics for tables. @item Changed binary items @code{0x...} to be regarded as integers by default. @item Fix for SCO and @code{SHOW PROCESSLIST}. @item Added @code{auto-rehash} on reconnect for the @code{mysql} client. @item Fixed a newly introduced bug in @code{MyISAM}, where the index file couldn't get bigger than 64M. @item Added @code{SHOW MASTER STATUS} and @code{SHOW SLAVE STATUS}. @end itemize @node News-3.23.21, News-3.23.20, News-3.23.22, News-3.23.x @appendixsubsec Changes in release 3.23.21 @itemize @bullet @item Added @code{mysql_character_set_name(MYSQL *mysql)} function to the MySQL C API. @item Made the update log @code{ASCII 0} safe. @item Added the @code{mysql_config} script. @item Fixed problem when using @code{<} or @code{>} with a char column that was only partly indexed. @item One would get a core dump if the log file was not readable by the MySQL user. @item Changed @code{mysqladmin} to use the @code{CREATE DATABASE}/@code{DROP DATABASE} commands instead of the old deprecated API calls. @item Fixed @code{chown} warning in @code{safe_mysqld}. @item Fixed a bug in @code{ORDER BY} that was introduced in 3.23.19. @item Only optimize the @code{DELETE FROM tbl_name} to do a drop+create of the table if we are in @code{AUTOCOMMIT} mode. (Needed for BDB tables). @item Added extra checks to avoid index corruption when the @code{ISAM}/@code{MyISAM} index files gets full during an @code{INSERT}/@code{UPDATE}. @item @code{myisamchk} didn't correctly update row checksum when used with @code{-ro} (This only gave an warning in subsequent runs). @item Fixed bug in @code{REPAIR TABLE} so that it works with tables without indexes. @item Fixed buffer overrun in @code{DROP DATABASE} @item @code{LOAD TABLE FROM MASTER} is sufficiently bug-free to announce it as a feature. @item @code{MATCH} and @code{AGAINST} are now reserved words. @end itemize @node News-3.23.20, News-3.23.19, News-3.23.21, News-3.23.x @appendixsubsec Changes in release 3.23.20 @itemize @bullet @item Fixed bug in 3.23.19; @code{DELETE FROM tbl_name} removed the .frm file. @item Added @code{SHOW CREATE TABLE}. @end itemize @node News-3.23.19, News-3.23.18, News-3.23.20, News-3.23.x @appendixsubsec Changes in release 3.23.19 @itemize @bullet @item Changed copyright for all files to GPL for the server code and utilities and LGPL for the client libraries. @item Fixed bug where all rows matching weren't updated on a @code{MyISAM} table when doing update based on key on a table with many keys and some key changed values. @item The Linux MySQL RPM's and binaries are now statically linked with a linuxthread version that has faster mutex handling when used with MySQL. @item @code{ORDER BY} can now uses @code{REF} keys to find subset the rows that needs to be sorted. @item Changed name of @code{print_defaults} to @code{my_print_defaults} to avoid name confusion. @item Fixed @code{NULLIF()} to work according to ANSI SQL99. @item Added @code{net_read_timeout} and @code{net_write_timeout} as startup parameters to @code{mysqld}. @item Fixed bug that destroyed index when doing @code{myisamchk --sort-records} on a table with prefix compressed index. @item Added pack_isam and myisampack to the standard MySQL distribution. @item Added the syntax @code{BEGIN WORK} (the same as @code{BEGIN}). @item Fixed core dump bug when using @code{ORDER BY} on a @code{CONV()} expression. @item Added @code{LOAD TABLE FROM MASTER} @item Added @code{FLUSH MASTER} and @code{FLUSH SLAVE} @item Fixed big/little endian problem in the replication @end itemize @node News-3.23.18, News-3.23.17, News-3.23.19, News-3.23.x @appendixsubsec Changes in release 3.23.18 @itemize @bullet @item Fixed a problem from 3.23.17 when choosing character set on the client side. @item Added @code{FLUSH TABLES WITH READ LOCK} to make a global lock suitable to make a copy of MySQL data files. @item @code{CREATE TABLE ... SELECT ... PROCEDURE} now works. @item Internal temporary tables will now uses compressed index when using @code{GROUP BY} on @code{VARCHAR/CHAR} columns. @item Fixed a problem when locking the same table with both a @code{READ} and a @code{WRITE} lock. @item Fixed problem with myisamchk and @code{RAID} tables. @end itemize @node News-3.23.17, News-3.23.16, News-3.23.18, News-3.23.x @appendixsubsec Changes in release 3.23.17 @itemize @bullet @item Fixed a bug in @code{find_in_set()} when the first argument was @code{NULL}. @item Added table locks to Berkeley DB. @item Fixed a bug with @code{LEFT JOIN} and @code{ORDER BY} where the first table had only one matching row. @item Added 4 sample @code{my.cnf} example files in the @file{support-files} directory. @item Fixed @code{duplicated key} problem when doing big @code{GROUP BY}'s. (This bug was probably introduced in 3.23.15). @item Changed syntax for @code{INNER JOIN} to match ANSI SQL. @item Added @code{NATURAL JOIN} syntax. @item A lot of fixes in the @code{BDB} interface. @item Added handling of @code{--no-defaults} and @code{--defaults-file} to @code{safe_mysqld.sh} and @code{mysql_install_db.sh}. @item Fixed bug in reading compressed tables with many threads. @item Fixed that @code{USE INDEX} works with @code{PRIMARY} keys. @item Added @code{BEGIN} statement to start a transaction in @code{AUTOCOMMIT} mode. @item Added symbolic links support for Windows. @item Changed protocol to let client know if the server is in AUTOCOMMIT mode and if there is a pending transaction. If there is a pending transaction the client library will give an error before reconnecting to the server to let the client know that the server did a rollback. The protocol is still backward compatible with old clients @item @code{KILL} now works on a thread that is locked on a 'write' to a dead client. @item Fixed memory leak in the replication slave thread. @item Added new option @code{log-slave-updates} to allow daisy-chaining the slaves. @item Fixed compile error on FreeBSD and other systems where @code{pthread_t} is not the same as @code{int}. @item Fixed master shutdown aborting the slave thread. @item Fixed a race condition in @code{INSERT DELAYED} code when doing @code{ALTER TABLE}. @item Added deadlock detection sanity checks to @code{INSERT DELAYED} @end itemize @node News-3.23.16, News-3.23.15, News-3.23.17, News-3.23.x @appendixsubsec Changes in release 3.23.16 @itemize @bullet @item Added option @code{TYPE=QUICK} to @code{CHECK} and @code{REPAIR}. @item Fixed bug in @code{REPAIR TABLE} when the table was in use by other threads. @item Added a thread cache to make it possible to debug MySQL with @code{gdb} when one does a lot of reconnects. This will also improve systems where you can't use persistent connections. @item Lots of fixes in the Berkeley DB interface. @item @code{UPDATE IGNORE} will not abort if an update results in a @code{DUPLICATE_KEY} error. @item Put @code{CREATE TEMPORARY TABLE} commands in the update log. @item Fixed bug in handling of masked IP numbers in the privilege tables. @item Fixed bug with @code{delayed_key_writes} tables and @code{CHECK TABLE}. @item Added @code{replicate-do-db} and @code{replicate-ignore-db} options to restrict which databases get replicated @item Added @code{SQL_LOG_BIN} option @end itemize @node News-3.23.15, News-3.23.14, News-3.23.16, News-3.23.x @appendixsubsec Changes in release 3.23.15 @itemize @bullet @item To start @code{mysqld} as @code{root}, you must now use the @code{--user=root} option. @item Added interface to Berkeley DB. (This is not yet functional; Play with it at your own risk!) @item Replication between master and slaves. @item Fixed bug that other threads could steal a lock when a thread had a lock on a table and did a @code{FLUSH TABLES} command. @item Added the @code{slow_launch_time} variable and the @code{Slow_launch_threads} status variable to @code{mysqld}. These can be examined with @code{mysqladmin variables} and @code{mysqladmin extended-status}. @item Added functions @code{INET_NTOA()} and @code{INET_ATON()}. @item The default type of @code{IF()} now depends on the second and third arguments and not only on the second argument. @item Fixed case when @code{myisamchk} could go into a loop when trying to repair a crashed table. @item Don't write @code{INSERT DELAYED} to update log if @code{SQL_LOG_UPDATE=0}. @item Fixed problem with @code{REPLACE} on @code{HEAP} tables. @item Added possible character sets and time zone to @code{SHOW VARIABLES}. @item Fixed bug in locking code that could result it locking problems with concurrent inserts under high load. @item Fixed a problem with @code{DELETE} of many rows on a table with compressed keys where MySQL scanned the index to find the rows. @item Fixed problem with @code{CHECK} on table with deleted keyblocks. @item Fixed a bug in reconnect (at the client side) where it didn't free memory properly in some contexts. @item Fixed problems in update log when using @code{LAST_INSERT_ID()} to update a table with an auto_increment key. @item Added function @code{NULLIF()}. @item Fixed bug when using @code{LOAD DATA INFILE} on a table with @code{BLOB/TEXT} columns. @item Optimised MyISAM to be faster when inserting keys in sorted order. @item @code{EXPLAIN SELECT ...} now also prints out whether MySQL needs to create a temporary table or use file sorting when resolving the @code{SELECT}. @item Added optimization to skip @code{ORDER BY} parts where the part is a constant expression in the @code{WHERE} part. Indexes can now be used even if the @code{ORDER BY} doesn't match the index exactly, as long as all the not used index parts and all the extra @code{ORDER BY} columns are constants in the @code{WHERE} clause. @xref{MySQL indexes}. @item @code{UPDATE} and @code{DELETE} on a whole unique key in the @code{WHERE} part, is now faster than before. @item Changed @code{RAID_CHUNKSIZE} to be in 1024 byte increments. @item Fixed coredump in LOAD_FILE(NULL). @end itemize @node News-3.23.14, News-3.23.13, News-3.23.15, News-3.23.x @appendixsubsec Changes in release 3.23.14 @itemize @bullet @item Fixed a bug in @code{CONCAT()} where one of the arguments was a function that returned a modified argument. @item Fixed a critical bug in @code{myisamchk}, where it updated the header in the index file when one only checked the table. This confused the @code{mysqld} daemon if it updated the same table at the same time. Now the status in the index file is only updated if one uses @code{--update-state}. With older @code{myisamchk} versions you should use @code{--read-only} when only checking tables, if there is the slightest chance that the @code{mysqld} server is working on the table at the same time! @item Fixed that @code{DROP TABLE} is logged in the update log. @item Fixed problem when searching on @code{DECIMAL()} key field where the column data contained leading zeros. @item Fix bug in @code{myisamchk} when the auto_increment isn't the first key. @item Allow @code{DATETIME} in ISO8601 format: 2000-03-12T12:00:00 @item Dynamic character sets. A @code{mysqld} binary can now handle many different character sets (you can choose which when starting @code{mysqld}). @item Added command @code{REPAIR TABLE}. @item Added C API function @code{mysql_thread_safe()}. @item Added the @code{UMASK_DIR} environment variable. @item Added function @code{CONNECTION_ID()}. @item When using @code{=} on @code{BLOB} or @code{VARCHAR BINARY} keys, where only a part of the column was indexed, the whole column of the result row wasn't compared. @item Fix for sjis character set and @code{ORDER BY}. @item When running in ANSI mode, don't allow columns to be used that aren't in the @code{GROUP BY} part. @end itemize @node News-3.23.13, News-3.23.12, News-3.23.14, News-3.23.x @appendixsubsec Changes in release 3.23.13 @itemize @bullet @item Fixed problem when doing locks on the same table more than 2 times in the same @code{LOCK TABLE} command; This fixed the problem one got when running the test-ATIS test with @code{--fast} or @code{--check-only-changed}. @item Added option @code{SQL_BUFFER_RESULT} to @code{SELECT}. @item Removed end space from double/float numbers in results from temporary tables. Added @code{CHECK TABLE} command. @item Added changes for MyISAM in 3.23.12 that didn't get into the source distribution because of CVS problems. @item Fixed bug so that @code{mysqladmin shutdown} will wait for the local server to close down. @item Fixed a possible endless loop when calculating timestamp. @item Added @code{print_defaults} to the @file{.rpm} files. Removed @code{mysqlbug} from the client @file{.rpm} file. @end itemize @node News-3.23.12, News-3.23.11, News-3.23.13, News-3.23.x @appendixsubsec Changes in release 3.23.12 @itemize @bullet @item Fixed bug in @code{MyISAM} involving @code{REPLACE ... SELECT ...} which could give a corrupted table. @item Fixed bug in @code{myisamchk} where it wrongly reset the auto_increment value. @item LOTS of patches for Linux Alpha. MySQL now appears to be relatively stable on Alpha. @item Changed @code{DISTINCT} on @code{HEAP} temporary tables to use hashed keys to quickly find duplicated rows. This mostly concerns queries of type @code{SELECT DISTINCT ... GROUP BY ...}. This fixes a problem where not all duplicates were removed in queries of the above type. In addition, the new code is MUCH faster. @item Added patches to make MySQL compile on Mac OS X. @item Added option @code{IF NOT EXISTS} to @code{CREATE DATABASE}. @item Added options @code{--all-databases} and @code{--databases} to @code{mysqldump} to allow dumping of many databases at the same time. @item Fixed bug in compressed @code{DECIMAL()} index in @code{MyISAM} tables. @item Fixed bug when storing 0 into a timestamp. @item When doing @code{mysqladmin shutdown} on a local connection, @code{mysqladmin} now waits until the pidfile is gone before terminating. @item Fixed core dump with some @code{COUNT(DISTINCT ...)} queries. @item Fixed that @code{myisamchk} works properly with RAID:ed tables. @item Fixed problem with @code{LEFT JOIN} and @code{key_field IS NULL}. @item Fixed bug in @code{net_clear()} which could give the error @code{Aborted connection} in the MySQL clients. @item Added options @code{USE INDEX (key_list)} and @code{IGNORE INDEX (key_list)} as join parameters in @code{SELECT}. @item @code{DELETE} and @code{RENAME} should now work on @code{RAID} tables. @end itemize @node News-3.23.11, News-3.23.10, News-3.23.12, News-3.23.x @appendixsubsec Changes in release 3.23.11 @itemize @bullet @item Allow the @code{ALTER TABLE tbl_name ADD (field_list)} syntax. @item Fixed problem with optimizer that could sometimes use wrong keys. @item Fixed that @code{GRANT/REVOKE ALL PRIVILEGES} doesn't affect @code{GRANT OPTION}. @item Removed extra @code{)} from the output of @code{SHOW GRANTS} @item Fixed problem when storing numbers in timestamps. @item Fix problem with timezones that have half hour offsets. @item Allow the syntax @code{UNIQUE INDEX} in @code{CREATE} statements. @item @code{mysqlhotcopy} - fast on-line hot-backup utility for local MySQL databases. By Tim Bunce. @item New more secure @code{mysqlaccess}. Thanks to Steve Harvey for this. @item Added options @code{--i-am-a-dummy} and @code{--safe-updates} to @code{mysql}. @item Added variables @code{select_limit} and @code{max_join_size} to @code{mysql}. @item Added sql variables: @code{SQL_MAX_JOIN_SIZE} and @code{SQL_SAFE_UPDATES}. @item Added @code{READ LOCAL} lock that doesn't lock the table for concurrent inserts. (This is used by @code{mysqldump}). @item Changed that @code{LOCK TABLES ... READ} doesn't anymore allow concurrent inserts. @item Added option @code{--skip-delay-key-write} to @code{mysqld}. @item Fixed security problem in the protocol regarding password checking. @item @code{_rowid} can now be used as an alias for an integer type unique indexed column. @item Added back blocking of @code{SIGPIPE} when compiling with @code{--thread-safe-clients} to make things safe for old clients. @end itemize @node News-3.23.10, News-3.23.9, News-3.23.11, News-3.23.x @appendixsubsec Changes in release 3.23.10 @itemize @bullet @item Fixed bug in 3.23.9 where memory wasn't properly freed when doing @code{LOCK TABLES}. @end itemize @node News-3.23.9, News-3.23.8, News-3.23.10, News-3.23.x @appendixsubsec Changes in release 3.23.9 @itemize @bullet @item Fixed problem that affected queries that did arithmetic on group functions. @item Fixed problem with timestamps and @code{INSERT DELAYED}. @item Fixed that @code{date_column BETWEEN const_date AND const_date} works. @item Fixed problem when only changing a 0 to @code{NULL} in a table with @code{BLOB/TEXT} columns. @item Fixed bug in range optimizer when using many key parts and or on the middle key parts: @code{WHERE K1=1 and K3=2 and (K2=2 and K4=4 or K2=3 and K4=5)} @item Added command @code{source} to @code{mysql} to allow reading of batch files inside the @code{mysql} client. Original patch by Matthew Vanecek. @item Fixed critical problem with the @code{WITH GRANT OPTION} option. @item Don't give an unnecessary @code{GRANT} error when using tables from many databases in the same query. @item Added VIO wrapper (needed for SSL support ; By Andrei Errapart and T�nu Samuel). @item Fixed optimizer problem on @code{SELECT} when using many overlapping indexes. MySQL should now be able to choose keys even better when there is many keys to choose from. @item Changed optimizer to prefer a range key instead of a ref key when the range key can uses more columns than the ref key (which only can use columns with =). For example, the following type of queries should now be faster: @code{SELECT * from key_part_1=const and key_part_2 > const2} @item Fixed bug that a change of all @code{VARCHAR} columns to @code{CHAR} columns didn't change row type from dynamic to fixed. @item Disabled floating-point exceptions for FreeBSD to fix core dump when doing @code{SELECT floor(pow(2,63))}. @item Changed @code{mysqld} startup option @code{--delay-key-write} to @code{--delay-key-write-for-all-tables} @item Added @code{read-next-on-key} to @code{HEAP} tables. This should fix all problems with @code{HEAP} tables when using not @code{UNIQUE} keys. @item Added print of default arguments options to all clients. @item Added @code{--log-slow-queries} to @code{mysqld} to log all queries that take a long time to a separate log file with a time of how long the query took. @item Fixed core dump when doing @code{WHERE key_column=RAND(...)} @item Fixed optimization bug in @code{SELECT ... LEFT JOIN ... key_column IS NULL}, when @code{key_column} could contain @code{NULL} values. @item Fixed problem with 8-bit characters as separators in @code{LOAD DATA INFILE}. @end itemize @node News-3.23.8, News-3.23.7, News-3.23.9, News-3.23.x @appendixsubsec Changes in release 3.23.8 @itemize @bullet @item Fixed problem when handling indexfiles larger than 8G. @item Added latest patches to mit-pthreads for NetBSD. @item Fixed problem with timezones that are < GMT -11. @item Fixed a bug when deleting packed keys in @code{NISAM}. @item Fixed problem with @code{ISAM} when doing some @code{ORDER BY ... DESC} queries. @item Fixed bug when doing a join on a text key which didn't cover the whole key. @item Option @code{--delay-key-write} didn't enable delayed key writing. @item Fixed update of @code{TEXT} column which only involved case changes. @item Fixed that @code{INSERT DELAYED} doesn't update timestamps that are given. @item Added function @code{YEARWEEK()} and options @code{x}, @code{X}, @code{v} and @code{V} to @code{DATE_FORMAT()}. @item Fixed problem with @code{MAX(indexed_column)} and HEAP tables. @item Fixed problem with @code{BLOB NULL} keys and @code{LIKE} "prefix%". @item Fixed problem with @code{MyISAM} and fixed-length rows < 5 bytes. @item Fixed problem that could cause MySQL to touch freed memory when doing very complicated @code{GROUP BY} queries. @item Fixed core dump if you got a crashed table where an @code{ENUM} field value was too big. @end itemize @node News-3.23.7, News-3.23.6, News-3.23.8, News-3.23.x @appendixsubsec Changes in release 3.23.7 @itemize @bullet @item Fixed workaround under Linux to avoid problems with @code{pthread_mutex_timedwait}, which is used with @code{INSERT DELAYED}. @xref{Linux}. @item Fixed that one will get a 'disk full' error message if one gets disk full when doing sorting (instead of waiting until we got more disk space). @item Fixed a bug in @code{MyISAM} with keys > 250 characters. @item In @code{MyISAM} one can now do an @code{INSERT} at the same time as other threads are reading from the table. @item Added variable @code{max_write_lock_count} to @code{mysqld} to force a @code{READ} lock after a certain number of @code{WRITE} locks. @item Inverted flag @code{delayed_key_write} on @code{show variables}. @item Renamed variable @code{concurrency} to @code{thread_concurrency}. @item The following functions are now multi-byte-safe: @code{LOCATE(substr,str)}, @code{POSITION(substr IN str)}, @code{LOCATE(substr,str,pos)}, @code{INSTR(str,substr)}, @code{LEFT(str,len)}, @code{RIGHT(str,len)}, @code{SUBSTRING(str,pos,len)}, @code{SUBSTRING(str FROM pos FOR len)}, @code{MID(str,pos,len)}, @code{SUBSTRING(str,pos)}, @code{SUBSTRING(str FROM pos)}, @code{SUBSTRING_INDEX(str,delim,count)}, @code{RTRIM(str)}, @code{TRIM([[BOTH | TRAILING] [remstr] FROM] str)}, @code{REPLACE(str,from_str,to_str)}, @code{REVERSE(str)}, @code{INSERT(str,pos,len,newstr)}, @code{LCASE(str)}, @code{LOWER(str)}, @code{UCASE(str)} and @code{UPPER(str)}; Patch by Wei He. @item Fix core dump when releasing a lock from a non-existent table. @item Remove locks on tables before starting to remove duplicates. @item Added option @code{FULL} to @code{SHOW PROCESSLIST}. @item Added option @code{--verbose} to @code{mysqladmin}. @item Fixed problem when automatically converting HEAP to MyISAM. @item Fixed bug in HEAP tables when doing insert + delete + insert + scan the table. @item Fixed bugs on Alpha with @code{REPLACE()} and @code{LOAD DATA INFILE}. @item Added @code{mysqld} variable @code{interactive_timeout}. @item Changed the argument to @code{mysql_data_seek()} from @code{ulong} to @code{ulonglong}. @end itemize @node News-3.23.6, News-3.23.5, News-3.23.7, News-3.23.x @appendixsubsec Changes in release 3.23.6 @itemize @bullet @item Added @code{mysqld} option @code{-O lower_case_table_names=@{0|1@}} to allow users to force table names to lowercase. @item Added @code{SELECT ... INTO DUMPFILE}. @item Added @code{mysqld} option @code{--ansi} to make some functions @code{ANSI SQL} compatible. @item Temporary tables now starts with @code{#sql}. @item Added quoting of identifiers with @code{`} (@code{"} in @code{--ansi} mode). @item Changed to use snprintf() when printing floats to avoid some buffer overflows on FreeBSD. @item Made @code{[floor()} overflow safe on FreeBSD. @item Added option @code{--quote-names} to @code{mysqldump} @item Fixed bug that one could make a part of a @code{PRIMARY KEY NOT NULL}. @item Fixed @code{encrypt()} to be thread safe and not reuse buffer. @item Added @code{mysql_odbc_escape_string()} function to support big5 characters in MyODBC. @item Rewrote the table handler to use classes. This introduces a lot of new code, but will make table handling faster and better. @item Added patch by Sasha for user-defined variables. @item Changed that @code{FLOAT} and @code{DOUBLE} (without any length modifiers) are not anymore fixed decimal point numbers. @item Changed the meaning of @code{FLOAT(X)}: Now this is the same as @code{FLOAT} if X <= 24 and a @code{DOUBLE} if 24 < X <= 53. @item @code{DECIMAL(X)} is now an alias for @code{DECIMAL(X,0)} and @code{DECIMAL} is now an alias for @code{DECIMAL(10,0)}. The same goes for @code{NUMERIC}. @item Added option @code{ROW_FORMAT=@{default | dynamic | static | compressed@}} to @code{CREATE_TABLE}. @item @code{DELETE FROM table_name} didn't work on temporary tables. @item Changed function @code{CHAR_LENGTH()} to be multi-byte character safe. @item Added function @code{ORD(string)}. @end itemize @node News-3.23.5, News-3.23.4, News-3.23.6, News-3.23.x @appendixsubsec Changes in release 3.23.5 @itemize @bullet @item Fixed some Y2K problems in the new date handling in 3.23. @item Fixed problem with @code{SELECT DISTINCT ... ORDER BY RAND()}. @item Added patches by Sergei A. Golubchik for text searching on the MyISAM level. @item Fixed cache overflow problem when using full joins without keys. @item Fixed some configure issues. @item Some small changes to make parsing faster. @item @code{ALTER TABLE} + adding a column after the last field didn't work. @item Fixed problem when using an auto_increment column in two keys @item One can now with MyISAM have the auto_increment part as a sub part: @code{CREATE TABLE foo (a int not null auto_increment, b char(5), primary key (b,a))} @item Fixed bug in MyISAM with packed char keys that could be @code{NULL}. @item @code{AS} on fieldname with @code{CREATE TABLE table_name SELECT ...} didn't work. @item Allow use of @code{NATIONAL} and @code{NCHAR} when defining character columns. This is the same as not using @code{BINARY}. @item Don't allow @code{NULL} columns in a @code{PRIMARY KEY} (only in @code{UNIQUE} keys). @item Clear @code{LAST_INSERT_ID} if one uses this in ODBC: @code{WHERE auto_increment_column IS NULL}. This seems to fix some problems with Access. @item @code{SET SQL_AUTO_IS_NULL=0|1} now turns on/off the handling of searching after the last inserted row with @code{WHERE auto_increment_column IS NULL}. @item Added new @code{mysqld} variable @code{concurrency} for Solaris. @item Added option @code{--relative} to @code{mysqladmin} to make @code{extended-status} more useful to monitor changes. @item Fixed bug when using @code{COUNT(DISTINCT ...)} on an empty table. @item Added support for the Chinese character set GBK. @item Fixed problem with @code{LOAD DATA INFILE} and @code{BLOB} columns. @item Added bit operator @code{~} (negation). @item Fixed problem with @code{UDF} functions. @end itemize @node News-3.23.4, News-3.23.3, News-3.23.5, News-3.23.x @appendixsubsec Changes in release 3.23.4 @itemize @bullet @item Inserting a @code{DATETIME} into a @code{TIME} column will not anymore try to store 'days' in it. @item Fixed problem with storage of float/double on little endian machines. (This affected @code{SUM()}.) @item Added connect timeout on TCP/IP connections. @item Fixed problem with @code{LIKE} "%" on an index that may have @code{NULL} values. @item @code{REVOKE ALL PRIVILEGES} didn't revoke all privileges. @item Allow creation of temporary tables with same name as the original table. @item When granting a user a grant option for a database, he couldn't grant privileges to other users. @item New command: @code{SHOW GRANTS FOR user} (by Sinisa). @item New @code{date_add} syntax: @code{date/datetime + INTERVAL # interval_type}. By Joshua Chamas. @item Fixed privilege check for @code{LOAD DATA REPLACE}. @item Automatic fixing of broken include files on Solaris 2.7 @item Some configure issues to fix problems with big file system detection. @item @code{REGEXP} is now case insensitive if you use non-binary strings. @end itemize @node News-3.23.3, News-3.23.2, News-3.23.4, News-3.23.x @appendixsubsec Changes in release 3.23.3 @itemize @bullet @item Added patches for MIT-pthreads on NetBSD. @item Fixed range bug in MyISAM. @item @code{ASC} is now the default again for @code{ORDER BY}. @item Added @code{LIMIT} to @code{UPDATE}. @item New client function: @code{mysql_change_user()}. @item Added character set to @code{SHOW VARIABLES}. @item Added support of @code{--[whitespace]} comments. @item Allow @code{INSERT into tbl_name VALUES ()}, that is, you may now specify an empty value list to insert a row in which each column is set to its default value. @item Changed @code{SUBSTRING(text FROM pos)} to conform to ANSI SQL. (Before this construct returned the rightmost 'pos' characters). @item @code{SUM()} with @code{GROUP BY} returned 0 on some systems. @item Changed output for @code{SHOW TABLE STATUS}. @item Added @code{DELAY_KEY_WRITE} option to @code{CREATE TABLE}. @item Allow @code{AUTO_INCREMENT} on any key part. @item Fixed problem with @code{YEAR(NOW())} and @code{YEAR(CURDATE())}. @item Added @code{CASE} construct. @item New function @code{COALESCE()}. @end itemize @node News-3.23.2, News-3.23.1, News-3.23.3, News-3.23.x @appendixsubsec Changes in release 3.23.2 @itemize @bullet @item Fixed range optimizer bug: @code{SELECT * FROM table_name WHERE key_part1 >= const AND (key_part2 = const OR key_part2 = const)}. The bug was that some rows could be duplicated in the result. @item Running @code{myisamchk} without @code{-a} updated the index distribution wrong. @item @code{SET SQL_LOW_PRIORITY_UPDATES=1} gave parse error before. @item You can now update indexes columns that are used in the @code{WHERE} clause. @code{UPDATE tbl_name SET KEY=KEY+1 WHERE KEY > 100} @item Date handling should now be a bit faster. @item Added handling of fuzzy dates (dates where day or month is 0): (Like: 1999-01-00) @item Fixed optimization of @code{SELECT ... WHERE key_part1=const1 AND key_part_2=const2 AND key_part1=const4 AND key_part2=const4} ; Indextype should be @code{range} instead of @code{ref}. @item Fixed @code{egcs} 1.1.2 optimizer bug (when using @code{BLOB}s) on Linux Alpha. @item Fixed problem with @code{LOCK TABLES} combined with @code{DELETE FROM table}. @item MyISAM tables now allow keys on @code{NULL} and @code{BLOB/TEXT} columns. @item The following join is now much faster: @code{SELECT ... FROM t1 LEFT JOIN t2 ON ... WHERE t2.not_null_column IS NULL}. @item @code{ORDER BY} and @code{GROUP BY} can be done on functions. @item Changed handling of 'const_item' to allow handling of @code{ORDER BY RAND()}. @item Indexes are now used for @code{WHERE key_column = function}. @item Indexes are now used for @code{WHERE key_column = column_name} even if the columns are not identically packed. @item Indexes are now used for @code{WHERE column_name IS NULL}. @item Changed heap tables to be stored in low_byte_first order (to make it easy to convert to MyISAM tables) @item Automatic change of HEAP temporary tables to MyISAM tables in case of 'table is full' errors. @item Added option @code{--init-file=file_name} to @code{mysqld}. @item @code{COUNT(DISTINCT value, [value, ...])} @item @code{CREATE TEMPORARY TABLE} now creates a temporary table, in its own namespace, that is automatically deleted if connection is dropped. @item New reserved words (required for @code{CASE}): @code{CASE, THEN, WHEN, ELSE and END}. @item New functions @code{EXPORT_SET()} and @code{MD5()}. @item Support for the GB2312 Chinese character set. @end itemize @node News-3.23.1, News-3.23.0, News-3.23.2, News-3.23.x @appendixsubsec Changes in release 3.23.1 @itemize @bullet @item Fixed some compilation problems. @end itemize @node News-3.23.0, , News-3.23.1, News-3.23.x @appendixsubsec Changes in release 3.23.0 @itemize @bullet @item A new table handler library (@code{MyISAM}) with a lot of new features. @xref{MyISAM}. @item You can create in-memory @code{HEAP} tables which are extremely fast for lookups. @item Support for big files (63 bit) on OSes that support big files. @item New function @code{LOAD_FILE(filename)} to get the contents of a file as a string value. @item New operator @code{<=>} which will act as @code{=} but will return TRUE if both arguments are @code{NULL}. This is useful for comparing changes between tables. @item Added the ODBC 3.0 @code{EXTRACT(interval FROM datetime)} function. @item Columns defined as @code{FLOAT(X)} is not rounded on storage and may be in scientific notation (1.0 E+10) when retrieved. @item @code{REPLACE} is now faster than before. @item Changed @code{LIKE} character comparison to behave as @code{=}; This means that @code{'e' LIKE '@'e'} (if the line doesn't display correctly, the latter 'e' means a French 'e' with a dot above) is now true. @item @code{SHOW TABLE STATUS} returns a lot of information about the tables. @item Added @code{LIKE} to the @code{SHOW STATUS} command. @item Added privilege column to @code{SHOW COLUMNS}. @item Added columns @code{packed} and @code{comment} to @code{SHOW INDEX}. @item Added comments to tables (with @code{CREATE TABLE ... COMMENT "xxx"}). @item Added @code{UNIQUE}, as in @code{CREATE TABLE table_name (col int not null UNIQUE)} @item New create syntax: @code{CREATE TABLE table_name SELECT ...} @item New create syntax: @code{CREATE TABLE IF NOT EXISTS ...} @item Allow creation of @code{CHAR(0)} columns. @item @code{DATE_FORMAT()} now requires @samp{%} before any format character. @item @code{DELAYED} is now a reserved word (sorry about that :( ). @item An example procedure is added: @code{analyse}, file: @file{sql_analyse.c}. This will describe the data in your query. Try the following: @example SELECT ... FROM ... WHERE ... PROCEDURE ANALYSE([max elements,[max memory]]) @end example This procedure is extremely useful when you want to check the data in your table! @item @code{BINARY} cast to force a string to be compared case sensitively. @item Added option @code{--skip-show-database} to @code{mysqld}. @item Check if a row has changed in an @code{UPDATE} now also works with @code{BLOB}/@code{TEXT} columns. @item Added the @code{INNER} join syntax. @strong{NOTE}: This made @code{INNER} a reserved word! @item Added support for netmasks to the hostname in the MySQL tables. You can specify a netmask using the @code{IP/NETMASK} syntax. @item If you compare a @code{NOT NULL DATE/DATETIME} column with @code{IS NULL}, this is changed to a compare against @code{0} to satisfy some ODBC applications. (By @email{shreeve@@uci.edu}). @item @code{NULL IN (...)} now returns @code{NULL} instead of @code{0}. This will ensure that @code{null_column NOT IN (...)} doesn't match @code{NULL} values. @item Fix storage of floating-point values in @code{TIME} columns. @item Changed parsing of @code{TIME} strings to be more strict. Now the fractional second part is detected (and currently skipped). The following formats are supported: @table @code @item [[DAYS] [H]H:]MM:]SS[.fraction] @item [[[[[H]H]H]H]MM]SS[.fraction] @end table @item Detect (and ignore) second fraction part from @code{DATETIME}. @item Added the @code{LOW_PRIORITY} attribute to @code{LOAD DATA INFILE}. @item The default index name now uses the same case as the used column name. @item Changed default number of connections to 100. @item Use bigger buffers when using @code{LOAD DATA INFILE}. @item @code{DECIMAL(x,y)} now works according to ANSI SQL. @item Added aggregate UDF functions. Thanks to Andreas F. Bobak @email{bobak@@relog.ch} for this! @item @code{LAST_INSERT_ID()} is now updated for @code{INSERT INTO ... SELECT}. @item Some small changes to the join table optimizer to make some joins faster. @item @code{SELECT DISTINCT} is much faster; It uses the new @code{UNIQUE} functionality in @code{MyISAM}. One difference compared to MySQL Version 3.22 is that the output of @code{DISTINCT} is not sorted anymore. @item All C client API macros are now functions to make shared libraries more reliable. Because of this, you can no longer call @code{mysql_num_fields()} on a @code{MYSQL} object, you must use @code{mysql_field_count()} instead. @item Added use of @code{LIBEWRAP}; Patch by Henning P . Schmiedehausen. @item Don't allow @code{AUTO_INCREMENT} for other than numerical columns. @item Using @code{AUTO_INCREMENT} will now automatically make the column @code{NOT NULL}. @item Show @code{NULL} as the default value for AUTO_INCREMENT columns. @item Added @code{SQL_BIG_RESULT}; @code{SQL_SMALL_RESULT} is now default. @item Added a shared library RPM. This enhancement was contributed by David Fox (dsfox@@cogsci.ucsd.edu). @item Added a @code{--enable-large-files/--disable-large-files} switch to @code{configure}. See @file{configure.in} for some systems where this is automatically turned off because of broken implementations. @item Upgraded @code{readline} to 4.0. @item New @code{CREATE TABLE} options: @code{PACK_KEYS} and @code{CHECKSUM}. @item Added @code{mysqld} option @code{--default-table-type}. @end itemize @node News-3.22.x, News-3.21.x, News-3.23.x, News @appendixsec Changes in release 3.22.x (Older; Still supported) @cindex changes, version 3.22 The 3.22 version has faster and safer connect code than version 3.21, as well as a lot of new nice enhancements. The reason for not including these changes As there aren't really any MAJOR changes, upgrading from 3.21 to 3.22 should be very easy and painless. @xref{Upgrading-from-3.21}. @menu * News-3.22.35:: Changes in release 3.22.35 * News-3.22.34:: Changes in release 3.22.34 * News-3.22.33:: Changes in release 3.22.33 * News-3.22.32:: Changes in release 3.22.32 * News-3.22.31:: Changes in release 3.22.31 * News-3.22.30:: Changes in release 3.22.30 * News-3.22.29:: Changes in release 3.22.29 * News-3.22.28:: Changes in release 3.22.28 * News-3.22.27:: Changes in release 3.22.27 * News-3.22.26:: Changes in release 3.22.26 * News-3.22.25:: Changes in release 3.22.25 * News-3.22.24:: Changes in release 3.22.24 * News-3.22.23:: Changes in release 3.22.23 * News-3.22.22:: Changes in release 3.22.22 * News-3.22.21:: Changes in release 3.22.21 * News-3.22.20:: Changes in release 3.22.20 * News-3.22.19:: Changes in release 3.22.19 * News-3.22.18:: Changes in release 3.22.18 * News-3.22.17:: Changes in release 3.22.17 * News-3.22.16:: Changes in release 3.22.16 * News-3.22.15:: Changes in release 3.22.15 * News-3.22.14:: Changes in release 3.22.14 * News-3.22.13:: Changes in release 3.22.13 * News-3.22.12:: Changes in release 3.22.12 * News-3.22.11:: Changes in release 3.22.11 * News-3.22.10:: Changes in release 3.22.10 * News-3.22.9:: Changes in release 3.22.9 * News-3.22.8:: Changes in release 3.22.8 * News-3.22.7:: Changes in release 3.22.7 * News-3.22.6:: Changes in release 3.22.6 * News-3.22.5:: Changes in release 3.22.5 * News-3.22.4:: Changes in release 3.22.4 * News-3.22.3:: Changes in release 3.22.3 * News-3.22.2:: Changes in release 3.22.2 * News-3.22.1:: Changes in release 3.22.1 * News-3.22.0:: Changes in release 3.22.0 @end menu @node News-3.22.35, News-3.22.34, News-3.22.x, News-3.22.x @appendixsubsec Changes in release 3.22.35 @itemize @bullet @item Fixed problem with @code{STD()}. @item Merged changes from the newest @code{ISAM} library from 3.23. @item Fixed problem with @code{INSERT DELAYED}. @item Fixed a bug core dump when using a @code{LEFT JOIN}/@code{STRAIGHT_JOIN} on a table with only one row. @end itemize @node News-3.22.34, News-3.22.33, News-3.22.35, News-3.22.x @appendixsubsec Changes in release 3.22.34 @itemize @bullet @item Fixed problem with @code{GROUP BY} on @code{TINYBLOB} columns; This caused bugzilla to not show rows in some queries. @item Had to do total recompile of the Windows binary version as VC++ didn't compile all relevant files for 3.22.33 :( @end itemize @node News-3.22.33, News-3.22.32, News-3.22.34, News-3.22.x @appendixsubsec Changes in release 3.22.33 @itemize @bullet @item Fixed problems in Windows when locking tables with @code{LOCK TABLE} @item Quicker kill of @code{SELECT DISTINCT} queries. @end itemize @node News-3.22.32, News-3.22.31, News-3.22.33, News-3.22.x @appendixsubsec Changes in release 3.22.32 @itemize @bullet @item Fixed problem when storing numbers in timestamps. @item Fix problem with timezones that have half hour offsets. @item @code{mysqlhotcopy} - fast on-line hot-backup utility for local MySQL databases. By Tim Bunce. @item New more secure @code{mysqlaccess}. Thanks to Steve Harvey for this. @item Fixed security problem in the protocol regarding password checking. @item Fixed problem that affected queries that did arithmetic on @code{GROUP} functions. @item Fixed a bug in the @code{ISAM} code when deleting rows on tables with packed indexes. @end itemize @node News-3.22.31, News-3.22.30, News-3.22.32, News-3.22.x @appendixsubsec Changes in release 3.22.31 @itemize @bullet @item A few small fixes for the Windows version. @end itemize @node News-3.22.30, News-3.22.29, News-3.22.31, News-3.22.x @appendixsubsec Changes in release 3.22.30 @itemize @bullet @item Fixed optimizer problem on @code{SELECT} when using many overlapping indexes. @item Disabled floating-point exceptions for FreeBSD to fix core dump when doing @code{SELECT floor(pow(2,63))}. @item Added print of default arguments options to all clients. @item Fixed critical problem with the @code{WITH GRANT OPTION} option. @item Fixed non-critical Y2K problem when writing short date to log files. @end itemize @node News-3.22.29, News-3.22.28, News-3.22.30, News-3.22.x @appendixsubsec Changes in release 3.22.29 @itemize @bullet @item Upgraded the configure and include files to match the latest 3.23 version. This should increase portability and make it easier to build shared libraries. @item Added latest patches to mit-pthreads for NetBSD. @item Fixed problem with timezones that are < GMT -11. @item Fixed a bug when deleting packed keys in NISAM. @item Fixed problem that could cause MySQL to touch freed memory when doing very complicated @code{GROUP BY} queries. @item Fixed core dump if you got a crashed table where an @code{ENUM} field value was too big. @item Added @code{mysqlshutdown.exe} and @code{mysqlwatch.exe} to the Windows distribution. @item Fixed problem when doing @code{ORDER BY} on a reference key. @item Fixed that @code{INSERT DELAYED} doesn't update timestamps that are given. @end itemize @node News-3.22.28, News-3.22.27, News-3.22.29, News-3.22.x @appendixsubsec Changes in release 3.22.28 @itemize @bullet @item Fixed problem with @code{LEFT JOIN} and @code{COUNT()} on a column which was declared @code{NULL} + and it had a @code{DEFAULT} value. @item Fixed core dump problem when using @code{CONCAT()} in a @code{WHERE} clause. @item Fixed problem with @code{AVG()} and @code{STD()} with @code{NULL} values. @end itemize @node News-3.22.27, News-3.22.26, News-3.22.28, News-3.22.x @appendixsubsec Changes in release 3.22.27 @itemize @bullet @item Fixed prototype in @file{my_ctype.h} when using other character sets. @item Some configure issues to fix problems with big file system detection. @item Fixed problem when sorting on big blob columns. @item @code{ROUND()} will now work on Windows. @end itemize @node News-3.22.26, News-3.22.25, News-3.22.27, News-3.22.x @appendixsubsec Changes in release 3.22.26 @itemize @bullet @item Fixed core dump with empty @code{BLOB/TEXT} column to @code{REVERSE()}. @item Extended @code{/*! */} with version numbers. @item Changed @code{SUBSTRING(text FROM pos)} to conform to ANSI SQL. (Before this construct returned the rightmost 'pos' characters). @item Fixed problem with @code{LOCK TABLES} combined with @code{DELETE FROM table} @item Fixed problem that INSERT ... SELECT didn't use SQL_BIG_TABLES. @item @code{SET SQL_LOW_PRIORITY_UPDATES=#} didn't work. @item Password wasn't updated correctly if privileges didn't change on: @code{GRANT ... IDENTIFIED BY} @item Fixed range optimizer bug in @code{SELECT * FROM table_name WHERE key_part1 >= const AND (key_part2 = const OR key_part2 = const)} @item Fixed bug in compression key handling in ISAM. @end itemize @node News-3.22.25, News-3.22.24, News-3.22.26, News-3.22.x @appendixsubsec Changes in release 3.22.25 @itemize @bullet @item Fixed some small problems with the installation. @end itemize @node News-3.22.24, News-3.22.23, News-3.22.25, News-3.22.x @appendixsubsec Changes in release 3.22.24 @itemize @bullet @item @code{DATA} is not a reserved word anymore. @item Fixed optimizer bug with tables with only one row. @item Fixed bug when using @code{LOCK TABLES table_name READ; FLUSH TABLES;} @item Applied some patches for HP-UX. @item @code{isamchk} should now work on Windows. @item Changed @file{configure} to not use big file handling on Linux as this crashes some RedHat 6.0 systems @end itemize @node News-3.22.23, News-3.22.22, News-3.22.24, News-3.22.x @appendixsubsec Changes in release 3.22.23 @itemize @bullet @item Upgraded to use Autoconf 2.13, Automake 1.4 and @code{libtool} 1.3.2. @item Better support for SCO in @code{configure}. @item Added option @code{--defaults-file=###} to option file handling to force use of only one specific option file. @item Extended @code{CREATE} syntax to ignore MySQL Version 3.23 keywords. @item Fixed deadlock problem when using @code{INSERT DELAYED} on a table locked with @code{LOCK TABLES}. @item Fixed deadlock problem when using @code{DROP TABLE} on a table that was locked by another thread. @item Add logging of @code{GRANT/REVOKE} commands in the update log. @item Fixed @code{isamchk} to detect a new error condition. @item Fixed bug in @code{NATURAL LEFT JOIN}. @end itemize @node News-3.22.22, News-3.22.21, News-3.22.23, News-3.22.x @appendixsubsec Changes in release 3.22.22 @itemize @bullet @item Fixed problem in the C API when you called @code{mysql_close()} directly after @code{mysql_init()}. @item Better client error message when you can't open socket. @item Fixed @code{delayed_insert_thread} counting when you couldn't create a new delayed_insert thread. @item Fixed bug in @code{CONCAT()} with many arguments. @item Added patches for DEC 3.2 and SCO. @item Fixed path-bug when installing MySQL as a service on NT. @item The MySQL-Windows version is now compiled with VC++ 6.0 instead of with VC++ 5.0. @item New installation setup for MySQL-Windows. @end itemize @node News-3.22.21, News-3.22.20, News-3.22.22, News-3.22.x @appendixsubsec Changes in release 3.22.21 @itemize @bullet @item Fixed problem with @code{DELETE FROM TABLE} when table was locked by another thread. @item Fixed bug in @code{LEFT JOIN} involving empty tables. @item Changed the @code{mysql.db} column from @code{char(32)} to @code{char(60)}. @item @code{MODIFY} and @code{DELAYED} are not reserved words anymore. @item Fixed a bug when storing days in a @code{TIME} column. @item Fixed a problem with @code{Host '...' is not allowed to connect to this MySQL server} after one had inserted a new MySQL user with a @code{GRANT} command. @item Changed to use @code{TCP_NODELAY} also on Linux (Should give faster TCP/IP connections). @end itemize @node News-3.22.20, News-3.22.19, News-3.22.21, News-3.22.x @appendixsubsec Changes in release 3.22.20 @itemize @bullet @item Fixed @code{STD()} for big tables when result should be 0. @item The update log didn't have newlines on some operating systems. @item @code{INSERT DELAYED} had some garbage at end in the update log. @end itemize @node News-3.22.19, News-3.22.18, News-3.22.20, News-3.22.x @appendixsubsec Changes in release 3.22.19 @itemize @bullet @item Fixed bug in @code{mysql_install_db} (from 3.22.17). @item Changed default key cache size to 8M. @item Fixed problem with queries that needed temporary tables with @code{BLOB} columns. @end itemize @node News-3.22.18, News-3.22.17, News-3.22.19, News-3.22.x @appendixsubsec Changes in release 3.22.18 @itemize @bullet @item Fixes a fatal problem in 3.22.17 on Linux; After @code{shutdown} all threads didn't die properly. @item Added option @code{-O flush_time=#} to @code{mysqld}. This is mostly useful on Windows and tells how often MySQL should close all unused tables and flush all updated tables to disk. @item Fixed problem that a @code{VARCHAR} column compared with @code{CHAR} column didn't use keys efficiently. @end itemize @node News-3.22.17, News-3.22.16, News-3.22.18, News-3.22.x @appendixsubsec Changes in release 3.22.17 @itemize @bullet @item Fixed a core dump problem when using @code{--log-update} and connecting without a default database. @item Fixed some @code{configure} and portability problems. @item Using @code{LEFT JOIN} on tables that had circular dependencies caused @code{mysqld} to hang forever. @end itemize @node News-3.22.16, News-3.22.15, News-3.22.17, News-3.22.x @appendixsubsec Changes in release 3.22.16 @itemize @bullet @item @code{mysqladmin processlist} could kill the server if a new user logged in. @item @code{DELETE FROM tbl_name WHERE key_column=col_name} didn't find any matching rows. Fixed. @item @code{DATE_ADD(column, ...)} didn't work. @item @code{INSERT DELAYED} could deadlock with status 'upgrading lock' @item Extended @code{ENCRYPT()} to take longer salt strings than 2 characters. @item @code{longlong2str} is now much faster than before. For @code{Intel x86} platforms, this function is written in optimized assembler. @item Added the @code{MODIFY} keyword to @code{ALTER TABLE}. @end itemize @node News-3.22.15, News-3.22.14, News-3.22.16, News-3.22.x @appendixsubsec Changes in release 3.22.15 @itemize @bullet @item @code{GRANT} used with @code{IDENTIFIED BY} didn't take effect until privileges were flushed. @item Name change of some variables in @code{SHOW STATUS}. @item Fixed problem with @code{ORDER BY} with 'only index' optimization when there were multiple key definitions for a used column. @item @code{DATE} and @code{DATETIME} columns are now up to 5 times faster than before. @item @code{INSERT DELAYED} can be used to let the client do other things while the server inserts rows into a table. @item @code{LEFT JOIN USING (col1,col2)} didn't work if one used it with tables from 2 different databases. @item @code{LOAD DATA LOCAL INFILE} didn't work in the Unix version because of a missing file. @item Fixed problems with @code{VARCHAR}/@code{BLOB} on very short rows (< 4 bytes); error 127 could occur when deleting rows. @item Updating @code{BLOB/TEXT} through formulas didn't work for short (< 256 char) strings. @item When you did a @code{GRANT} on a new host, @code{mysqld} could die on the first connect from this host. @item Fixed bug when one used @code{ORDER BY} on column name that was the same name as an alias. @item Added @code{BENCHMARK(loop_count,expression)} function to time expressions. @end itemize @node News-3.22.14, News-3.22.13, News-3.22.15, News-3.22.x @appendixsubsec Changes in release 3.22.14 @itemize @bullet @item Allow empty arguments to @code{mysqld} to make it easier to start from shell scripts. @item Setting a @code{TIMESTAMP} column to @code{NULL} didn't record the timestamp value in the update log. @item Fixed lock handler bug when one did @code{INSERT INTO TABLE ... SELECT ... GROUP BY}. @item Added a patch for @code{localtime_r()} on Windows so that it will not crash anymore if your date is > 2039, but instead will return a time of all zero. @item Names for user-defined functions are no longer case sensitive. @item Added escape of @code{^Z} (ASCII 26) to @code{\Z} as @code{^Z} doesn't work with pipes on Windows. @item @code{mysql_fix_privileges} adds a new column to the @code{mysql.func} to support aggregate UDF functions in future MySQL releases. @end itemize @node News-3.22.13, News-3.22.12, News-3.22.14, News-3.22.x @appendixsubsec Changes in release 3.22.13 @itemize @bullet @item Saving @code{NOW()}, @code{CURDATE()} or @code{CURTIME()} directly in a column didn't work. @item @code{SELECT COUNT(*) ... LEFT JOIN ...} didn't work with no @code{WHERE} part. @item Updated @file{config.guess} to allow MySQL to configure on UnixWare 7.0.x. @item Changed the implementation of @code{pthread_cond()} on the Windows version. @code{get_lock()} now correctly times out on Windows! @end itemize @node News-3.22.12, News-3.22.11, News-3.22.13, News-3.22.x @appendixsubsec Changes in release 3.22.12 @itemize @bullet @item Fixed problem when using @code{DATE_ADD()} and @code{DATE_SUB()} in a @code{WHERE} clause. @item You can now set the password for a user with the @code{GRANT ... TO user IDENTIFIED BY 'password'} syntax. @item Fixed bug in @code{GRANT} checking with @code{SELECT} on many tables. @item Added missing file @code{mysql_fix_privilege_tables} to the RPM distribution. This is not run by default because it relies on the client package. @item Added option @code{SQL_SMALL_RESULT} to @code{SELECT} to force use of fast temporary tables when you know that the result set will be small. @item Allow use of negative real numbers without a decimal point. @item Day number is now adjusted to maximum days in month if the resulting month after @code{DATE_ADD}/@code{DATE_SUB()} doesn't have enough days. @item Fix that @code{GRANT} compares columns in case-insensitive fashion. @item Fixed a bug in @file{sql_list.h} that made @code{ALTER TABLE} dump core in some contexts. @item The hostname in @code{user@@hostname} can now include @samp{.} and @samp{-} without quotes in the context of the @code{GRANT}, @code{REVOKE} and @code{SET PASSWORD FOR ...} statements. @item Fix for @code{isamchk} for tables which need big temporary files. @end itemize @node News-3.22.11, News-3.22.10, News-3.22.12, News-3.22.x @appendixsubsec Changes in release 3.22.11 @itemize @bullet @item @strong{IMPORTANT}: You must run the @code{mysql_fix_privilege_tables} script when you upgrade to this version! This is needed because of the new @code{GRANT} system. If you don't do this, you will get @code{Access denied} when you try to use @code{ALTER TABLE}, @code{CREATE INDEX} or @code{DROP INDEX}. @item @code{GRANT} to allow/deny users table and column access. @item Changed @code{USER()} to return @code{user@@host} @item Changed the syntax for how to set @code{PASSWORD} for another user. @item New command @code{FLUSH STATUS} that sets most status variables to zero. @item New status variables: @code{aborted_threads}, @code{aborted_connects}. @item New option variable: @code{connection_timeout}. @item Added support for Thai sorting (by Pruet Boonma @email{pruet@@ds90.intanon.nectec.or.th}). @item Slovak and japanese error messages. @item Configuration and portability fixes. @item Added option @code{SET SQL_WARNINGS=1} to get a warning count also for simple inserts. @item MySQL now uses @code{SIGTERM} instead of @code{SIGQUIT} with shutdown to work better on FreeBSD. @item Added option @code{\G} (print vertically) to @code{mysql}. @item @code{SELECT HIGH_PRIORITY ...} killed @code{mysqld}. @item @code{IS NULL} on a @code{AUTO_INCREMENT} column in a @code{LEFT JOIN} didn't work as expected. @item New function @code{MAKE_SET()}. @end itemize @node News-3.22.10, News-3.22.9, News-3.22.11, News-3.22.x @appendixsubsec Changes in release 3.22.10 @itemize @bullet @item @code{mysql_install_db} no longer starts the MySQL server! You should start @code{mysqld} with @code{safe_mysqld} after installing it! The MySQL RPM will, however, start the server as before. @item Added @code{--bootstrap} option to @code{mysqld} and recoded @code{mysql_install_db} to use it. This will make it easier to install MySQL with RPMs. @item Changed @code{+}, @code{-} (sign and minus), @code{*}, @code{/}, @code{%}, @code{ABS()} and @code{MOD()} to be @code{BIGINT} aware (64-bit safe). @item Fixed a bug in @code{ALTER TABLE} that caused @code{mysqld} to crash. @item MySQL now always reports the conflicting key values when a duplicate key entry occurs. (Before this was only reported for @code{INSERT}). @item New syntax: @code{INSERT INTO tbl_name SET col_name=value, col_name=value, ...} @item Most errors in the @file{.err} log are now prefixed with a time stamp. @item Added option @code{MYSQL_INIT_COMMAND} to @code{mysql_options()} to make a query on connect or reconnect. @item Added option @code{MYSQL_READ_DEFAULT_FILE} and @code{MYSQL_READ_DEFAULT_GROUP} to @code{mysql_options()} to read the following parameters from the MySQL option files: @code{port}, @code{socket}, @code{compress}, @code{password}, @code{pipe}, @code{timeout}, @code{user}, @code{init-command}, @code{host} and @code{database}. @item Added @code{maybe_null} to the UDF structure. @item Added option @code{IGNORE} to @code{INSERT} statements with many rows. @item Fixed some problems with sorting of the koi8 character sets; Users of koi8 @strong{MUST} run @code{isamchk -rq} on each table that has an index on a @code{CHAR} or @code{VARCHAR} column. @item New script @code{mysql_setpermission}, by Luuk de Boer, allows one to easily create new users with permissions for specific databases. @item Allow use of hexadecimal strings (0x...) when specifying a constant string (like in the column separators with @code{LOAD DATA INFILE}). @item Ported to OS/2 (thanks to Antony T. Curtis @email{antony.curtis@@olcs.net}). @item Added more variables to @code{SHOW STATUS} and changed format of output to be like @code{SHOW VARIABLES}. @item Added @code{extended-status} command to @code{mysqladmin} which will show the new status variables. @end itemize @node News-3.22.9, News-3.22.8, News-3.22.10, News-3.22.x @appendixsubsec Changes in release 3.22.9 @itemize @bullet @item @code{SET SQL_LOG_UPDATE=0} caused a lockup of the server. @item New SQL command: @code{FLUSH [ TABLES | HOSTS | LOGS | PRIVILEGES ] [, ...]} @item New SQL command: @code{KILL} @code{thread_id}. @item Added casts and changed include files to make MySQL easier to compile on AIX and DEC OSF1 4.x @item Fixed conversion problem when using @code{ALTER TABLE} from a @code{INT} to a short @code{CHAR()} column. @item Added @code{SELECT HIGH_PRIORITY}; This will get a lock for the @code{SELECT} even if there is a thread waiting for another @code{SELECT} to get a @code{WRITE LOCK}. @item Moved wild_compare to string class to be able to use @code{LIKE} on @code{BLOB}/@code{TEXT} columns with @code{\0}. @item Added @code{ESCAPE} option to @code{LIKE}. @item Added a lot more output to @code{mysqladmin debug}. @item You can now start @code{mysqld} on Windows with the @code{--flush} option. This will flush all tables to disk after each update. This makes things much safer on NT/Win98 but also @strong{MUCH} slower. @end itemize @node News-3.22.8, News-3.22.7, News-3.22.9, News-3.22.x @appendixsubsec Changes in release 3.22.8 @itemize @bullet @item Czech character sets should now work much better. You must also install @uref{http://www.mysql.com/Downloads/Patches/czech-3.22.8-patch}. This patch should also be installed if you are using a character set with uses @code{my_strcoll()}! The patch should always be safe to install (for any system), but as this patch changes ISAM internals it's not yet in the default distribution. @item @code{DATE_ADD()} and @code{DATE_SUB()} didn't work with group functions. @item @code{mysql} will now also try to reconnect on @code{USE DATABASE} commands. @item Fix problem with @code{ORDER BY} and @code{LEFT JOIN} and @code{const} tables. @item Fixed problem with @code{ORDER BY} if the first @code{ORDER BY} column was a key and the rest of the @code{ORDER BY} columns wasn't part of the key. @item Fixed a big problem with @code{OPTIMIZE TABLE}. @item MySQL clients on NT will now by default first try to connect with named pipes and after this with TCP/IP. @item Fixed a problem with @code{DROP TABLE} and @code{mysqladmin shutdown} on Windows (a fatal bug from 3.22.6). @item Fixed problems with @code{TIME columns} and negative strings. @item Added an extra thread signal loop on shutdown to avoid some error messages from the client. @item MySQL now uses the next available number as extension for the update log file. @item Added patches for UNIXWARE 7. @end itemize @node News-3.22.7, News-3.22.6, News-3.22.8, News-3.22.x @appendixsubsec Changes in release 3.22.7 @itemize @bullet @item Added @code{LIMIT} clause for the @code{DELETE} statement. @item You can now use the @code{/*! ... */} syntax to hide MySQL-specific keywords when you write portable code. MySQL will parse the code inside the comments as if the surrounding @code{/*!} and @code{*/} comment characters didn't exist. @item @code{OPTIMIZE TABLE tbl_name} can now be used to reclaim disk space after many deletes. Currently, this uses @code{ALTER TABLE} to regenerate the table, but in the future it will use an integrated @code{isamchk} for more speed. @item Upgraded @code{libtool} to get the configure more portable. @item Fixed slow @code{UPDATE} and @code{DELETE} operations when using @code{DATETIME} or @code{DATE} keys. @item Changed optimizer to make it better at deciding when to do a full join and when using keys. @item You can now use @code{mysqladmin proc} to display information about your own threads. Only users with the @strong{Process_priv} privilege can get information about all threads. @item Added handling of formats @code{YYMMDD}, @code{YYYYMMDD}, @code{YYMMDDHHMMSS} for numbers when using @code{DATETIME} and @code{TIMESTAMP} types. (Formerly these formats only worked with strings.) @item Added connect option @code{CLIENT_IGNORE_SPACE} to allow use of spaces after function names and before @samp{(} (Powerbuilder requires this). This will make all function names reserved words. @item Added the @code{--log-long-format} option to @code{mysqld} to enable timestamps and INSERT_ID's in the update log. @item Added @code{--where} option to @code{mysqldump} (patch by Jim Faucette). @item The lexical analyzer now uses ``perfect hashing'' for faster parsing of SQL statements. @end itemize @node News-3.22.6, News-3.22.5, News-3.22.7, News-3.22.x @appendixsubsec Changes in release 3.22.6 @itemize @bullet @item Faster @code{mysqldump}. @item For the @code{LOAD DATA INFILE} statement, you can now use the new @code{LOCAL} keyword to read the file from the client. @code{mysqlimport} will automatically use @code{LOCAL} when importing with the TCP/IP protocol. @item Fixed small optimize problem when updating keys. @item Changed makefiles to support shared libraries. @item MySQL-NT can now use named pipes, which means that you can now use MySQL-NT without having to install TCP/IP. @end itemize @node News-3.22.5, News-3.22.4, News-3.22.6, News-3.22.x @appendixsubsec Changes in release 3.22.5 @itemize @bullet @item All table lock handing is changed to avoid some very subtle deadlocks when using @code{DROP TABLE}, @code{ALTER TABLE}, @code{DELETE FROM TABLE} and @code{mysqladmin flush-tables} under heavy usage. Changed locking code to get better handling of locks of different types. @item Updated @code{DBI} to 1.00 and @code{DBD} to 1.2.0. @item Added a check that the error message file contains error messages suitable for the current version of @code{mysqld}. (To avoid errors if you accidentally try to use an old error message file.) @item All count structures in the client (@code{affected_rows()}, @code{insert_id()}, ...) are now of type @code{BIGINT} to allow 64-bit values to be used. This required a minor change in the MySQL protocol which should affect only old clients when using tables with @code{AUTO_INCREMENT} values > 16M. @item The return type of @code{mysql_fetch_lengths()} has changed from @code{uint *} to @code{ulong *}. This may give a warning for old clients but should work on most machines. @item Change @code{mysys} and @code{dbug} libraries to allocate all thread variables in one struct. This makes it easier to make a threaded @file{libmysql.dll} library. @item Use the result from @code{gethostname()} (instead of @code{uname()}) when constructing @file{.pid} file names. @item New better compressed server/client protocol. @item @code{COUNT()}, @code{STD()} and @code{AVG()} are extended to handle more than 4G rows. @item You can now store values in the range @code{-838:59:59} <= x <= @code{838:59:59} in a @code{TIME} column. @item @strong{WARNING: INCOMPATIBLE CHANGE!!} If you set a @code{TIME} column to too short a value, MySQL now assumes the value is given as: @code{[[[D ]HH:]MM:]SS} instead of @code{HH[:MM[:SS]]}. @item @code{TIME_TO_SEC()} and @code{SEC_TO_TIME()} can now handle negative times and hours up to 32767. @item Added new option @code{SET OPTION SQL_LOG_UPDATE=@{0|1@}} to allow users with the @strong{process} privilege to bypass the update log. (Modified patch from Sergey A Mukhin @email{violet@@rosnet.net}.) @item Fixed fatal bug in @code{LPAD()}. @item Initialize line buffer in @file{mysql.cc} to make @code{BLOB} reading from pipes safer. @item Added @code{-O max_connect_errors=#} option to @code{mysqld}. Connect errors are now reset for each correct connection. @item Increased the default value of @code{max_allowed_packet} to @code{1M} in @code{mysqld}. @item Added @code{--low-priority-updates} option to @code{mysqld}, to give table-modifying operations (@code{INSERT}, @code{REPLACE}, @code{UPDATE}, @code{DELETE}) lower priority than retrievals. You can now use @code{@{INSERT | REPLACE | UPDATE | DELETE@} LOW_PRIORITY ...} You can also use @code{SET OPTION SQL_LOW_PRIORITY_UPDATES=@{0|1@}} to change the priority for one thread. One side effect is that @code{LOW_PRIORITY} is now a reserved word. :( @item Add support for @code{INSERT INTO table ... VALUES(...),(...),(...)}, to allow inserting multiple rows with a single statement. @item @code{INSERT INTO tbl_name} is now also cached when used with @code{LOCK TABLES}. (Previously only @code{INSERT ... SELECT} and @code{LOAD DATA INFILE} were cached.) @item Allow @code{GROUP BY} functions with @code{HAVING}: @example mysql> SELECT col FROM table GROUP BY col HAVING COUNT(*)>0; @end example @item @code{mysqld} will now ignore trailing @samp{;} characters in queries. This is to make it easier to migrate from some other SQL servers that require the trailing @samp{;}. @item Fix for corrupted fixed-format output generated by @code{SELECT INTO OUTFILE}. @item @strong{WARNING: INCOMPATIBLE CHANGE!!} Added Oracle @code{GREATEST()} and @code{LEAST()} functions. You must now use these instead of the @code{MAX()} and @code{MIN()} functions to get the largest/smallest value from a list of values. These can now handle @code{REAL}, @code{BIGINT} and string (@code{CHAR} or @code{VARCHAR}) values. @item @strong{WARNING: INCOMPATIBLE CHANGE!!} @code{DAYOFWEEK()} had offset 0 for Sunday. Changed the offset to 1. @item Give an error for queries that mix @code{GROUP BY} columns and fields when there is no @code{GROUP BY} specification. @item Added @code{--vertical} option to @code{mysql}, for printing results in vertical mode. @item Index-only optimization; some queries are now resolved using only indexes. Until MySQL 4.0, this works only for numeric columns. @xref{MySQL indexes, , MySQL indexes}. @item Lots of new benchmarks. @item A new C API chapter and lots of other improvements in the manual. @end itemize @node News-3.22.4, News-3.22.3, News-3.22.5, News-3.22.x @appendixsubsec Changes in release 3.22.4 @itemize @bullet @item Added @code{--tmpdir} option to @code{mysqld}, for specifying the location of the temporary file directory. @item MySQL now automatically changes a query from an ODBC client: @example SELECT ... FROM table WHERE auto_increment_column IS NULL @end example to: @example SELECT ... FROM table WHERE auto_increment_column == LAST_INSERT_ID() @end example This allows some ODBC programs (Delphi, Access) to retrieve the newly inserted row to fetch the @code{AUTO_INCREMENT} id. @item @code{DROP TABLE} now waits for all users to free a table before deleting it. @item Fixed small memory leak in the new connect protocol. @item New functions @code{BIN()}, @code{OCT()}, @code{HEX()} and @code{CONV()} for converting between different number bases. @item Added function @code{SUBSTRING()} with 2 arguments. @item If you created a table with a record length smaller than 5, you couldn't delete rows from the table. @item Added optimization to remove const reference tables from @code{ORDER BY} and @code{GROUP BY}. @item @code{mysqld} now automatically disables system locking on Linux and Windows, and for systems that use MIT-pthreads. You can force the use of locking with the @code{--enable-locking} option. @item Added @code{--console} option to @code{mysqld}, to force a console window (for error messages) when using Windows. @item Fixed table locks for Windows. @item Allow @samp{$} in identifiers. @item Changed name of user-specific configuration file from @file{my.cnf} to @file{.my.cnf} (Unix only). @item Added @code{DATE_ADD()} and @code{DATE_SUB()} functions. @end itemize @node News-3.22.3, News-3.22.2, News-3.22.4, News-3.22.x @appendixsubsec Changes in release 3.22.3 @itemize @bullet @item Fixed a lock problem (bug in MySQL Version 3.22.1) when closing temporary tables. @item Added missing @code{mysql_ping()} to the client library. @item Added @code{--compress} option to all MySQL clients. @item Changed @code{byte} to @code{char} in @file{mysql.h} and @file{mysql_com.h}. @end itemize @node News-3.22.2, News-3.22.1, News-3.22.3, News-3.22.x @appendixsubsec Changes in release 3.22.2 @itemize @bullet @item Searching on multiple constant keys that matched more than 30% of the rows didn't always use the best possible key. @item New functions @code{<<}, @code{>>}, @code{RPAD()} and @code{LPAD()}. @item You can now save default options (like passwords) in a configuration file (@file{my.cnf}). @item Lots of small changes to get @code{ORDER BY} to work when no records are found when using fields that are not in @code{GROUP BY} (MySQL extension). @item Added @code{--chroot} option to @code{mysqld}, to start @code{mysqld} in a chroot environment (by Nikki Chumakov @email{nikkic@@cityline.ru}). @item Trailing spaces are now ignored when comparing case-sensitive strings; this should fix some problems with ODBC and flag 512! @item Fixed a core-dump bug in the range optimizer. @item Added @code{--one-thread} option to @code{mysqld}, for debugging with LinuxThreads (or @code{glibc}). (This replaces the @code{-T32} flag) @item Added @code{DROP TABLE IF EXISTS} to prevent an error from occurring if the table doesn't exist. @item @code{IF} and @code{EXISTS} are now reserved words (they would have to be sooner or later). @item Added lots of new options to @code{mysqldump}. @item Server error messages are now in @file{mysqld_error.h}. @item The server/client protocol now supports compression. @item All bug fixes from MySQL Version 3.21.32. @end itemize @node News-3.22.1, News-3.22.0, News-3.22.2, News-3.22.x @appendixsubsec Changes in release 3.22.1 @itemize @bullet @item Added new C API function @code{mysql_ping()}. @item Added new API functions @code{mysql_init()} and @code{mysql_options()}. You now MUST call @code{mysql_init()} before you call @code{mysql_real_connect()}. You don't have to call @code{mysql_init()} if you only use @code{mysql_connect()}. @item Added @code{mysql_options(...,MYSQL_OPT_CONNECT_TIMEOUT,...)} so you can set a timeout for connecting to a server. @item Added @code{--timeout} option to @code{mysqladmin}, as a test of @code{mysql_options()}. @item Added @code{AFTER column} and @code{FIRST} options to @code{ALTER TABLE ... ADD columns}. This makes it possible to add a new column at some specific location within a row in an existing table. @item @code{WEEK()} now takes an optional argument to allow handling of weeks when the week starts on Monday (some European countries). By default, @code{WEEK()} assumes the week starts on Sunday. @item @code{TIME} columns weren't stored properly (bug in MySQL Version 3.22.0). @item @code{UPDATE} now returns information about how many rows were matched and updated, and how many ``warnings'' occurred when doing the update. @item Fixed incorrect result from @code{FORMAT(-100,2)}. @item @code{ENUM} and @code{SET} columns were compared in binary (case-sensitive) fashion; changed to be case insensitive. @end itemize @node News-3.22.0, , News-3.22.1, News-3.22.x @appendixsubsec Changes in release 3.22.0 @itemize @bullet @item New (backward compatible) connect protocol that allows you to specify the database to use when connecting, to get much faster connections to a specific database. The @code{mysql_real_connect()} call is changed to: @example mysql_real_connect(MYSQL *mysql, const char *host, const char *user, const char *passwd, const char *db, uint port, const char *unix_socket, uint client_flag) @end example @item Each connection is handled by its own thread, rather than by the master @code{accept()} thread. This fixes permanently the telnet bug that was a topic on the mail list some time ago. @item All TCP/IP connections are now checked with backward resolution of the hostname to get better security. @code{mysqld} now has a local hostname resolver cache so connections should actually be faster than before, even with this feature. @item A site automatically will be blocked from future connections if someone repeatedly connects with an ``improper header'' (like when one uses telnet). @item You can now refer to tables in different databases with references of the form @code{tbl_name@@db_name} or @code{db_name.tbl_name}. This makes it possible to give a user read access to some tables and write access to others simply by keeping them in different databases! @item Added @code{--user} option to @code{mysqld}, to allow it to run as another Unix user (if it is started as the Unix @code{root} user). @item Added caching of users and access rights (for faster access rights checking) @item Normal users (not anonymous ones) can change their password with @code{mysqladmin password 'new_password'}. This uses encrypted passwords that are not logged in the normal MySQL log! @item All important string functions are now coded in assembler for x86 Linux machines. This gives a speedup of 10% in many cases. @item For tables that have many columns, the column names are now hashed for much faster column name lookup (this will speed up some benchmark tests a lot!) @item Some benchmarks are changed to get better individual timing. (Some loops were so short that a specific test took < 2 seconds. The loops have been changed to take about 20 seconds to make it easier to compare different databases. A test that took 1-2 seconds before now takes 11-24 seconds, which is much better) @item Re-arranged @code{SELECT} code to handle some very specific queries involving group functions (like @code{COUNT(*)}) without a @code{GROUP BY} but with @code{HAVING}. The following now works: @example mysql> SELECT count(*) as C FROM table HAVING C > 1; @end example @item Changed the protocol for field functions to be faster and avoid some calls to @code{malloc()}. @item Added @code{-T32} option to @code{mysqld}, for running all queries under the main thread. This makes it possible to debug @code{mysqld} under Linux with @code{gdb}! @item Added optimization of @code{not_null_column IS NULL} (needed for some Access queries). @item Allow @code{STRAIGHT_JOIN} to be used between two tables to force the optimizer to join them in a specific order. @item String functions now return @code{VARCHAR} rather than @code{CHAR} and the column type is now @code{VARCHAR} for fields saved as @code{VARCHAR}. This should make the @strong{MyODBC} driver better, but may break some old MySQL clients that don't handle @code{FIELD_TYPE_VARCHAR} the same way as @code{FIELD_TYPE_CHAR}. @item @code{CREATE INDEX} and @code{DROP INDEX} are now implemented through @code{ALTER TABLE}. @code{CREATE TABLE} is still the recommended (fast) way to create indexes. @item Added @code{--set-variable} option @code{wait_timeout} to @code{mysqld}. @item Added time column to @code{mysqladmin processlist} to show how long a query has taken or how long a thread has slept. @item Added lots of new variables to @code{show variables} and some new to @code{show status}. @item Added new type @code{YEAR}. @code{YEAR} is stored in 1 byte with allowable values of 0, and 1901 to 2155. @item Added new @code{DATE} type that is stored in 3 bytes rather than 4 bytes. All new tables are created with the new date type if you don't use the @code{--old-protocol} option to @code{mysqld}. @item Fixed bug in record caches; for some queries, you could get @code{Error from table handler: #} on some operating systems. @item Added @code{--enable-assembler} option to @code{configure}, for x86 machines (tested on Linux + @code{gcc}). This will enable assembler functions for the most important string functions for more speed! @end itemize @node News-3.21.x, News-3.20.x, News-3.22.x, News @appendixsec Changes in release 3.21.x @cindex changes, version 3.21 Version 3.21 is quite old now, and should be avoided if possible. This information is kept here for historical purposes only. @menu * News-3.21.33:: Changes in release 3.21.33 * News-3.21.32:: Changes in release 3.21.32 * News-3.21.31:: Changes in release 3.21.31 * News-3.21.30:: Changes in release 3.21.30 * News-3.21.29:: Changes in release 3.21.29 * News-3.21.28:: Changes in release 3.21.28 * News-3.21.27:: Changes in release 3.21.27 * News-3.21.26:: Changes in release 3.21.26 * News-3.21.25:: Changes in release 3.21.25 * News-3.21.24:: Changes in release 3.21.24 * News-3.21.23:: Changes in release 3.21.23 * News-3.21.22:: Changes in release 3.21.22 * News-3.21.21a:: Changes in release 3.21.21a * News-3.21.21:: Changes in release 3.21.21 * News-3.21.20:: Changes in release 3.21.20 * News-3.21.19:: Changes in release 3.21.19 * News-3.21.18:: Changes in release 3.21.18 * News-3.21.17:: Changes in release 3.21.17 * News-3.21.16:: Changes in release 3.21.16 * News-3.21.15:: Changes in release 3.21.15 * News-3.21.14b:: Changes in release 3.21.14b * News-3.21.14a:: Changes in release 3.21.14a * News-3.21.13:: Changes in release 3.21.13 * News-3.21.12:: Changes in release 3.21.12 * News-3.21.11:: Changes in release 3.21.11 * News-3.21.10:: Changes in release 3.21.10 * News-3.21.9:: Changes in release 3.21.9 * News-3.21.8:: Changes in release 3.21.8 * News-3.21.7:: Changes in release 3.21.7 * News-3.21.6:: Changes in release 3.21.6 * News-3.21.5:: Changes in release 3.21.5 * News-3.21.4:: Changes in release 3.21.4 * News-3.21.3:: Changes in release 3.21.3 * News-3.21.2:: Changes in release 3.21.2 * News-3.21.0:: Changes in release 3.21.0 @end menu @node News-3.21.33, News-3.21.32, News-3.21.x, News-3.21.x @appendixsubsec Changes in release 3.21.33 @itemize @bullet @item Fixed problem when sending @code{SIGHUP} to @code{mysqld}; @code{mysqld} core dumped when starting from boot on some systems. @item Fixed problem with losing a little memory for some connections. @item @code{DELETE FROM tbl_name} without a @code{WHERE} condition is now done the long way when you use @code{LOCK TABLES} or if the table is in use, to avoid race conditions. @item @code{INSERT INTO TABLE (timestamp_column) VALUES (NULL);} didn't set timestamp. @end itemize @node News-3.21.32, News-3.21.31, News-3.21.33, News-3.21.x @appendixsubsec Changes in release 3.21.32 @itemize @bullet @item Fixed some possible race conditions when doing many reopen/close on the same tables under heavy load! This can happen if you execute @code{mysqladmin refresh} often. This could in some very rare cases corrupt the header of the index file and cause error 126 or 138. @item Fixed fatal bug in @code{refresh()} when running with the @code{--skip-locking} option. There was a ``very small'' time gap after a @code{mysqladmin refresh} when a table could be corrupted if one thread updated a table while another thread did @code{mysqladmin refresh} and another thread started a new update ont the same table before the first thread had finished. A refresh (or @code{--flush-tables}) will now not return until all used tables are closed! @item @code{SELECT DISTINCT} with a @code{WHERE} clause that didn't match any rows returned a row in some contexts (bug only in 3.21.31). @item @code{GROUP BY} + @code{ORDER BY} returned one empty row when no rows where found. @item Fixed a bug in the range optimizer that wrote @code{Use_count: Wrong count for ...} in the error log file. @end itemize @node News-3.21.31, News-3.21.30, News-3.21.32, News-3.21.x @appendixsubsec Changes in release 3.21.31 @itemize @bullet @item Fixed a sign extension problem for the @code{TINYINT} type on Irix. @item Fixed problem with @code{LEFT("constant_string",function)}. @item Fixed problem with @code{FIND_IN_SET()}. @item @code{LEFT JOIN} core dumped if the second table is used with a constant @code{WHERE/ON} expression that uniquely identifies one record. @item Fixed problems with @code{DATE_FORMAT()} and incorrect dates. @code{DATE_FORMAT()} now ignores @code{'%'} to make it possible to extend it more easily in the future. @end itemize @node News-3.21.30, News-3.21.29, News-3.21.31, News-3.21.x @appendixsubsec Changes in release 3.21.30 @itemize @bullet @item @code{mysql} now returns an exit code > 0 if the query returned an error. @item Saving of command line history to file in @code{mysql} client. By Tommy Larsen @email{tommy@@mix.hive.no}. @item Fixed problem with empty lines that were ignored in @file{mysql.cc}. @item Save the pid of the signal handler thread in the pid file instead of the pid of the main thread. @item Added patch by @email{tommy@@valley.ne.jp} to support Japanese characters SJIS and UJIS. @item Changed @code{safe_mysqld} to redirect startup messages to @code{'hostname'.err} instead of @code{'hostname'.log} to reclaim file space on @code{mysqladmin refresh}. @item @code{ENUM} always had the first entry as default value. @item @code{ALTER TABLE} wrote two entries to the update log. @item @code{sql_acc()} now closes the @code{mysql} grant tables after a reload to save table space and memory. @item Changed @code{LOAD DATA} to use less memory with tables and @code{BLOB} columns. @item Sorting on a function which made a division / 0 produced a wrong set in some cases. @item Fixed @code{SELECT} problem with @code{LEFT()} when using the czech character set. @item Fixed problem in @code{isamchk}; it couldn't repair a packed table in a very unusual case. @item @code{SELECT} statements with @code{&} or @code{|} (bit functions) failed on columns with @code{NULL} values. @item When comparing a field = field, where one of the fields was a part key, only the length of the part key was compared. @end itemize @node News-3.21.29, News-3.21.28, News-3.21.30, News-3.21.x @appendixsubsec Changes in release 3.21.29 @itemize @bullet @item @code{LOCK TABLES} + @code{DELETE from tbl_name} never removed locks properly. @item Fixed problem when grouping on an @code{OR} function. @item Fixed permission problem with @code{umask()} and creating new databases. @item Fixed permission problem on result file with @code{SELECT ... INTO OUTFILE ...} @item Fixed problem in range optimizer (core dump) for a very complex query. @item Fixed problem when using @code{MIN(integer)} or @code{MAX(integer)} in @code{GROUP BY}. @item Fixed bug on Alpha when using integer keys. (Other keys worked on Alpha). @item Fixed bug in @code{WEEK("XXXX-xx-01")}. @end itemize @node News-3.21.28, News-3.21.27, News-3.21.29, News-3.21.x @appendixsubsec Changes in release 3.21.28 @itemize @bullet @item Fixed socket permission (clients couldn't connect to Unix socket on Linux). @item Fixed bug in record caches; for some queries, you could get @code{Error from table handler: #} on some operating systems. @end itemize @node News-3.21.27, News-3.21.26, News-3.21.28, News-3.21.x @appendixsubsec Changes in release 3.21.27 @itemize @bullet @item Added user level lock functions @code{GET_LOCK(string,timeout)}, @code{RELEASE_LOCK(string)}. @item Added @code{opened_tables} to @code{show status}. @item Changed connect timeout to 3 seconds to make it somewhat harder for crackers to kill @code{mysqld} through telnet + TCP/IP. @item Fixed bug in range optimizer when using @code{WHERE key_part_1 >= something AND key_part_2 <= something_else}. @item Changed @code{configure} for detection of FreeBSD 3.0 9803xx and above @item @code{WHERE} with string_column_key = constant_string didn't always find all rows if the column had many values differing only with characters of the same sort value (like e and @'e). @item Strings keys looked up with 'ref' were not compared in case-sensitive fashion. @item Added @code{umask()} to make log files non-readable for normal users. @item Ignore users with old (8-byte) password on startup if not using @code{--old-protocol} option to @code{mysqld}. @item @code{SELECT} which matched all key fields returned the values in the case of the matched values, not of the found values. (Minor problem.) @end itemize @node News-3.21.26, News-3.21.25, News-3.21.27, News-3.21.x @appendixsubsec Changes in release 3.21.26 @itemize @bullet @item @code{FROM_DAYS(0)} now returns "0000-00-00". @item In @code{DATE_FORMAT()}, PM and AM were swapped for hours 00 and 12. @item Extended the default maximum key size to 256. @item Fixed bug when using @code{BLOB}/@code{TEXT} in @code{GROUP BY} with many tables. @item An @code{ENUM} field that is not declared @code{NOT NULL} has @code{NULL} as the default value. (Previously, the default value was the first enumeration value.) @item Fixed bug in the join optimizer code when using many part keys on the same key: @code{INDEX (Organization,Surname(35),Initials(35))}. @item Added some tests to the table order optimizer to get some cases with @code{SELECT ... FROM many_tables} much faster. @item Added a retry loop around @code{accept()} to possibly fix some problems on some Linux machines. @end itemize @node News-3.21.25, News-3.21.24, News-3.21.26, News-3.21.x @appendixsubsec Changes in release 3.21.25 @itemize @bullet @item Changed @code{typedef 'string'} to @code{typedef 'my_string'} for better portability. @item You can now kill threads that are waiting on a disk-full condition. @item Fixed some problems with UDF functions. @item Added long options to @code{isamchk}. Try @code{isamchk --help}. @item Fixed a bug when using 8 bytes long (alpha); @code{filesort()} didn't work. Affects @code{DISTINCT}, @code{ORDER BY} and @code{GROUP BY} on 64-bit processors. @end itemize @node News-3.21.24, News-3.21.23, News-3.21.25, News-3.21.x @appendixsubsec Changes in release 3.21.24 @itemize @bullet @item Dynamic loadable functions. Based on source from Alexis Mikhailov. @item You couldn't delete from a table if no one had done a @code{SELECT} on the table. @item Fixed problem with range optimizer with many @code{OR} operators on key parts inside each other. @item Recoded @code{MIN()} and @code{MAX()} to work properly with strings and @code{HAVING}. @item Changed default umask value for new files from @code{0664} to @code{0660}. @item Fixed problem with @code{LEFT JOIN} and constant expressions in the @code{ON} part. @item Added Italian error messages from @email{brenno@@dewinter.com}. @item @code{configure} now works better on OSF1 (tested on 4.0D). @item Added hooks to allow @code{LIKE} optimization with international character support. @item Upgraded @code{DBI} to 0.93. @end itemize @node News-3.21.23, News-3.21.22, News-3.21.24, News-3.21.x @appendixsubsec Changes in release 3.21.23 @itemize @bullet @item The following symbols are now reserved words: @code{TIME}, @code{DATE}, @code{TIMESTAMP}, @code{TEXT}, @code{BIT}, @code{ENUM}, @code{NO}, @code{ACTION}, @code{CHECK}, @code{YEAR}, @code{MONTH}, @code{DAY}, @code{HOUR}, @code{MINUTE}, @code{SECOND}, @code{STATUS}, @code{VARIABLES}. @item Setting a @code{TIMESTAMP} to @code{NULL} in @code{LOAD DATA INFILE ...} didn't set the current time for the @code{TIMESTAMP}. @item Fix @code{BETWEEN} to recognize binary strings. Now @code{BETWEEN} is case sensitive. @item Added @code{--skip-thread-priority} option to @code{mysqld}, for systems where @code{mysqld}'s thread scheduling doesn't work properly (BSDI 3.1). @item Added ODBC functions @code{DAYNAME()} and @code{MONTHNAME()}. @item Added function @code{TIME_FORMAT()}. This works like @code{DATE_FORMAT()}, but takes a time string (@code{'HH:MM:DD'}) as argument. @item Fixed unlikely(?) key optimizer bug when using @code{OR}s of key parts inside @code{AND}s. @item Added command @code{variables} to @code{mysqladmin}. @item A lot of small changes to the binary releases. @item Fixed a bug in the new protocol from MySQL Version 3.21.20. @item Changed @code{ALTER TABLE} to work with Windows (Windows can't rename open files). Also fixed a couple of small bugs in the Windows version. @item All standard MySQL clients are now ported to MySQL-Windows. @item MySQL can now be started as a service on NT. @end itemize @node News-3.21.22, News-3.21.21a, News-3.21.23, News-3.21.x @appendixsubsec Changes in release 3.21.22 @itemize @bullet @item Starting with this version, all MySQL distributions will be configured, compiled and tested with @code{crash-me} and the benchmarks on the following platforms: SunOS 5.6 sun4u, SunOS 5.5.1 sun4u, SunOS 4.14 sun4c, SunOS 5.6 i86pc, Irix 6.3 mips5k, HP-UX 10.20 hppa, AIX 4.2.1 ppc, OSF1 V4.0 alpha, FreeBSD 2.2.2 i86pc and BSDI 3.1 i386. @item Fix @code{COUNT(*)} problems when the @code{WHERE} clause didn't match any records. (Bug from 3.21.17.) @item Removed that @code{NULL = NULL} is true. Now you must use @code{IS NULL} or @code{IS NOT NULL} to test whether or not a value is @code{NULL}. (This is according to ANSI SQL but may break old applications that are ported from @code{mSQL}.) You can get the old behavior by compiling with @code{-DmSQL_COMPLIANT}. @item Fixed bug that core dumped when using many @code{LEFT OUTER JOIN} clauses. @item Fixed bug in @code{ORDER BY} on string formula with possible @code{NULL} values. @item Fixed problem in range optimizer when using <= on sub index. @item Added functions @code{DAYOFYEAR()}, @code{DAYOFMONTH()}, @code{MONTH()}, @code{YEAR()}, @code{WEEK()}, @code{QUARTER()}, @code{HOUR()}, @code{MINUTE()}, @code{SECOND()} and @code{FIND_IN_SET()}. @item Added command @code{SHOW VARIABLES}. @item Added support of ``long constant strings'' from ANSI SQL: @example mysql> SELECT 'first ' 'second'; -> 'first second' @end example @item Upgraded mSQL-Mysql-modules to 1.1825. @item Upgraded @code{mysqlaccess} to 2.02. @item Fixed problem with Russian character set and @code{LIKE}. @item Ported to OpenBSD 2.1. @item New Dutch error messages. @end itemize @node News-3.21.21a, News-3.21.21, News-3.21.22, News-3.21.x @appendixsubsec Changes in release 3.21.21a @itemize @bullet @item Configure changes for some operating systems. @end itemize @node News-3.21.21, News-3.21.20, News-3.21.21a, News-3.21.x @appendixsubsec Changes in release 3.21.21 @itemize @bullet @item Fixed optimizer bug when using @code{WHERE data_field = date_field2 AND date_field2 = constant}. @item Added command @code{SHOW STATUS}. @item Removed @file{manual.ps} from the source distribution to make it smaller. @end itemize @node News-3.21.20, News-3.21.19, News-3.21.21, News-3.21.x @appendixsubsec Changes in release 3.21.20 @itemize @bullet @item Changed the maximum table name and column name lengths from 32 to 64. @item Aliases can now be of ``any'' length. @item Fixed @code{mysqladmin stat} to return the right number of queries. @item Changed protocol (downward compatible) to mark if a column has the @code{AUTO_INCREMENT} attribute or is a @code{TIMESTAMP}. This is needed for the new Java driver. @item Added Hebrew sorting order by Zeev Suraski. @item Solaris 2.6: Fixed @code{configure} bugs and increased maximum table size from 2G to 4G. @end itemize @node News-3.21.19, News-3.21.18, News-3.21.20, News-3.21.x @appendixsubsec Changes in release 3.21.19 @itemize @bullet @item Upgraded @code{DBD} to 1823. This version implements @code{mysql_use_result} in @code{DBD-Mysql}. @item Benchmarks updated for empress (by Luuk). @item Fixed a case of slow range searching. @item Configure fixes (@file{Docs} directory). @item Added function @code{REVERSE()} (by Zeev Suraski). @end itemize @node News-3.21.18, News-3.21.17, News-3.21.19, News-3.21.x @appendixsubsec Changes in release 3.21.18 @itemize @bullet @item Issue error message if client C functions are called in wrong order. @item Added automatic reconnect to the @file{libmysql.c} library. If a write command fails, an automatic reconnect is done. @item Small sort sets no longer use temporary files. @item Upgraded @code{DBI} to 0.91. @item Fixed a couple of problems with @code{LEFT OUTER JOIN}. @item Added @code{CROSS JOIN} syntax. @code{CROSS} is now a reserved word. @item Recoded @code{yacc}/@code{bison} stack allocation to be even safer and to allow MySQL to handle even bigger expressions. @item Fixed a couple of problems with the update log. @item @code{ORDER BY} was slow when used with key ranges. @end itemize @node News-3.21.17, News-3.21.16, News-3.21.18, News-3.21.x @appendixsubsec Changes in release 3.21.17 @itemize @bullet @item Changed documentation string of @code{--with-unix-socket-path} to avoid confusion. @item Added ODBC and ANSI SQL style @code{LEFT OUTER JOIN}. @item The following are new reserved words: @code{LEFT}, @code{NATURAL}, @code{USING}. @item The client library now uses the value of the environment variable @code{MYSQL_HOST} as the default host if it's defined. @item @code{SELECT col_name, SUM(expr)} now returns @code{NULL} for @code{col_name} when there are matching rows. @item Fixed problem with comparing binary strings and @code{BLOB}s with ASCII characters over 127. @item Fixed lock problem: when freeing a read lock on a table with multiple read locks, a thread waiting for a write lock would have been given the lock. This shouldn't affect data integrity, but could possibly make @code{mysqld} restart if one thread was reading data that another thread modified. @item @code{LIMIT offset,count} didn't work in @code{INSERT ... SELECT}. @item Optimized key block caching. This will be quicker than the old algorithm when using bigger key caches. @end itemize @node News-3.21.16, News-3.21.15, News-3.21.17, News-3.21.x @appendixsubsec Changes in release 3.21.16 @itemize @bullet @item Added ODBC 2.0 & 3.0 functions @code{POWER()}, @code{SPACE()}, @code{COT()}, @code{DEGREES()}, @code{RADIANS()}, @code{ROUND(2 arg)} and @code{TRUNCATE()}. @item @strong{WARNING: INCOMPATIBLE CHANGE!!} @code{LOCATE()} parameters were swapped according to ODBC standard. Fixed. @item Added function @code{TIME_TO_SEC()}. @item In some cases, default values were not used for @code{NOT NULL} fields. @item Timestamp wasn't always updated properly in @code{UPDATE SET ...} statements. @item Allow empty strings as default values for @code{BLOB} and @code{TEXT}, to be compatible with @code{mysqldump}. @end itemize @node News-3.21.15, News-3.21.14b, News-3.21.16, News-3.21.x @appendixsubsec Changes in release 3.21.15 @itemize @bullet @item @strong{WARNING: INCOMPATIBLE CHANGE!!} @code{mysqlperl} is now from Msql-Mysql-modules. This means that @code{connect()} now takes @code{host}, @code{database}, @code{user}, @code{password} arguments! The old version took @code{host}, @code{database}, @code{password}, @code{user}. @item Allow @code{DATE '1997-01-01'}, @code{TIME '12:10:10'} and @code{TIMESTAMP '1997-01-01 12:10:10'} formats required by ANSI SQL. @strong{WARNING: INCOMPATIBLE CHANGE!!} This has the unfortunate side-effect that you no longer can have columns named @code{DATE}, @code{TIME} or @code{TIMESTAMP}. :( Old columns can still be accessed through @code{tablename.columnname}!) @item Changed Makefiles to hopefully work better with BSD systems. Also, @file{manual.dvi} is now included in the distribution to avoid having stupid @code{make} programs trying to rebuild it. @item @code{readline} library upgraded to version 2.1. @item A new sortorder german-1. That is a normal ISO-Latin1 with a german sort order. @item Perl @code{DBI}/@code{DBD} is now included in the distribution. @code{DBI} is now the recommended way to connect to MySQL from Perl. @item New portable benchmark suite with @code{DBD}, with test results from @code{mSQL} 2.0.3, MySQL, PostgreSQL 6.2.1 and Solid server 2.2. @item @code{crash-me} is now included with the benchmarks; This is a Perl program designed to find as many limits as possible in a SQL server. Tested with @code{mSQL}, PostgreSQL, Solid and MySQL. @item Fixed bug in range-optimizer that crashed MySQL on some queries. @item Table and column name completion for @code{mysql} command line tool, by Zeev Suraski and Andi Gutmans. @item Added new command @code{REPLACE} that works like @code{INSERT} but replaces conflicting records with the new record. @code{REPLACE INTO TABLE ... SELECT ...} works also. @item Added new commands @code{CREATE DATABASE db_name} and @code{DROP DATABASE db_name}. @item Added @code{RENAME} option to @code{ALTER TABLE}: @code{ALTER TABLE name RENAME TO new_name}. @item @code{make_binary_distribution} now includes @file{libgcc.a} in @file{libmysqlclient.a}. This should make linking work for people who don't have @code{gcc}. @item Changed @code{net_write()} to @code{my_net_write()} because of a name conflict with Sybase. @item @cindex ODBC compatibility @cindex compatibility, with ODBC New function @code{DAYOFWEEK()} compatible with ODBC. @item Stack checking and @code{bison} memory overrun checking to make MySQL safer with weird queries. @end itemize @node News-3.21.14b, News-3.21.14a, News-3.21.15, News-3.21.x @appendixsubsec Changes in release 3.21.14b @itemize @bullet @item Fixed a couple of small @code{configure} problems on some platforms. @end itemize @node News-3.21.14a, News-3.21.13, News-3.21.14b, News-3.21.x @appendixsubsec Changes in release 3.21.14a @itemize @bullet @item Ported to SCO Openserver 5.0.4 with FSU Pthreads. @item HP-UX 10.20 should work. @item Added new function @code{DATE_FORMAT()}. @item Added @code{NOT IN}. @item Added automatic removal of 'ODBC function conversions': @code{@{fn now() @}} @item Handle ODBC 2.50.3 option flags. @item Fixed comparison of @code{DATE} and @code{TIME} values with @code{NULL}. @item Changed language name from germany to german to be consistent with the other language names. @item Fixed sorting problem on functions returning a @code{FLOAT}. Previously, the values were converted to @code{INT}s before sorting. @item Fixed slow sorting when sorting on key field when using @code{key_column=constant}. @item Sorting on calculated @code{DOUBLE} values sorted on integer results instead. @item @code{mysql} no longer needs a database argument. @item Changed the place where @code{HAVING} should be. According to ANSI, it should be after @code{GROUP BY} but before @code{ORDER BY}. MySQL Version 3.20 incorrectly had it last. @item Added Sybase command @code{USE DATABASE} to start using another database. @item Added automatic adjusting of number of connections and table cache size if the maximum number of files that can be opened is less than needed. This should fix that @code{mysqld} doesn't crash even if you haven't done a @code{ulimit -n 256} before starting @code{mysqld}. @item Added lots of limit checks to make it safer when running with too little memory or when doing weird queries. @end itemize @node News-3.21.13, News-3.21.12, News-3.21.14a, News-3.21.x @appendixsubsec Changes in release 3.21.13 @itemize @bullet @item Added retry of interrupted reads and clearing of @code{errno}. This makes Linux systems much safer! @item Fixed locking bug when using many aliases on the same table in the same @code{SELECT}. @item Fixed bug with @code{LIKE} on number key. @item New error message so you can check whether the connection was lost while the command was running or whether the connection was down from the start. @item Added @code{--table} option to @code{mysql} to print in table format. Moved time and row information after query result. Added automatic reconnect of lost connections. @item Added @code{!=} as a synonym for @code{<>}. @item Added function @code{VERSION()} to make easier logs. @item New multi-user test @file{tests/fork_test.pl} to put some strain on the thread library. @end itemize @node News-3.21.12, News-3.21.11, News-3.21.13, News-3.21.x @appendixsubsec Changes in release 3.21.12 @itemize @bullet @item Fixed @code{ftruncate()} call in MIT-pthreads. This made @code{isamchk} destroy the @file{.ISM} files on (Free)BSD 2.x systems. @item Fixed broken @code{__P_} patch in MIT-pthreads. @item Many memory overrun checks. All string functions now return @code{NULL} if the returned string should be longer than @code{max_allowed_packet} bytes. @item Changed the name of the @code{INTERVAL} type to @code{ENUM}, because @code{INTERVAL} is used in ANSI SQL. @item In some cases, doing a @code{JOIN} + @code{GROUP} + @code{INTO OUTFILE}, the result wasn't grouped. @item @code{LIKE} with @code{'_'} as last character didn't work. Fixed. @item Added extended ANSI SQL @code{TRIM()} function. @item Added @code{CURTIME()}. @item Added @code{ENCRYPT()} function by Zeev Suraski. @item Fixed better @code{FOREIGN KEY} syntax skipping. New reserved words: @code{MATCH}, @code{FULL}, @code{PARTIAL}. @item @code{mysqld} now allows IP number and hostname to the @code{--bind-address} option. @item Added @code{SET OPTION CHARACTER SET cp1251_koi8} to enable conversions of data to/from cp1251_koi8. @item Lots of changes for Win95 port. In theory, this version should now be easily portable to Win95. @item Changed the @code{CREATE COLUMN} syntax of @code{NOT NULL} columns to be after the @code{DEFAULT} value, as specified in the ANSI SQL standard. This will make @code{mysqldump} with @code{NOT NULL} and default values incompatible with MySQL Version 3.20. @item Added many function name aliases so the functions can be used with ODBC or ANSI SQL92 syntax. @item Fixed syntax of @code{ALTER TABLE tbl_name ALTER COLUMN col_name SET DEFAULT NULL}. @item Added @code{CHAR} and @code{BIT} as synonyms for @code{CHAR(1)}. @item Fixed core dump when updating as a user who has only @strong{select} privilege. @item @code{INSERT ... SELECT ... GROUP BY} didn't work in some cases. An @code{Invalid use of group function} error occurred. @item When using @code{LIMIT}, @code{SELECT} now always uses keys instead of record scan. This will give better performance on @code{SELECT} and a @code{WHERE} that matches many rows. @item Added Russian error messages. @end itemize @node News-3.21.11, News-3.21.10, News-3.21.12, News-3.21.x @appendixsubsec Changes in release 3.21.11 @itemize @bullet @item Configure changes. @item MySQL now works with the new thread library on BSD/OS 3.0. @item Added new group functions @code{BIT_OR()} and @code{BIT_AND()}. @item Added compatibility functions @code{CHECK} and @code{REFERENCES}. @code{CHECK} is now a reserved word. @item Added @code{ALL} option to @code{GRANT} for better compatibility. (@code{GRANT} is still a dummy function.) @item Added partly-translated dutch messages. @item Fixed bug in @code{ORDER BY} and @code{GROUP BY} with @code{NULL} columns. @item Added function @code{last_insert_id()} to retrieve last @code{AUTO_INCREMENT} value. This is intended for clients to ODBC that can't use the @code{mysql_insert_id()} API function, but can be used by any client. @item Added @code{--flush-logs} option to @code{mysqladmin}. @item Added command @code{STATUS} to @code{mysql}. @item Fixed problem with @code{ORDER BY}/@code{GROUP BY} because of bug in @code{gcc}. @item Fixed problem with @code{INSERT ... SELECT ... GROUP BY}. @end itemize @node News-3.21.10, News-3.21.9, News-3.21.11, News-3.21.x @appendixsubsec Changes in release 3.21.10 @itemize @bullet @item New @code{mysqlaccess}. @item @code{CREATE} now supports all ODBC types and the @code{mSQL} @code{TEXT} type. All ODBC 2.5 functions are also supported (added @code{REPEAT}). This provides better portability. @item Added text types @code{TINYTEXT}, @code{TEXT}, @code{MEDIUMTEXT} and @code{LONGTEXT}. These are actually @code{BLOB}types, but all searching is done in case-insensitive fashion. @item All old @code{BLOB} fields are now @code{TEXT} fields. This only changes that all searching on strings is done in case-sensitive fashion. You must do an @code{ALTER TABLE} and change the field type to @code{BLOB} if you want to have tests done in case-sensitive fashion. @item Fixed some @code{configure} issues. @item Made the locking code a bit safer. Fixed very unlikely deadlock situation. @item Fixed a couple of bugs in the range optimizer. Now the new range benchmark @code{test-select} works. @end itemize @node News-3.21.9, News-3.21.8, News-3.21.10, News-3.21.x @appendixsubsec Changes in release 3.21.9 @itemize @bullet @item Added @code{--enable-unix-socket=pathname} option to @code{configure}. @item Fixed a couple of portability problems with include files. @item Fixed bug in range calculation that could return empty set when searching on multiple key with only one entry (very rare). @item Most things ported to FSU Pthreads, which should allow MySQL to run on SCO. @xref{SCO}. @end itemize @node News-3.21.8, News-3.21.7, News-3.21.9, News-3.21.x @appendixsubsec Changes in release 3.21.8 @itemize @bullet @item Works now in Solaris 2.6. @item Added handling of calculation of @code{SUM()} functions. For example, you can now use @code{SUM(column)/COUNT(column)}. @item Added handling of trigometric functions: @code{PI()}, @code{ACOS()}, @code{ASIN()}, @code{ATAN()}, @code{COS()}, @code{SIN()} and @code{TAN()}. @item New languages: norwegian, norwegian-ny and portuguese. @item Fixed parameter bug in @code{net_print()} in @file{procedure.cc}. @item Fixed a couple of memory leaks. @item Now allow also the old @code{SELECT ... INTO OUTFILE} syntax. @item Fixed bug with @code{GROUP BY} and @code{SELECT} on key with many values. @item @code{mysql_fetch_lengths()} sometimes returned incorrect lengths when you used @code{mysql_use_result()}. This affected at least some cases of @code{mysqldump --quick}. @item Fixed bug in optimization of @code{WHERE const op field}. @item Fixed problem when sorting on @code{NULL} fields. @item Fixed a couple of 64-bit (Alpha) problems. @item Added @code{--pid-file=#} option to @code{mysqld}. @item Added date formatting to @code{FROM_UNIXTIME()}, originally by Zeev Suraski. @item Fixed bug in @code{BETWEEN} in range optimizer (Did only test = of the first argument). @item Added machine-dependent files for MIT-pthreads i386-SCO. There is probably more to do to get this to work on SCO 3.5. @end itemize @node News-3.21.7, News-3.21.6, News-3.21.8, News-3.21.x @appendixsubsec Changes in release 3.21.7 @itemize @bullet @item Changed @file{Makefile.am} to take advantage of Automake 1.2. @item Added the beginnings of a benchmark suite. @item Added more secure password handling. @item Added new client function @code{mysql_errno()}, to get the error number of the error message. This makes error checking in the client much easier. This makes the new server incompatible with the 3.20.x server when running without @code{--old-protocol}. The client code is backward compatible. More information can be found in the @file{README} file! @item Fixed some problems when using very long, illegal names. @end itemize @node News-3.21.6, News-3.21.5, News-3.21.7, News-3.21.x @appendixsubsec Changes in release 3.21.6 @itemize @bullet @item Fixed more portability issues (incorrect @code{sigwait} and @code{sigset} defines). @item @code{configure} should now be able to detect the last argument to @code{accept()}. @end itemize @node News-3.21.5, News-3.21.4, News-3.21.6, News-3.21.x @appendixsubsec Changes in release 3.21.5 @itemize @bullet @item Should now work with FreeBSD 3.0 if used with @file{FreeBSD-3.0-libc_r-1.0.diff}, which can be found at @uref{http://www.mysql.com/Downloads/Patches/}. @item Added new option @code{-O tmp_table_size=#} to @code{mysqld}. @item New function @code{FROM_UNIXTIME(timestamp)} which returns a date string in 'YYYY-MM-DD HH:MM:DD' format. @item New function @code{SEC_TO_TIME(seconds)} which returns a string in 'HH:MM:SS' format. @item New function @code{SUBSTRING_INDEX()}, originally by Zeev Suraski. @end itemize @node News-3.21.4, News-3.21.3, News-3.21.5, News-3.21.x @appendixsubsec Changes in release 3.21.4 @itemize @bullet @item Should now configure and compile on OSF1 4.0 with the DEC compiler. @item Configuration and compilation on BSD/OS 3.0 works, but due to some bugs in BSD/OS 3.0, @code{mysqld} doesn't work on it yet. @item Configuration and compilation on FreeBSD 3.0 works, but I couldn't get @code{pthread_create} to work. @end itemize @node News-3.21.3, News-3.21.2, News-3.21.4, News-3.21.x @appendixsubsec Changes in release 3.21.3 @itemize @bullet @item Added reverse check lookup of hostnames to get better security. @item Fixed some possible buffer overflows if filenames that are too long are used. @item @code{mysqld} doesn't accept hostnames that start with digits followed by a @code{'.'}, because the hostname may look like an IP number. @item Added @code{--skip-networking} option to @code{mysqld}, to only allow socket connections. (This will not work with MIT-pthreads!) @item Added check of too long table names for alias. @item Added check if database name is okay. @item Added check if too long table names. @item Removed incorrect @code{free()} that killed the server on @code{CREATE DATABASE} or @code{DROP DATABASE}. @item Changed some @code{mysqld} @code{-O} options to better names. @item Added @code{-O join_cache_size=#} option to @code{mysqld}. @item Added @code{-O max_join_size=#} option to @code{mysqld}, to be able to set a limit how big queries (in this case big = slow) one should be able to handle without specifying @code{SET OPTION SQL_BIG_SELECTS=1}. A # = is about 10 examined records. The default is ``unlimited''. @item When comparing a @code{TIME}, @code{DATE}, @code{DATETIME} or @code{TIMESTAMP} column to a constant, the constant is converted to a time value before performing the comparison. This will make it easier to get ODBC (particularly Access97) to work with the above types. It should also make dates easier to use and the comparisons should be quicker than before. @item Applied patch from Jochen Wiedmann that allows @code{query()} in @code{mysqlperl} to take a query with @code{\0} in it. @item Storing a timestamp with a 2-digit year (@code{YYMMDD}) didn't work. @item Fix that timestamp wasn't automatically updated if set in an @code{UPDATE} clause. @item Now the automatic timestamp field is the FIRST timestamp field. @item @code{SELECT * INTO OUTFILE}, which didn't correctly if the outfile already existed. @item @code{mysql} now shows the thread ID when starting or doing a reconnect. @item Changed the default sort buffer size from 2M to 1M. @end itemize @node News-3.21.2, News-3.21.0, News-3.21.3, News-3.21.x @appendixsubsec Changes in release 3.21.2 @itemize @bullet @item The range optimizer is coded, but only 85% tested. It can be enabled with @code{--new}, but it crashes core a lot yet... @item More portable. Should compile on AIX and alpha-digital. At least the @code{isam} library should be relatively 64-bit clean. @item New @code{isamchk} which can detect and fix more problems. @item New options for @code{isamlog}. @item Using new version of Automake. @item Many small portability changes (from the AIX and alpha-digital port) Better checking of pthread(s) library. @item czech error messages by @email{snajdr@@pvt.net}. @item Decreased size of some buffers to get fewer problems on systems with little memory. Also added more checks to handle ``out of memory'' problems. @item @code{mysqladmin}: you can now do @code{mysqladmin kill 5,6,7,8} to kill multiple threads. @item When the maximum connection limit is reached, one extra connection by a user with the @strong{PROCESS_ACL} privilege is granted. @item Added @code{-O backlog=#} option to @code{mysqld}. @item Increased maximum packet size from 512K to 1024K for client. @item Almost all of the function code is now tested in the internal test suite. @item @code{ALTER TABLE} now returns warnings from field conversions. @item Port changed to 3306 (got it reserved from ISI). @item Added a fix for Visual FoxBase so that any schema name from a table specification is automatically removed. @item New function @code{ASCII()}. @item Removed function @code{BETWEEN(a,b,c)}. Use the standard ANSI syntax instead: @code{expr BETWEEN expr AND expr}. @item MySQL no longer has to use an extra temporary table when sorting on functions or @code{SUM()} functions. @item Fixed bug that you couldn't use @code{tbl_name.field_name} in @code{UPDATE}. @item Fixed @code{SELECT DISTINCT} when using 'hidden group'. For example: @example mysql> SELECT DISTINCT MOD(some_field,10) FROM test GROUP BY some_field; @end example Note: @code{some_field} is normally in the @code{SELECT} part. ANSI SQL should require it. @end itemize @node News-3.21.0, , News-3.21.2, News-3.21.x @appendixsubsec Changes in release 3.21.0 @itemize @bullet @item New reserved words used: @code{INTERVAL}, @code{EXPLAIN}, @code{READ}, @code{WRITE}, @code{BINARY}. @item Added ODBC function @code{CHAR(num,...)}. @item New operator @code{IN}. This uses a binary search to find a match. @item New command @code{LOCK TABLES tbl_name [AS alias] @{READ|WRITE@} ...} @item Added @code{--log-update} option to @code{mysqld}, to get a log suitable for incremental updates. @item New command @code{EXPLAIN SELECT ...} to get information about how the optimizer will do the join. @item For easier client code, the client should no longer use @code{FIELD_TYPE_TINY_BLOB}, @code{FIELD_TYPE_MEDIUM_BLOB}, @code{FIELD_TYPE_LONG_BLOB} or @code{FIELD_TYPE_VAR_STRING} (as previously returned by @code{mysql_list_fields}). You should instead only use @code{FIELD_TYPE_BLOB} or @code{FIELD_TYPE_STRING}. If you want exact types, you should use the command @code{SHOW FIELDS}. @item Added varbinary syntax: @code{0x######} which can be used as a string (default) or a number. @item @code{FIELD_TYPE_CHAR} is renamed to @code{FIELD_TYPE_TINY}. @item Changed all fields to C++ classes. @item Removed FORM struct. @item Fields with @code{DEFAULT} values no longer need to be @code{NOT NULL}. @item New field types: @table @code @item ENUM A string which can take only a couple of defined values. The value is stored as a 1-3 byte number that is mapped automatically to a string. This is sorted according to string positions! @item SET A string which may have one or many string values separated with ','. The string is stored as a 1-, 2-, 3-, 4- or 8-byte number where each bit stands for a specific set member. This is sorted according to the unsigned value of the stored packed number. @end table @item Now all function calculation is done with @code{double} or @code{long long}. This will provide the full 64-bit range with bit functions and fix some conversions that previously could result in precision losses. One should avoid using @code{unsigned long long} columns with full 64-bit range (numbers bigger than 9223372036854775807) because calculations are done with @code{signed long long}. @item @code{ORDER BY} will now put @code{NULL} field values first. @code{GROUP BY} will also work with @code{NULL} values. @item Full @code{WHERE} with expressions. @item New range optimizer that can resolve ranges when some keypart prefix is constant. Example: @example mysql> SELECT * FROM tbl_name WHERE key_part_1="customer" AND key_part_2>=10 AND key_part_2<=10; @end example @end itemize @node News-3.20.x, News-3.19.x, News-3.21.x, News @appendixsec Changes in release 3.20.x @cindex changes, version 3.20 Version 3.20 is quite old now, and should be avoided if possible. This information is kept here for historical purposes only. Changes from 3.20.18 to 3.20.32b are not documented here because the 3.21 release branched here. And the relevant changes are also documented as changes to the 3.21 version. @menu * News-3.20.18:: Changes in release 3.20.18 * News-3.20.17:: Changes in release 3.20.17 * News-3.20.16:: Changes in release 3.20.16 * News-3.20.15:: Changes in release 3.20.15 * News-3.20.14:: Changes in release 3.20.14 * News-3.20.13:: Changes in release 3.20.13 * News-3.20.11:: Changes in release 3.20.11 * News-3.20.10:: Changes in release 3.20.10 * News-3.20.9:: Changes in release 3.20.9 * News-3.20.8:: Changes in release 3.20.8 * News-3.20.7:: Changes in release 3.20.7 * News-3.20.6:: Changes in release 3.20.6 * News-3.20.3:: Changes in release 3.20.3 * News-3.20.0:: Changes in releases 3.20.0 @end menu @node News-3.20.18, News-3.20.17, News-3.20.x, News-3.20.x @appendixsubsec Changes in release 3.20.18 @itemize @bullet @item Added @code{-p#} (remove @code{#} directories from path) to @code{isamlog}. All files are written with a relative path from the database directory Now @code{mysqld} shouldn't crash on shutdown when using the @code{--log-isam} option. @item New @code{mysqlperl} version. It is now compatible with @code{msqlperl-0.63}. @item New @code{DBD} module available at @uref{http://www.mysql.com/Downloads/Contrib/} site. @item Added group function @code{STD()} (standard deviation). @item The @code{mysqld} server is now compiled by default without debugging information. This will make the daemon smaller and faster. @item Now one usually only has to specify the @code{--basedir} option to @code{mysqld}. All other paths are relative in a normal installation. @item @code{BLOB} columns sometimes contained garbage when used with a @code{SELECT} on more than one table and @code{ORDER BY}. @item Fixed that calculations that are not in @code{GROUP BY} work as expected (ANSI SQL extension). Example: @example mysql> SELECT id,id+1 FROM table GROUP BY id; @end example @item The test of using @code{MYSQL_PWD} was reversed. Now @code{MYSQL_PWD} is enabled as default in the default release. @item Fixed conversion bug which caused @code{mysqld} to core dump with Arithmetic error on Sparc-386. @item Added @code{--unbuffered} option to @code{mysql}, for new @code{mysqlaccess}. @item When using overlapping (unnecessary) keys and join over many tables, the optimizer could get confused and return 0 records. @end itemize @node News-3.20.17, News-3.20.16, News-3.20.18, News-3.20.x @appendixsubsec Changes in release 3.20.17 @itemize @bullet @item You can now use @code{BLOB} columns and the functions @code{IS NULL} and @code{IS NOT NULL} in the @code{WHERE} clause. @item All communication packets and row buffers are now allocated dynamically on demand. The default value of @code{max_allowed_packet} is now 64K for the server and 512K for the client. This is mainly used to catch incorrect packets that could trash all memory. The server limit may be changed when it is started. @item Changed stack usage to use less memory. @item Changed @code{safe_mysqld} to check for running daemon. @item The @code{ELT()} function is renamed to @code{FIELD()}. The new @code{ELT()} function returns a value based on an index: @code{FIELD()} is the inverse of @code{ELT()} Example: @code{ELT(2,"A","B","C")} returns @code{"B"}. @code{FIELD("B","A","B","C")} returns @code{2}. @item @code{COUNT(field)}, where @code{field} could have a @code{NULL} value, now works. @item A couple of bugs fixed in @code{SELECT ... GROUP BY}. @item Fixed memory overrun bug in @code{WHERE} with many unoptimizable brace levels. @item Fixed some small bugs in the grant code. @item If hostname isn't found by @code{get_hostname}, only the IP is checked. Previously, you got @code{Access denied}. @item Inserts of timestamps with values didn't always work. @item @code{INSERT INTO ... SELECT ... WHERE} could give the error @code{Duplicated field}. @item Added some tests to @code{safe_mysqld} to make it ``safer''. @item @code{LIKE} was case sensitive in some places and case insensitive in others. Now @code{LIKE} is always case insensitive. @item @file{mysql.cc}: Allow @code{'#'} anywhere on the line. @item New command @code{SET OPTION SQL_SELECT_LIMIT=#}. See the FAQ for more details. @item New version of the @code{mysqlaccess} script. @item Change @code{FROM_DAYS()} and @code{WEEKDAY()} to also take a full @code{TIMESTAMP} or @code{DATETIME} as argument. Before they only took a number of type @code{YYYYMMDD} or @code{YYMMDD}. @item Added new function @code{UNIX_TIMESTAMP(timestamp_column)}. @end itemize @node News-3.20.16, News-3.20.15, News-3.20.17, News-3.20.x @appendixsubsec Changes in release 3.20.16 @itemize @bullet @item More changes in MIT-pthreads to get them safer. Fixed also some link bugs at least in SunOS. @item Changed @code{mysqld} to work around a bug in MIT-pthreads. This makes multiple small @code{SELECT} operations 20 times faster. Now @code{lock_test.pl} should work. @item Added @code{mysql_FetchHash(handle)} to @code{mysqlperl}. @item The @code{mysqlbug} script is now distributed built to allow for reporting bugs that appear during the build with it. @item Changed @file{libmysql.c} to prefer @code{getpwuid()} instead of @code{cuserid()}. @item Fixed bug in @code{SELECT} optimizer when using many tables with the same column used as key to different tables. @item Added new latin2 and Russian KOI8 character tables. @item Added support for a dummy @code{GRANT} command to satisfy Powerbuilder. @end itemize @node News-3.20.15, News-3.20.14, News-3.20.16, News-3.20.x @appendixsubsec Changes in release 3.20.15 @itemize @bullet @item Fixed fatal bug @code{packets out of order} when using MIT-pthreads. @item Removed possible loop when a thread waits for command from client and @code{fcntl()} fails. Thanks to Mike Bretz for finding this bug. @item Changed alarm loop in @file{mysqld.cc} because shutdown didn't always succeed in Linux. @item Removed use of @code{termbits} from @file{mysql.cc}. This conflicted with @code{glibc} 2.0. @item Fixed some syntax errors for at least BSD and Linux. @item Fixed bug when doing a @code{SELECT} as superuser without a database. @item Fixed bug when doing @code{SELECT} with group calculation to outfile. @end itemize @node News-3.20.14, News-3.20.13, News-3.20.15, News-3.20.x @appendixsubsec Changes in release 3.20.14 @itemize @bullet @item If one gives @code{-p} or @code{--password} option to @code{mysql} without an argument, the user is solicited for the password from the tty. @item Added default password from @code{MYSQL_PWD} (by Elmar Haneke). @item Added command @code{kill} to @code{mysqladmin} to kill a specific MySQL thread. @item Sometimes when doing a reconnect on a down connection this succeeded first on second try. @item Fixed adding an @code{AUTO_INCREMENT} key with @code{ALTER_TABLE}. @item @code{AVG()} gave too small value on some @code{SELECT}s with @code{GROUP BY} and @code{ORDER BY}. @item Added new @code{DATETIME} type (by Giovanni Maruzzelli @email{maruzz@@matrice.it}). @item Fixed that define @code{DONT_USE_DEFAULT_FIELDS} works. @item Changed to use a thread to handle alarms instead of signals on Solaris to avoid race conditions. @item Fixed default length of signed numbers. (George Harvey @email{georgeh@@pinacl.co.uk}.) @item Allow anything for @code{CREATE INDEX}. @item Add prezeros when packing numbers to @code{DATE}, @code{TIME} and @code{TIMESTAMP}. @item Fixed a bug in @code{OR} of multiple tables (gave empty set). @item Added many patches to MIT-pthreads. This fixes at least one lookup bug. @end itemize @node News-3.20.13, News-3.20.11, News-3.20.14, News-3.20.x @appendixsubsec Changes in release 3.20.13 @itemize @bullet @item Added ANSI SQL94 @code{DATE} and @code{TIME} types. @item Fixed bug in @code{SELECT} with @code{AND}-@code{OR} levels. @item Added support for Slovenian characters. The @file{Contrib} directory contains source and instructions for adding other character sets. @item Fixed bug with @code{LIMIT} and @code{ORDER BY}. @item Allow @code{ORDER BY} and @code{GROUP BY} on items that aren't in the @code{SELECT} list. (Thanks to Wim Bonis @email{bonis@@kiss.de}, for pointing this out.) @item Allow setting of timestamp values in @code{INSERT}. @item Fixed bug with @code{SELECT ... WHERE ... = NULL}. @item Added changes for @code{glibc} 2.0. To get @code{glibc} to work, you should add the @file{gibc-2.0-sigwait-patch} before compiling @code{glibc}. @item Fixed bug in @code{ALTER TABLE} when changing a @code{NOT NULL} field to allow @code{NULL} values. @item Added some ANSI92 synonyms as field types to @code{CREATE TABLE}. @code{CREATE TABLE} now allows @code{FLOAT(4)} and @code{FLOAT(8)} to mean @code{FLOAT} and @code{DOUBLE}. @item New utility program @code{mysqlaccess} by @email{Yves.Carlier@@rug.ac.be}. This program shows the access rights for a specific user and the grant rows that determine this grant. @item Added @code{WHERE const op field} (by @email{bonis@@kiss.de}). @end itemize @node News-3.20.11, News-3.20.10, News-3.20.13, News-3.20.x @appendixsubsec Changes in release 3.20.11 @itemize @bullet @item When using @code{SELECT ... INTO OUTFILE}, all temporary tables are ISAM instead of HEAP to allow big dumps. @item Changed date functions to be string functions. This fixed some ``funny'' side effects when sorting on dates. @item Extended @code{ALTER TABLE} according to SQL92. @item Some minor compatibility changes. @item Added @code{--port} and @code{--socket} options to all utility programs and @code{mysqld}. @item Fixed MIT-pthreads @code{readdir_r()}. Now @code{mysqladmin create database} and @code{mysqladmin drop database} should work. @item Changed MIT-pthreads to use our @code{tempnam()}. This should fix the ``sort aborted'' bug. @item Added sync of records count in @code{sql_update}. This fixed slow updates on first connection. (Thanks to Vaclav Bittner for the test.) @end itemize @node News-3.20.10, News-3.20.9, News-3.20.11, News-3.20.x @appendixsubsec Changes in release 3.20.10 @itemize @bullet @item New insert type: @code{INSERT INTO ... SELECT ...} @item @code{MEDIUMBLOB} fixed. @item Fixed bug in @code{ALTER TABLE} and @code{BLOB}s. @item @code{SELECT ... INTO OUTFILE} now creates the file in the current database directory. @item @code{DROP TABLE} now can take a list of tables. @item Oracle synonym @code{DESCRIBE} (@code{DESC}). @item Changes to @code{make_binary_distribution}. @item Added some comments to installation instructions about @code{configure}'s C++ link test. @item Added @code{--without-perl} option to @code{configure}. @item Lots of small portability changes. @end itemize @node News-3.20.9, News-3.20.8, News-3.20.10, News-3.20.x @appendixsubsec Changes in release 3.20.9 @itemize @bullet @item @code{ALTER TABLE} didn't copy null bit. As a result, fields that were allowed to have @code{NULL} values were always @code{NULL}. @item @code{CREATE} didn't take numbers as @code{DEFAULT}. @item Some compatibility changes for SunOS. @item Removed @file{config.cache} from old distribution. @end itemize @node News-3.20.8, News-3.20.7, News-3.20.9, News-3.20.x @appendixsubsec Changes in release 3.20.8 @itemize @bullet @item Fixed bug with @code{ALTER TABLE} and multi-part keys. @end itemize @node News-3.20.7, News-3.20.6, News-3.20.8, News-3.20.x @appendixsubsec Changes in release 3.20.7 @itemize @bullet @item New commands: @code{ALTER TABLE}, @code{SELECT ... INTO OUTFILE} and @code{LOAD DATA INFILE}. @item New function: @code{NOW()}. @item Added new field @strong{file_priv} to @code{mysql/user} table. @item New script @code{add_file_priv} which adds the new field @strong{file_priv} to the @code{user} table. This script must be executed if you want to use the new @code{SELECT ... INTO} and @code{LOAD DATA INFILE ...} commands with a version of MySQL earlier than 3.20.7. @item Fixed bug in locking code, which made @code{lock_test.pl} test fail. @item New files @file{NEW} and @file{BUGS}. @item Changed @file{select_test.c} and @file{insert_test.c} to include @file{config.h}. @item Added command @code{status} to @code{mysqladmin} for short logging. @item Increased maximum number of keys to 16 and maximum number of key parts to 15. @item Use of sub keys. A key may now be a prefix of a string field. @item Added @code{-k} option to @code{mysqlshow}, to get key information for a table. @item Added long options to @code{mysqldump}. @end itemize @node News-3.20.6, News-3.20.3, News-3.20.7, News-3.20.x @appendixsubsec Changes in release 3.20.6 @itemize @bullet @item Portable to more systems because of MIT-pthreads, which will be used automatically if @code{configure} cannot find a @code{-lpthreads} library. @item Added GNU-style long options to almost all programs. Test with @code{@kbd{program} --help}. @item Some shared library support for Linux. @item The FAQ is now in @file{.texi} format and is available in @file{.html}, @file{.txt} and @file{.ps} formats. @item Added new SQL function @code{RAND([init])}. @item Changed @code{sql_lex} to handle @code{\0} unquoted, but the client can't send the query through the C API, because it takes a str pointer. You must use @code{mysql_real_query()} to send the query. @item Added API function @code{mysql_get_client_info()}. @item @code{mysqld} now uses the @code{N_MAX_KEY_LENGTH} from @file{nisam.h} as the maximum allowed key length. @item The following now works: @example mysql> SELECT filter_nr,filter_nr FROM filter ORDER BY filter_nr; @end example Previously, this resulted in the error: @code{Column: 'filter_nr' in order clause is ambiguous}. @item @code{mysql} now outputs @code{'\0'}, @code{'\t'}, @code{'\n'} and @code{'\\'} when encountering ASCII 0, tab, newline or @code{'\'} while writing tab-separated output. This is to allow printing of binary data in a portable format. To get the old behavior, use @code{-r} (or @code{--raw}). @item Added german error messages (60 of 80 error messages translated). @item Added new API function @code{mysql_fetch_lengths(MYSQL_RES *)}, which returns an array of column lengths (of type @code{uint}). @item Fixed bug with @code{IS NULL} in @code{WHERE} clause. @item Changed the optimizer a little to get better results when searching on a key part. @item Added @code{SELECT} option @code{STRAIGHT_JOIN} to tell the optimizer that it should join tables in the given order. @item Added support for comments starting with @code{'--'} in @file{mysql.cc} (Postgres syntax). @item You can have @code{SELECT} expressions and table columns in a @code{SELECT} which are not used in the group part. This makes it efficient to implement lookups. The column that is used should be a constant for each group because the value is calculated only once for the first row that is found for a group. @example mysql> SELECT id,lookup.text,sum(*) FROM test,lookup WHERE test.id=lookup.id GROUP BY id; @end example @item Fixed bug in @code{SUM(function)} (could cause a core dump). @item Changed @code{AUTO_INCREMENT} placement in the SQL query: @example INSERT into table (auto_field) values (0); @end example inserted 0, but it should insert an @code{AUTO_INCREMENT} value. @item @file{mysqlshow.c}: Added number of records in table. Had to change the client code a little to fix this. @item @code{mysql} now allows doubled @code{''} or @code{""} within strings for embedded @code{'} or @code{"}. @item New math functions: @code{EXP()}, @code{LOG()}, @code{SQRT()}, @code{ROUND()}, @code{CEILING()}. @end itemize @node News-3.20.3, News-3.20.0, News-3.20.6, News-3.20.x @appendixsubsec Changes in release 3.20.3 @itemize @bullet @item The @code{configure} source now compiles a thread-free client library @code{-lmysqlclient}. This is the only library that needs to be linked with client applications. When using the binary releases, you must link with @code{-lmysql -lmysys -ldbug -lstrings} as before. @item New @code{readline} library from @code{bash-2.0}. @item LOTS of small changes to @code{configure} and makefiles (and related source). @item It should now be possible to compile in another directory using @code{VPATH}. Tested with GNU Make 3.75. @item @code{safe_mysqld} and @code{mysql.server} changed to be more compatible between the source and the binary releases. @item @code{LIMIT} now takes one or two numeric arguments. If one argument is given, it indicates the maximum number of rows in a result. If two arguments are given, the first argument indicates the offset of the first row to return, the second is the maximum number of rows. With this it's easy to do a poor man's next page/previous page WWW application. @item Changed name of SQL function @code{FIELDS()} to @code{ELT()}. Changed SQL function @code{INTERVALL()} to @code{INTERVAL()}. @item Made @code{SHOW COLUMNS} a synonym for @code{SHOW FIELDS}. Added compatibility syntax @code{FRIEND KEY} to @code{CREATE TABLE}. In MySQL, this creates a non-unique key on the given columns. @item Added @code{CREATE INDEX} and @code{DROP INDEX} as compatibility functions. In MySQL, @code{CREATE INDEX} only checks if the index exists and issues an error if it doesn't exist. @code{DROP INDEX} always succeeds. @item @file{mysqladmin.c}: added client version to version information. @item Fixed core dump bug in @code{sql_acl} (core on new connection). @item Removed @code{host}, @code{user} and @code{db} tables from database @code{test} in the distribution. @item @code{FIELD_TYPE_CHAR} can now be signed (-128 - 127) or unsigned (0 - 255) Previously, it was always unsigned. @item Bug fixes in @code{CONCAT()} and @code{WEEKDAY()}. @item Changed a lot of source to get @code{mysqld} to be compiled with SunPro compiler. @item SQL functions must now have a @code{'('} immediately after the function name (no intervening space). For example, @code{'user('} is regarded as beginning a function call, and @code{'user ('} is regarded as an identifier @code{user} followed by a @code{'('}, not as a function call. @end itemize @node News-3.20.0, , News-3.20.3, News-3.20.x @appendixsubsec Changes in release 3.20.0 @itemize @bullet @item The source distribution is done with @code{configure} and Automake. It will make porting much easier. The @code{readline} library is included in the distribution. @item Separate client compilation: the client code should be very easy to compile on systems which don't have threads. @item The old Perl interface code is automatically compiled and installed. Automatic compiling of @code{DBD} will follow when the new @code{DBD} code is ported. @item Dynamic language support: @code{mysqld} can now be started with Swedish or English (default) error messages. @item New functions: @code{INSERT()}, @code{RTRIM()}, @code{LTRIM()} and @code{FORMAT()}. @item @code{mysqldump} now works correctly for all field types (even @code{AUTO_INCREMENT}). The format for @code{SHOW FIELDS FROM tbl_name} is changed so the @code{Type} column contains information suitable for @code{CREATE TABLE}. In previous releases, some @code{CREATE TABLE} information had to be patched when re-creating tables. @item Some parser bugs from 3.19.5 (@code{BLOB} and @code{TIMESTAMP}) are corrected. @code{TIMESTAMP} now returns different date information depending on its create length. @item Changed parser to allow a database, table or field name to start with a number or @code{'_'}. @item All old C code from Unireg changed to C++ and cleaned up. This makes the daemon a little smaller and easier to understand. @item A lot of small bug fixes done. @item New @file{INSTALL} files (not final version) and some information regarding porting. @end itemize @node News-3.19.x, , News-3.20.x, News @appendixsec Changes in release 3.19.x @cindex changes, version 3.19 Version 3.19 is quite old now, and should be avoided if possible. This information is kept here for historical purposes only. @menu * News-3.19.5:: Changes in release 3.19.5 * News-3.19.4:: Changes in release 3.19.4 * News-3.19.3:: Changes in release 3.19.3 @end menu @node News-3.19.5, News-3.19.4, News-3.19.x, News-3.19.x @appendixsubsec Changes in release 3.19.5 @itemize @bullet @item Some new functions, some more optimization on joins. @item Should now compile clean on Linux (2.0.x). @item Added functions @code{DATABASE()}, @code{USER()}, @code{POW()}, @code{LOG10()} (needed for ODBC). @item In a @code{WHERE} with an @code{ORDER BY} on fields from only one table, the table is now preferred as first table in a multi-join. @item @code{HAVING} and @code{IS NULL} or @code{IS NOT NULL} now works. @item A group on one column and a sort on a group function (@code{SUM()}, @code{AVG()}...) didn't work together. Fixed. @item @code{mysqldump}: Didn't send password to server. @end itemize @node News-3.19.4, News-3.19.3, News-3.19.5, News-3.19.x @appendixsubsec Changes in release 3.19.4 @itemize @bullet @item Fixed horrible locking bug when inserting in one thread and reading in another thread. @item Fixed one-off decimal bug. 1.00 was output as 1.0. @item Added attribute @code{'Locked'} to process list as info if a query is locked by another query. @item Fixed full magic timestamp. Timestamp length may now be 14, 12, 10, 8, 6, 4 or 2 bytes. @item Sort on some numeric functions could sort incorrectly on last number. @item @code{IF(arg,syntax_error,syntax_error)} crashed. @item Added functions @code{CEILING()}, @code{ROUND()}, @code{EXP()}, @code{LOG()} and @code{SQRT()}. @item Enhanced @code{BETWEEN} to handle strings. @end itemize @node News-3.19.3, , News-3.19.4, News-3.19.x @appendixsubsec Changes in release 3.19.3 @itemize @bullet @item Fixed @code{SELECT} with grouping on @code{BLOB} columns not to return incorrect @code{BLOB} info. Grouping, sorting and distinct on @code{BLOB} columns will not yet work as expected (probably it will group/sort by the first 7 characters in the @code{BLOB}). Grouping on formulas with a fixed string size (use @code{MID()} on a @code{BLOB}) should work. @item When doing a full join (no direct keys) on multiple tables with @code{BLOB} fields, the @code{BLOB} was garbage on output. @item Fixed @code{DISTINCT} with calculated columns. @end itemize @node Porting, Environment variables, News, Top @appendix Comments on porting to other systems @cindex porting, to other systems A working Posix thread library is needed for the server. On Solaris 2.5 we use Sun PThreads (the native thread support in 2.4 and earlier versions are not good enough) and on Linux we use LinuxThreads by Xavier Leroy, @email{Xavier.Leroy@@inria.fr}. The hard part of porting to a new Unix variant without good native thread support is probably to port MIT-pthreads. See @file{mit-pthreads/README} and @uref{http://www.humanfactor.com/pthreads/, Programming POSIX Threads}. The MySQL distribution includes a patched version of Provenzano's Pthreads from MIT (see @uref{http://www.mit.edu:8001/people/proven/pthreads.html, MIT Pthreads web page}). This can be used for some operating systems that do not have POSIX threads. It is also possible to use another user level thread package named FSU Pthreads (see @uref{http://www.informatik.hu-berlin.de/~mueller/pthreads.html, FSU Pthreads home page}). This implementation is being used for the SCO port. See the @file{thr_lock.c} and @file{thr_alarm.c} programs in the @file{mysys} directory for some tests/examples of these problems. Both the server and the client need a working C++ compiler (we use @code{gcc} and have tried SparcWorks). Another compiler that is known to work is the Irix @code{cc}. To compile only the client use @code{./configure --without-server}. There is currently no support for only compiling the server, nor is it likly to be added unless someone has a good reason for it. If you want/need to change any @file{Makefile} or the configure script you must get Automake and Autoconf. We have used the @code{automake-1.2} and @code{autoconf-2.12} distributions. All steps needed to remake everything from the most basic files. @example /bin/rm */.deps/*.P /bin/rm -f config.cache aclocal autoheader aclocal automake autoconf ./configure --with-debug=full --prefix='your installation directory' # The makefiles generated above need GNU make 3.75 or newer. # (called gmake below) gmake clean all install init-db @end example If you run into problems with a new port, you may have to do some debugging of MySQL! @xref{Debugging server}. @strong{NOTE:} Before you start debugging @code{mysqld}, first get the test programs @code{mysys/thr_alarm} and @code{mysys/thr_lock} to work. This will ensure that your thread installation has even a remote chance to work! @menu * Debugging server:: Debugging a MySQL server * Debugging client:: Debugging a MySQL client * The DBUG package:: The DBUG package * Locking methods:: Locking methods * RTS-threads:: Comments about RTS threads * Thread packages:: Differences between different thread packages @end menu @node Debugging server, Debugging client, Porting, Porting @appendixsec Debugging a MySQL server @cindex server, debugging @cindex debugging, server @cindex crash If you are using some functionality that is very new in MySQL, you can try to run @code{mysqld} with the @code{--skip-new} (which will disable all new, potentially unsafe functionality) or with @code{--safe-mode} which disables a lot of optimization that may cause problems. @xref{Crashing}. If @code{mysqld} doesn't want to start, you should check that you don't have any @code{my.cnf} files that interfere with your setup! You can check your @code{my.cnf} arguments with @code{mysqld --print-defaults} and avoid using them by starting with @code{mysqld --no-defaults ...}. If @code{mysqld} starts to eat up CPU or memory or if it ``hangs'', you can use @code{mysqladmin processlist status} to find out if someone is executing a query that takes a long time. It may be a good idea to run @code{mysqladmin -i10 processlist status} in some window if you are experiencing performance problems or problems when new clients can't connect. The command @code{mysqladmin debug} will dump some information about locks in use, used memory and query usage to the mysql log file. This may help solve some problems. This command also provides some useful information even if you haven't compiled MySQL for debugging! If the problem is that some tables are getting slower and slower you should try to optimize the table with @code{OPTIMIZE TABLE} or @code{myisamchk}. @xref{MySQL Database Administration}. You should also check the slow queries with @code{EXPLAIN}. You should also read the OS-specific section in this manual for problems that may be unique to your environment. @xref{Operating System Specific Notes}. @menu * Compiling for debugging:: Compiling MYSQL for debugging. * Making trace files:: Creating trace files * Using gdb on mysqld:: Debugging mysqld under gdb * Using stack trace:: Using a stack trace * Using log files:: Using log files to find cause of errors in mysqld * Reproduceable test case:: Making a test case when you experience table corruption @end menu @node Compiling for debugging, Making trace files, Debugging server, Debugging server @appendixsubsec Compiling MYSQL for debugging. If you have some very specific problem, you can always try to debug MySQL. To do this you must configure MySQL with the @code{--with-debug} or the @code{--with-debug=full} option. You can check whether or not MySQL was compiled with debugging by doing: @code{mysqld --help}. If the @code{--debug} flag is listed with the options then you have debugging enabled. @code{mysqladmin ver} also lists the @code{mysqld} version as @code{mysql ... --debug} in this case. If you are using gcc or egcs, the recommended configure line is: @example CC=gcc CFLAGS="-O2" CXX=gcc CXXFLAGS="-O2 -felide-constructors -fno-exceptions -fno-rtti" ./configure --prefix=/usr/local/mysql --with-debug --with-extra-charsets=complex @end example This will avoid problems with the @code{libstdc++} library and with C++ exceptions (many compilers have problems with C++ exceptions in threaded code) and compile a MySQL version with support for all character sets. If you suspect a memory overrun error, you can configure MySQL with @code{--with-debug=full}, which will install a memory allocation (@code{SAFEMALLOC}) checker. Running with @code{SAFEMALLOC} is however quite slow, so if you get performance problems you should start @code{mysqld} with the @code{--skip-safemalloc} option. This will disable the memory overrun checks for each call to @code{malloc} and @code{free}. If @code{mysqld} stops crashing when you compile it with @code{--with-debug}, you have probably found a compiler bug or a timing bug within MySQL. In this case you can try to add @code{-g} to the @code{CFLAGS} and @code{CXXFLAGS} variables above and not use @code{--with-debug}. If @code{mysqld} now dies, you can at least attach to it with @code{gdb} or use @code{gdb} on the core file to find out what happened. When you configure MySQL for debugging you automatically enable a lot of extra safety check functions that monitor the health of @code{mysqld}. If they find something ``unexpected,'' an entry will be written to @code{stderr}, which @code{safe_mysqld} directs to the error log! This also means that if you are having some unexpected problems with MySQL and are using a source distribution, the first thing you should do is to configure MySQL for debugging! (The second thing, of course, is to send mail to @email{mysql@@lists.mysql.com} and ask for help. Please use the @code{mysqlbug} script for all bug reports or questions regarding the MySQL version you are using! In the Windows MySQL distribution, @code{mysqld.exe} is by default compiled with support for trace files. @node Making trace files, Using gdb on mysqld, Compiling for debugging, Debugging server @appendixsubsec Creating trace files If the @code{mysqld} server doesn't start or if you can cause the @code{mysqld} server to crash quickly, you can try to create a trace file to find the problem. To do this you have to have a @code{mysqld} that is compiled for debugging. You can check this by executing @code{mysqld -V}. If the version number ends with @code{-debug}, it's compiled with support for trace files. Start the @code{mysqld} server with a trace log in @file{/tmp/mysqld.trace} (or @file{C:\mysqld.trace} on Windows): @code{mysqld --debug} On Windows you should also use the @code{--standalone} flag to not start @code{mysqld} as a service: In a DOS window do: @example mysqld --debug --standalone @end example After this you can use the @code{mysql.exe} command line tool in a second DOS window to reproduce the problem. You can take down the above @code{mysqld} server with @code{mysqladmin shutdown}. Note that the trace file will get very @emph{BIG}! If you want to have a smaller trace file, you can use something like: @code{mysqld --debug=d,info,error,query,general,where:O,/tmp/mysqld.trace} which only prints information with the most interesting tags in @file{/tmp/mysqld.trace}. If you make a bug report about this, please only send the lines from the trace file to the appropriate mailing list where something seems to go wrong! If you can't locate the wrong place, you can ftp the trace file, together with a full bug report, to @uref{ftp://support.mysql.com/pub/mysql/secret} so that a MySQL developer can take a look a this. The trace file is made with the @strong{DBUG} package by Fred Fish. @xref{The DBUG package}. @node Using gdb on mysqld, Using stack trace, Making trace files, Debugging server @appendixsubsec Debugging mysqld under gdb @cindex gdb, using On most system you can also start @code{mysqld} from @code{gdb} to get more information if @code{mysqld} crashes. With some older @code{gdb} versions on Linux you must use @code{run --one-thread} if you want to be able to debug @code{mysqld} threads. In this case you can only have one thread active at a time. When running @code{mysqld} under gdb, you should disable the stack trace with @code{--skip-stack-trace} to be able to catch segfaults within gdb. It's very hard to debug MySQL under @code{gdb} if you do a lot of new connections the whole time as @code{gdb} doesn't free the memory for old threads. You can avoid this problem by starting @code{mysqld} with @code{-O thread_cache_size= 'max_connections +1'}. In most cases just using @code{-O thread_cache_size=5'} will help a lot! If you want to get a core dump on Linux if @code{mysqld} dies with a SIGSEGV signal, you can start @code{mysqld} with the @code{--core-file} option. This core file can be used to make a backtrace that may help you find out why @code{mysqld} died: @example shell> gdb mysqld core gdb> backtrace full gdb> exit @end example @xref{Crashing}. If you are using gdb 4.17.x or above on Linux, you should install a @file{.gdb} file, with the following information, in your current directory: @example set print sevenbit off handle SIGUSR1 nostop noprint handle SIGUSR2 nostop noprint handle SIGWAITING nostop noprint handle SIGLWP nostop noprint handle SIGPIPE nostop handle SIGALRM nostop handle SIGHUP nostop handle SIGTERM nostop noprint @end example If you have problems debugging threads with gdb, you should download gdb 5.x and try this instead. The new gdb version has very improved thread handling! Here is an example how to debug mysqld: @example shell> gdb /usr/local/libexec/mysqld gdb> run ... backtrace full # Do this when mysqld crashes @end example Include the above output in a mail generated with @code{mysqlbug} and mail this to @code{mysql@@lists.mysql.com}. If @code{mysqld} hangs you can try to use some system tools like @code{strace} or @code{/usr/proc/bin/pstack} to examine where @code{mysqld} has hung. @example strace /tmp/log libexec/mysqld @end example @findex DBI->trace @findex trace DBI method @tindex DBI_TRACE environment variable @tindex Environment variable, DBI_TRACE If you are using the Perl @code{DBI} interface, you can turn on debugging information by using the @code{trace} method or by setting the @code{DBI_TRACE} environment variable. @xref{Perl DBI Class, , Perl @code{DBI} Class}. @node Using stack trace, Using log files, Using gdb on mysqld, Debugging server @appendixsubsec Using a stack trace On some operating systems, the error log will contain a stack trace if @code{mysqld} dies unexpectedly. You can use this to find out where (and maybe why) @code{mysqld} died. @xref{Error log}. To get a stack trace, you should NOT compile @code{mysqld} with the @code{-fomit-frame-pointer} option to gcc. @xref{Compiling for debugging}. If the error file contains something like the following: @example mysqld got signal 11; The manual section 'Debugging a MySQL server' tells you how to use a stack trace and/or the core file to produce a readable backtrace that may help in finding out why mysqld died Attemping backtrace. You can use the following information to find out where mysqld died. If you see no messages after this, something went terribly wrong stack range sanity check, ok, backtrace follows 0x40077552 0x81281a0 0x8128f47 0x8127be0 0x8127995 0x8104947 0x80ff28f 0x810131b 0x80ee4bc 0x80c3c91 0x80c6b43 0x80c1fd9 0x80c1686 @end example you can find where @code{mysqld} died by doing the following: @enumerate @item Copy the above numbers to a file, for example @file{mysqld.stack}. @item Make a symbol file for the @code{mysqld} server: @example nm -n libexec/mysqld > /tmp/mysqld.sym @end example Note that many MySQL binary distributions comes with the above file, named @code{mysqld.sym.gz}. In this case you must unpack this by doing: @example gunzip < bin/mysqld.sym.gz > /tmp/mysqld.sym @end example @item Execute @code{resolve_stack_dump -s /tmp/mysqld.sym -n mysqld.stack}. This will print out where @code{mysqld} died. If this doesn't help you find out why @code{mysqld} died, you should make a bug report and include the output from the above commend with the bug report. Note however that in most cases it will not help us to just have a stack trace to find the reason for the problem. To be able to locate the bug or provide a workaround, we would in most cases need to know the query that killed @code{mysqld} and preferable a test case so that we can repeat the problem! @xref{Bug reports}. @end enumerate @node Using log files, Reproduceable test case, Using stack trace, Debugging server @appendixsubsec Using log files to find cause of errors in mysqld Note that before starting @code{mysqld} with @code{--log} you should check all your tables with @code{myisamchk}. @xref{MySQL Database Administration}. If @code{mysqld} dies or hangs, you should start @code{mysqld} with @code{--log}. When @code{mysqld} dies again, you can examine the end of the log file for the query that killed @code{mysqld}. If you are using @code{--log} without a file name, the log is stored in the database directory as 'hostname'.log In most cases it's the last query in the log file that killed @code{mysqld}, but if possible you should verify this by restarting @code{mysqld} and executing the found query from the @code{mysql} command line tools. If this works, you should also test all complicated queries that didn't complete. You can also try the command @code{EXPLAIN} on all @code{SELECT} statements that takes a long time to ensure that @code{mysqld} is using indexes properly. @xref{EXPLAIN, , @code{EXPLAIN}}. You can find the queries that take a long time to execute by starting @code{mysqld} with @code{--log-slow-queries}. @xref{Slow query log}. If you find the text @code{mysqld restarted} in the error log file (normally named @file{hostname.err}) you have probably found a query that causes @code{mysqld} to fail. If this happens you should check all your tables with @code{myisamchk} (@pxref{MySQL Database Administration}), and test the queries in the MySQL log files to see if one doesn't work. If you find such a query, try first upgrading to the newest MySQL version. If this doesn't help and you can't find anything in the @code{mysql} mail archive, you should report the bug to @email{mysql@@lists.mysql.com}. Links to mail archives are available online at the @uref{http://www.mysql.com/documentation/, MySQL documentation page}. If you have started @code{mysqld} with @code{--with-myisam-recover}, MySQL will automatically check and try to repair @code{MyISAM} tables if they are marked as 'not closed properly' or 'crashed'. If this happens, MySQL will write an entry in the @code{hostname.err} file @code{'Warning: Checking table ...'} which is followed by @code{Warning: Repairing table} if the table needs to be repaired. If you get a lot of these errors, without @code{mysqld} having died unexpectedly just before, then something is wrong and needs to be investigated further. @xref{Command-line options}. It's of course not a good sign if @code{mysqld} did died unexpectedly, but in this case one shouldn't investigate the @code{Checking table...} messages but instead try to find out why @code{mysqld} died. @node Reproduceable test case, , Using log files, Debugging server @appendixsubsec Making a test case when you experience table corruption If you get corrupted tables or if @code{mysqld} always fails after some update commands, you can test if this bug is reproducible by doing the following: @itemize @bullet @item Take down the MySQL daemon (with @code{mysqladmin shutdown}). @item Make a backup of the tables (to guard against the very unlikely case that the repair will do something bad). @item Check all tables with @code{myisamchk -s database/*.MYI}. Repair any wrong tables with @code{myisamchk -r database/table.MYI}. @item Make a second backup of the tables. @item Remove (or move away) any old log files from the MySQL data directory if you need more space. @item Start @code{mysqld} with @code{--log-bin}. @xref{Binary log}. If you want to find a query that crashes @code{mysqld}, you should use @code{--log --log-bin}. @item When you have gotten a crashed table, stop the @code{mysqld server}. @item Restore the backup. @item Restart the @code{mysqld} server @strong{without} @code{--log-bin} @item Re-execute the commands with @code{mysqlbinlog update-log-file | mysql}. The update log is saved in the MySQL database directory with the name @code{hostname-bin.#}. @item If the tables are corrupted again or you can get @code{mysqld} to die with the above command, you have found reproducible bug that should be easy to fix! FTP the tables and the binary log to @uref{ftp://support.mysql.com/pub/mysql/secret} and send a mail to @email{bugs@@lists.mysql.com} or (if you are a support customer) to @email{support@@mysql.com} about the problem and the MySQL team will fix it as soon as possible. @end itemize You can also use the script @code{mysql_find_rows} to just execute some of the update statements if you want to narrow down the problem. @node Debugging client, The DBUG package, Debugging server, Porting @appendixsec Debugging a MySQL client @cindex debugging, client @cindex clients, debugging To be able to debug a MySQL client with the integrated debug package, you should configure MySQL with @code{--with-debug}. @xref{configure options}. @tindex MYSQL_DEBUG environment variable @tindex Environment variable, MYSQL_DEBUG Before running a client, you should set the @code{MYSQL_DEBUG} environment variable: @example shell> MYSQL_DEBUG=d:t:O,/tmp/client.trace shell> export MYSQL_DEBUG @end example This causes clients to generate a trace file in @file{/tmp/client.trace}. If you have problems with your own client code, you should attempt to connect to the server and run your query using a client that is known to work. Do this by running @code{mysql} in debugging mode (assuming you have compiled MySQL with debugging on): @example shell> mysql --debug=d:t:O,/tmp/client.trace @end example This will provide useful information in case you mail a bug report. @xref{Bug reports}. If your client crashes at some 'legal' looking code, you should check that your @file{mysql.h} include file matches your mysql library file. A very common mistake is to use an old @file{mysql.h} file from an old MySQL installation with new MySQL library. @node The DBUG package, Locking methods, Debugging client, Porting @appendixsec The DBUG package. @cindex DBUG package The MySQL server and most MySQL clients are compiled with the DBUG package originally made by Fred Fish. When one has configured MySQL for debugging, this package makes it possible to get a trace file of what the program is debugging. @xref{Making trace files}. One uses the debug package by invoking the program with the @code{--debug="..."} or the @code{-#...} option. Most MySQL programs has a default debug string that will be used if you don't specify an option to @code{--debug}. The default trace file is usually @code{/tmp/programname.trace} on Unix and @code{\programname.trace} on Windows. The debug control string is a sequence of colon separated fields as follows: @example <field_1>:<field_2>:...:<field_N> @end example Each field consists of a mandatory flag character followed by an optional "," and comma-separated list of modifiers: @example flag[,modifier,modifier,...,modifier] @end example The currently recognized flag characters are: @multitable @columnfractions .1 .9 @item d @tab Enable output from DBUG_<N> macros for the current state. May be followed by a list of keywords which selects output only for the DBUG macros with that keyword. An empty list of keywords implies output for all macros. @item D @tab Delay after each debugger output line. The argument is the number of tenths of seconds to delay, subject to machine capabilities. That is, @code{-#D,20} is delay two seconds. @item f @tab Limit debugging and/or tracing, and profiling to the list of named functions. Note that a null list will disable all functions. The appropriate "d" or "t" flags must still be given, this flag only limits their actions if they are enabled. @item F @tab Identify the source file name for each line of debug or trace output. @item i @tab Identify the process with the pid or thread id for each line of debug or trace output. @item g @tab Enable profiling. Create a file called 'dbugmon.out' containing information that can be used to profile the program. May be followed by a list of keywords that select profiling only for the functions in that list. A null list implies that all functions are considered. @item L @tab Identify the source file line number for each line of debug or trace output. @item n @tab Print the current function nesting depth for each line of debug or trace output. @item N @tab Number each line of dbug output. @item o @tab Redirect the debugger output stream to the specified file. The default output is stderr. @item O @tab As @code{O} but the file is really flushed between each write. When needed the file is closed and reopened between each write. @item p @tab Limit debugger actions to specified processes. A process must be identified with the DBUG_PROCESS macro and match one in the list for debugger actions to occur. @item P @tab Print the current process name for each line of debug or trace output. @item r @tab When pushing a new state, do not inherit the previous state's function nesting level. Useful when the output is to start at the left margin. @item S @tab Do function _sanity(_file_,_line_) at each debugged function until _sanity() returns something that differs from 0. (Mostly used with safemalloc to find memory leaks) @item t @tab Enable function call/exit trace lines. May be followed by a list (containing only one modifier) giving a numeric maximum trace level, beyond which no output will occur for either debugging or tracing macros. The default is a compile time option. @end multitable Some examples of debug control strings which might appear on a shell command line (the "-#" is typically used to introduce a control string to an application program) are: @example -#d:t -#d:f,main,subr1:F:L:t,20 -#d,input,output,files:n -#d:t:i:O,\\mysqld.trace @end example In MySQL, common tags to print (with the @code{d} option) are: @code{enter},@code{exit},@code{error},@code{warning},@code{info} and @code{loop}. @node Locking methods, RTS-threads, The DBUG package, Porting @appendixsec Locking methods @cindex locking methods @cindex methods, locking Currently MySQL only supports table locking for @code{ISAM}/@code{MyISAM} and @code{HEAP} tables and page level locking for @code{BDB} tables. @xref{Internal locking}. With @code{MyISAM} tables one can freely mix @code{INSERT} and @code{SELECT} without locks (@code{Versioning}). Starting in version 3.23.33, you can analyze the table lock contention on your system by checking @code{Table_locks_waited} and @code{Table_locks_immediate} environment variables. Some database users claim that MySQL cannot support near the number of concurrent users because it lacks row-level locking. This may be true for some specific applications, but is not generally true. As always this depends totally on what the application does and what is the access/update pattern of the data. Pros for row locking: @itemize @bullet @item Fewer lock conflicts when accessing different rows in many threads. @item Less changes for rollbacks. @item Makes it possible to lock a single row a long time. @end itemize Cons: @itemize @bullet @item Takes more memory than page level or table locks. @item Is slower than page level or table locks when used one a big part of the table, because one has to do many more locks. @item Is definitely much worse than other locks if you do often do @code{GROUP BY} on a large part of the data or if one has to often scan the whole table. @item With higher level locks one can also more easily support locks of different types to tune the application as the lock overhead is less notable as for row level locks. @end itemize Table locks are superior to page level / row level locks in the following cases: @itemize @bullet @item Mostly reads @item Read and updates on strict keys; This is where one updates or deletes a row that can be fetched with one key read: @example UPDATE table_name SET column=value WHERE unique_key# DELETE FROM table_name WHERE unique_key=# @end example @item @code{SELECT} combined with @code{INSERT} (and very few @code{UPDATE}'s and @code{DELETE}'s. @item Many scans / @code{GROUP BY} on the whole table without any writers. @end itemize Other options than row / page level locking: Versioning (like we use in MySQL for concurrent inserts) where you can have one writer at the same time as many readers. This means that the database/table supports different views for the data depending on when one started to access it. Other names for this are time travel, copy on write or copy on demand. Copy on demand is in many case much better than page or row level locking; The worst case does, however, use much more memory than when using normal locks. Instead of using row level locks one can use application level locks. (Like get_lock/release_lock in MySQL). This works of course only in well-behaved applications. In many cases one can do an educated guess which locking type is best for the application but generally it's very hard to say that a given lock type is better than another; Everything depends on the application and different part of the application may require different lock types. Here are some tips about locking in MySQL: On web application most applications do lots of selects, very few deletes, updates mainly on keys and inserts in some specific tables. The base MySQL setup is VERY tuned for this. Concurrent users is not a problem if one doesn't mix updates and selects that needs to examine many rows in the same table. If one mixes inserts and deletes on the same table then @code{INSERT DELAYED} may be of great help. One can also use @code{LOCK TABLES} to speed up things (many updates within a single lock is much faster than updates without locks). Splitting thing to different tables will also helps. If you get speed problems with the table locks in MySQL, you may be able to solve these to convert some of your tables to @code{BDB} tables. @xref{BDB}. The optimization section in the manual covers a lot of different aspects of how to tune ones application. @xref{Tips}. @node RTS-threads, Thread packages, Locking methods, Porting @appendixsec Comments about RTS threads @cindex RTS-threads @cindex threads, RTS I have tried to use the RTS thread packages with MySQL but stumbled on the following problems: They use an old version of a lot of POSIX calls and it is very tedious to make wrappers for all functions. I am inclined to think that it would be easier to change the thread libraries to the newest POSIX specification. Some wrappers are already written. See @file{mysys/my_pthread.c} for more info. At least the following should be changed: @code{pthread_get_specific} should use one argument. @code{sigwait} should take two arguments. A lot of functions (at least @code{pthread_cond_wait}, @code{pthread_cond_timedwait}) should return the error code on error. Now they return -1 and set @code{errno}. Another problem is that user-level threads use the @code{ALRM} signal and this aborts a lot of functions (@code{read}, @code{write}, @code{open}...). MySQL should do a retry on interrupt on all of these but it is not that easy to verify it. The biggest unsolved problem is the following: To get thread-level alarms I changed @file{mysys/thr_alarm.c} to wait between alarms with @code{pthread_cond_timedwait()}, but this aborts with error @code{EINTR}. I tried to debug the thread library as to why this happens, but couldn't find any easy solution. If someone wants to try MySQL with RTS threads I suggest the following: @itemize @bullet @item Change functions MySQL uses from the thread library to POSIX. This shouldn't take that long. @item Compile all libraries with the @code{-DHAVE_rts_threads}. @item Compile @code{thr_alarm}. @item If there are some small differences in the implementation, they may be fixed by changing @file{my_pthread.h} and @file{my_pthread.c}. @item Run @code{thr_alarm}. If it runs without any ``warning'', ``error'' or aborted messages, you are on the right track. Here is a successful run on Solaris: @example Main thread: 1 Thread 0 (5) started Thread: 5 Waiting process_alarm Thread 1 (6) started Thread: 6 Waiting process_alarm process_alarm thread_alarm Thread: 6 Slept for 1 (1) sec Thread: 6 Waiting process_alarm process_alarm thread_alarm Thread: 6 Slept for 2 (2) sec Thread: 6 Simulation of no alarm needed Thread: 6 Slept for 0 (3) sec Thread: 6 Waiting process_alarm process_alarm thread_alarm Thread: 6 Slept for 4 (4) sec Thread: 6 Waiting process_alarm thread_alarm Thread: 5 Slept for 10 (10) sec Thread: 5 Waiting process_alarm process_alarm thread_alarm Thread: 6 Slept for 5 (5) sec Thread: 6 Waiting process_alarm process_alarm ... thread_alarm Thread: 5 Slept for 0 (1) sec end @end example @end itemize @node Thread packages, , RTS-threads, Porting @appendixsec Differences between different thread packages @cindex thread packages, differences between MySQL is very dependent on the thread package used. So when choosing a good platform for MySQL, the thread package is very important. There are at least three types of thread packages: @itemize @bullet @item User threads in a single process. Thread switching is managed with alarms and the threads library manages all non-thread-safe functions with locks. Read, write and select operations are usually managed with a thread-specific select that switches to another thread if the running threads have to wait for data. If the user thread packages are integrated in the standard libs (FreeBSD and BSDI threads) the thread package requires less overhead than thread packages that have to map all unsafe calls (MIT-pthreads, FSU Pthreads and RTS threads). In some environments (for example, SCO), all system calls are thread safe so the mapping can be done very easily (FSU Pthreads on SCO). Downside: All mapped calls take a little time and it's quite tricky to be able to handle all situations. There are usually also some system calls that are not handled by the thread package (like MIT-pthreads and sockets). Thread scheduling isn't always optimal. @item User threads in separate processes. Thread switching is done by the kernel and all data are shared between threads. The thread package manages the standard thread calls to allow sharing data between threads. LinuxThreads is using this method. Downside: Lots of processes. Thread creating is slow. If one thread dies the rest are usually left hanging and you must kill them all before restarting. Thread switching is somewhat expensive. @item Kernel threads. Thread switching is handled by the thread library or the kernel and is very fast. Everything is done in one process, but on some systems, @code{ps} may show the different threads. If one thread aborts, the whole process aborts. Most system calls are thread safe and should require very little overhead. Solaris, HP-UX, AIX and OSF1 have kernel threads. @end itemize In some systems kernel threads are managed by integrating user level threads in the system libraries. In such cases, the thread switching can only be done by the thread library and the kernel isn't really ``thread aware''. @node Environment variables, Regexp, Porting, Top @appendix Environment Variables @cindex environment variables, list of Here is a list of all the environment variables that are used directly or indirectly by MySQL. Most of these can also be found in other places in this manual. Note that any options on the command line will take precedence over values specified in configuration files and environment variables, and values in configuration files take precedence over values in environment variables. In many cases it's preferable to use a configure file instead of environment variables to modify the behavior of MySQL. @xref{Option files}. @tindex CCX environment variable @tindex Environment variable, CCX @tindex CC environment variable @tindex Environment variable, CC @tindex CFLAGS environment variable @tindex Environment variable, CFLAGS @tindex CXXFLAGS environment variable @tindex Environment variable, CXXFLAGS @tindex DBI_USER environment variable @tindex Environment variable, DBI_USER @tindex DBI_TRACE environment variable @tindex Environment variable, DBI_TRACE @tindex HOME environment variable @tindex Environment variable, HOME @tindex LD_RUN_PATH environment variable @tindex Environment variable, LD_RUN_PATH @tindex MYSQL_DEBUG environment variable @tindex Environment variable, MYSQL_DEBUG @tindex MYSQL_HISTFILE environment variable @tindex Environment variable, MYSQL_HISTFILE @tindex MYSQL_HOST environment variable @tindex Environment variable, MYSQL_HOST @tindex MYSQL_PWD environment variable @tindex Environment variable, MYSQL_PWD @tindex MYSQL_TCP_PORT environment variable @tindex Environment variable, MYSQL_TCP_PORT @tindex MYSQL_UNIX_PORT environment variable @tindex Environment variable, MYSQL_UNIX_PORT @tindex PATH environment variable @tindex Environment variable, PATH @tindex TMPDIR environment variable @tindex Environment variable, TMPDIR @tindex TZ environment variable @tindex Environment variable, TZ @tindex UMASK_DIR environment variable @tindex Environment variable, UMASK_DIR @tindex UMASK environment variable @tindex Environment variable, UMASK @tindex USER environment variable @tindex Environment variable, USER @multitable @columnfractions .2 .8 @item @code{CCX} @tab Set this to your C++ compiler when running configure. @item @code{CC} @tab Set this to your C compiler when running configure. @item @code{CFLAGS} @tab Flags for your C compiler when running configure. @item @code{CXXFLAGS} @tab Flags for your C++ compiler when running configure. @item @code{DBI_USER} @tab The default user name for Perl DBI. @item @code{DBI_TRACE} @tab Used when tracing Perl DBI. @item @code{HOME} @tab The default path for the @code{mysql} history file is @file{$HOME/.mysql_history}. @item @code{LD_RUN_PATH} @tab Used to specify where your @code{libmysqlclient.so} is. @item @code{MYSQL_DEBUG} @tab Debug-trace options when debugging. @item @code{MYSQL_HISTFILE} @tab The path to the @code{mysql} history file. @item @code{MYSQL_HOST} @tab Default host name used by the @code{mysql} command-line prompt. @item @code{MYSQL_PWD} @tab The default password when connecting to @code{mysqld}. Note that use of this is insecure! @item @code{MYSQL_TCP_PORT} @tab The default TCP/IP port. @item @code{MYSQL_UNIX_PORT} @tab The default socket; used for connections to @code{localhost}. @item @code{PATH} @tab Used by the shell to finds the MySQL programs. @item @code{TMPDIR} @tab The directory where temporary tables/files are created. @item @code{TZ} @tab This should be set to your local time zone. @xref{Timezone problems}. @item @code{UMASK_DIR} @tab The user-directory creation mask when creating directories. Note that this is ANDed with @code{UMASK}! @item @code{UMASK} @tab The user-file creation mask when creating files. @item @code{USER} @tab The default user on Windows to use when connecting to @code{mysqld}. @end multitable @node Regexp, Unireg, Environment variables, Top @appendix Description of MySQL regular expression syntax @cindex regex @cindex regular expression syntax, described @cindex syntax, regular expression A regular expression (regex) is a powerful way of specifying a complex search. MySQL uses Henry Spencer's implementation of regular expressions, which is aimed at conformance with POSIX 1003.2. MySQL uses the extended version. This is a simplistic reference that skips the details. To get more exact information, see Henry Spencer's @code{regex(7)} manual page that is included in the source distribution. @xref{Credits}. A regular expression describes a set of strings. The simplest regexp is one that has no special characters in it. For example, the regexp @code{hello} matches @code{hello} and nothing else. Non-trivial regular expressions use certain special constructs so that they can match more than one string. For example, the regexp @code{hello|word} matches either the string @code{hello} or the string @code{word}. As a more complex example, the regexp @code{B[an]*s} matches any of the strings @code{Bananas}, @code{Baaaaas}, @code{Bs}, and any other string starting with a @code{B}, ending with an @code{s}, and containing any number of @code{a} or @code{n} characters in between. A regular expression may use any of the following special characters/constructs: @table @code @item ^ Match the beginning of a string. @example mysql> select "fo\nfo" REGEXP "^fo$"; -> 0 mysql> select "fofo" REGEXP "^fo"; -> 1 @end example @item $ Match the end of a string. @example mysql> select "fo\no" REGEXP "^fo\no$"; -> 1 mysql> select "fo\no" REGEXP "^fo$"; -> 0 @end example @item . Match any character (including newline). @example mysql> select "fofo" REGEXP "^f.*"; -> 1 mysql> select "fo\nfo" REGEXP "^f.*"; -> 1 @end example @item a* Match any sequence of zero or more @code{a} characters. @example mysql> select "Ban" REGEXP "^Ba*n"; -> 1 mysql> select "Baaan" REGEXP "^Ba*n"; -> 1 mysql> select "Bn" REGEXP "^Ba*n"; -> 1 @end example @item a+ Match any sequence of one or more @code{a} characters. @example mysql> select "Ban" REGEXP "^Ba+n"; -> 1 mysql> select "Bn" REGEXP "^Ba+n"; -> 0 @end example @item a? Match either zero or one @code{a} character. @example mysql> select "Bn" REGEXP "^Ba?n"; -> 1 mysql> select "Ban" REGEXP "^Ba?n"; -> 1 mysql> select "Baan" REGEXP "^Ba?n"; -> 0 @end example @item de|abc Match either of the sequences @code{de} or @code{abc}. @example mysql> select "pi" REGEXP "pi|apa"; -> 1 mysql> select "axe" REGEXP "pi|apa"; -> 0 mysql> select "apa" REGEXP "pi|apa"; -> 1 mysql> select "apa" REGEXP "^(pi|apa)$"; -> 1 mysql> select "pi" REGEXP "^(pi|apa)$"; -> 1 mysql> select "pix" REGEXP "^(pi|apa)$"; -> 0 @end example @item (abc)* Match zero or more instances of the sequence @code{abc}. @example mysql> select "pi" REGEXP "^(pi)*$"; -> 1 mysql> select "pip" REGEXP "^(pi)*$"; -> 0 mysql> select "pipi" REGEXP "^(pi)*$"; -> 1 @end example @item @{1@} @itemx @{2,3@} The is a more general way of writing regexps that match many occurrences of the previous atom. @table @code @item a* Can be written as @code{a@{0,@}}. @item a+ Can be written as @code{a@{1,@}}. @item a? Can be written as @code{a@{0,1@}}. @end table To be more precise, an atom followed by a bound containing one integer @code{i} and no comma matches a sequence of exactly @code{i} matches of the atom. An atom followed by a bound containing one integer @code{i} and a comma matches a sequence of @code{i} or more matches of the atom. An atom followed by a bound containing two integers @code{i} and @code{j} matches a sequence of @code{i} through @code{j} (inclusive) matches of the atom. Both arguments must be in the range from @code{0} to @code{RE_DUP_MAX} (default 255), inclusive. If there are two arguments, the second must be greater than or equal to the first. @item [a-dX] @itemx [^a-dX] Matches any character which is (or is not, if ^ is used) either @code{a}, @code{b}, @code{c}, @code{d} or @code{X}. To include a literal @code{]} character, it must immediately follow the opening bracket @code{[}. To include a literal @code{-} character, it must be written first or last. So @code{[0-9]} matches any decimal digit. Any character that does not have a defined meaning inside a @code{[]} pair has no special meaning and matches only itself. @example mysql> select "aXbc" REGEXP "[a-dXYZ]"; -> 1 mysql> select "aXbc" REGEXP "^[a-dXYZ]$"; -> 0 mysql> select "aXbc" REGEXP "^[a-dXYZ]+$"; -> 1 mysql> select "aXbc" REGEXP "^[^a-dXYZ]+$"; -> 0 mysql> select "gheis" REGEXP "^[^a-dXYZ]+$"; -> 1 mysql> select "gheisa" REGEXP "^[^a-dXYZ]+$"; -> 0 @end example @item [[.characters.]] The sequence of characters of that collating element. The sequence is a single element of the bracket expression's list. A bracket expression containing a multi-character collating element can thus match more than one character, for example, if the collating sequence includes a @code{ch} collating element, then the regular expression @code{[[.ch.]]*c} matches the first five characters of @code{chchcc}. @item [=character_class=] An equivalence class, standing for the sequences of characters of all collating elements equivalent to that one, including itself. For example, if @code{o} and @code{(+)} are the members of an equivalence class, then @code{[[=o=]]}, @code{[[=(+)=]]}, and @code{[o(+)]} are all synonymous. An equivalence class may not be an endpoint of a range. @item [:character_class:] Within a bracket expression, the name of a character class enclosed in @code{[:} and @code{:]} stands for the list of all characters belonging to that class. Standard character class names are: @multitable @columnfractions .33 .33 .33 @item alnum @tab digit @tab punct @item alpha @tab graph @tab space @item blank @tab lower @tab upper @item cntrl @tab print @tab xdigit @end multitable These stand for the character classes defined in the @code{ctype(3)} manual page. A locale may provide others. A character class may not be used as an endpoint of a range. @example mysql> select "justalnums" REGEXP "[[:alnum:]]+"; -> 1 mysql> select "!!" REGEXP "[[:alnum:]]+"; -> 0 @end example @item [[:<:]] @itemx [[:>:]] These match the null string at the beginning and end of a word respectively. A word is defined as a sequence of word characters which is neither preceded nor followed by word characters. A word character is an alnum character (as defined by @code{ctype(3)}) or an underscore (@code{_}). @example mysql> select "a word a" REGEXP "[[:<:]]word[[:>:]]"; -> 1 mysql> select "a xword a" REGEXP "[[:<:]]word[[:>:]]"; -> 0 @end example @end table @example mysql> select "weeknights" REGEXP "^(wee|week)(knights|nights)$"; -> 1 @end example @node Unireg, GPL license, Regexp, Top @appendix What is Unireg? @cindex Unireg, described @cindex interface builder Unireg is our tty interface builder, but it uses a low-level connection to our ISAM (which is used by MySQL) and because of this it is very quick. It has existed since 1979 (on Unix in C since ~1986). Unireg has the following components: @itemize @bullet @item One table viewer with updates/browsing. @item Multi table viewer (with one scrolling region). @item Table creator. (With lots of column tags you can't create with MySQL) This is WYSIWYG (for a tty). You design a screen and Unireg prompts for the column specification. @item Report generator. @item A lot of utilities (quick import/export of tables to/from text files, analysis of table contents...). @item Powerful multi-table updates (which we use a lot) with a BASIC-like language with LOTS of functions. @item Dynamic languages (at present in Swedish and Finnish). If somebody wants an English version there are a few files that would have to be translated. @item The ability to run updates interactively or in a batch. @item Emacs-like key definitions with keyboard macros. @item All this in a binary of 800K. @item The @code{convform} utility. Converts @file{.frm} and text files between different character sets. @item The @code{myisampack} utility. Packs an ISAM table (makes it 50-80% smaller). The table can be read by MySQL like an ordinary table. Only one record has to be decompressed per access. Cannot handle @code{BLOB} or @code{TEXT} columns or updates (yet). @end itemize We update most of our production databases with the Unireg interface and serve web pages through MySQL (and in some extreme cases the Unireg report generator). Unireg takes about 3M of disk space and works on at least the following platforms: SunOS 4.x, Solaris, Linux, HP-UX, ICL Unix, DNIX, SCO and MS-DOS. Unireg is currently only available in Swedish and Finnish. The price tag for Unireg is 10,000 Swedish kr (about $1500 US), but this includes support. Unireg is distributed as a binary. (But all the ISAM sources can be found in MySQL). Usually we compile the binary for the customer at their site. All new development is concentrated to MySQL. @page @c This node name is special @node GPL license, LGPL license, Unireg, Top @appendix GNU GENERAL PUBLIC LICENSE @cindex GPL, General Public License @cindex GPL, GNU General Public License @center Version 2, June 1991 @display Copyright @copyright{} 1989, 1991 Free Software Foundation, Inc. 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. @end display @appendixsec Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software---to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. @iftex @appendixsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION @end iftex @ifinfo @center GNU GENERAL PUBLIC LICENSE @center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION @end ifinfo @enumerate 0 @item This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The ``Program'', below, refers to any such program or work, and a ``work based on the Program'' means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term ``modification''.) Each licensee is addressed as ``you''. Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does. @item You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. @item You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: @enumerate a @item You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. @item You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. @item If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) @end enumerate These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. @item You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: @enumerate a @item Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, @item Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, @item Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) @end enumerate The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. @item You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. @item You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. @item Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. @item If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. @item If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. @item The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and ``any later version'', you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. @item If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. @iftex @heading NO WARRANTY @end iftex @ifinfo @center NO WARRANTY @end ifinfo @item BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. @item IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. @end enumerate @iftex @heading END OF TERMS AND CONDITIONS @end iftex @ifinfo @center END OF TERMS AND CONDITIONS @end ifinfo @page @appendixsec How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the ``copyright'' line and a pointer to where the full notice is found. @smallexample @var{one line to give the program's name and a brief idea of what it does.} Copyright (C) @var{yyyy} @var{name of author} This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. @end smallexample Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: @smallexample Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author} Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. @end smallexample The hypothetical commands @samp{show w} and @samp{show c} should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than @samp{show w} and @samp{show c}; they could even be mouse-clicks or menu items---whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a ``copyright disclaimer'' for the program, if necessary. Here is a sample; alter the names: @example Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomovision' (which makes passes at compilers) written by James Hacker. @var{signature of Ty Coon}, 1 April 1989 Ty Coon, President of Vice @end example This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License. @page @node LGPL license, Placeholder, GPL license, Top @appendix GNU LESSER GENERAL PUBLIC LICENSE @cindex LGPL, Lesser General Public License @cindex LGPL, GNU Library General Public License @center Version 2.1, February 1999 @display Copyright @copyright{} 1991, 1999 Free Software Foundation, Inc. 59 Temple Place -- Suite 330, Boston, MA 02111-1307, USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. [This is the first released version of the Lesser GPL. It also counts as the successor of the GNU Library Public License, version 2, hence the version number 2.1.] @end display @appendixsec Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public Licenses are intended to guarantee your freedom to share and change free software---to make sure the software is free for all its users. This license, the Lesser General Public License, applies to some specially designated software---typically libraries---of the Free Software Foundation and other authors who decide to use it. You can use it too, but we suggest you first think carefully about whether this license or the ordinary General Public License is the better strategy to use in any particular case, based on the explanations below. When we speak of free software, we are referring to freedom of use, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish); that you receive source code or can get it if you want it; that you can change the software and use pieces of it in new free programs; and that you are informed that you can do these things. To protect your rights, we need to make restrictions that forbid distributors to deny you these rights or to ask you to surrender these rights. These restrictions translate to certain responsibilities for you if you distribute copies of the library or if you modify it. For example, if you distribute copies of the library, whether gratis or for a fee, you must give the recipients all the rights that we gave you. You must make sure that they, too, receive or can get the source code. If you link other code with the library, you must provide complete object files to the recipients, so that they can relink them with the library after making changes to the library and recompiling it. And you must show them these terms so they know their rights. We protect your rights with a two-step method: (1) we copyright the library, and (2) we offer you this license, which gives you legal permission to copy, distribute and/or modify the library. To protect each distributor, we want to make it very clear that there is no warranty for the free library. Also, if the library is modified by someone else and passed on, the recipients should know that what they have is not the original version, so that the original author's reputation will not be affected by problems that might be introduced by others. Finally, software patents pose a constant threat to the existence of any free program. We wish to make sure that a company cannot effectively restrict the users of a free program by obtaining a restrictive license from a patent holder. Therefore, we insist that any patent license obtained for a version of the library must be consistent with the full freedom of use specified in this license. Most GNU software, including some libraries, is covered by the ordinary GNU General Public License. This license, the GNU Lesser General Public License, applies to certain designated libraries, and is quite different from the ordinary General Public License. We use this license for certain libraries in order to permit linking those libraries into non-free programs. When a program is linked with a library, whether statically or using a shared library, the combination of the two is legally speaking a combined work, a derivative of the original library. The ordinary General Public License therefore permits such linking only if the entire combination fits its criteria of freedom. The Lesser General Public License permits more lax criteria for linking other code with the library. We call this license the @dfn{Lesser} General Public License because it does @emph{Less} to protect the user's freedom than the ordinary General Public License. It also provides other free software developers Less of an advantage over competing non-free programs. These disadvantages are the reason we use the ordinary General Public License for many libraries. However, the Lesser license provides advantages in certain special circumstances. For example, on rare occasions, there may be a special need to encourage the widest possible use of a certain library, so that it becomes a de-facto standard. To achieve this, non-free programs must be allowed to use the library. A more frequent case is that a free library does the same job as widely used non-free libraries. In this case, there is little to gain by limiting the free library to free software only, so we use the Lesser General Public License. In other cases, permission to use a particular library in non-free programs enables a greater number of people to use a large body of free software. For example, permission to use the GNU C Library in non-free programs enables many more people to use the whole GNU operating system, as well as its variant, the GNU/Linux operating system. Although the Lesser General Public License is Less protective of the users' freedom, it does ensure that the user of a program that is linked with the Library has the freedom and the wherewithal to run that program using a modified version of the Library. The precise terms and conditions for copying, distribution and modification follow. Pay close attention to the difference between a ``work based on the library'' and a ``work that uses the library''. The former contains code derived from the library, whereas the latter must be combined with the library in order to run. @iftex @appendixsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION @end iftex @ifinfo @center GNU LESSER GENERAL PUBLIC LICENSE @center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION @end ifinfo @enumerate 0 @item This License Agreement applies to any software library or other program which contains a notice placed by the copyright holder or other authorized party saying it may be distributed under the terms of this Lesser General Public License (also called ``this License''). Each licensee is addressed as ``you''. A ``library'' means a collection of software functions and/or data prepared so as to be conveniently linked with application programs (which use some of those functions and data) to form executables. The ``Library'', below, refers to any such software library or work which has been distributed under these terms. A ``work based on the Library'' means either the Library or any derivative work under copyright law: that is to say, a work containing the Library or a portion of it, either verbatim or with modifications and/or translated straightforwardly into another language. (Hereinafter, translation is included without limitation in the term ``modification''.) ``Source code'' for a work means the preferred form of the work for making modifications to it. For a library, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the library. Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running a program using the Library is not restricted, and output from such a program is covered only if its contents constitute a work based on the Library (independent of the use of the Library in a tool for writing it). Whether that is true depends on what the Library does and what the program that uses the Library does. @item You may copy and distribute verbatim copies of the Library's complete source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and distribute a copy of this License along with the Library. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. @item You may modify your copy or copies of the Library or any portion of it, thus forming a work based on the Library, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: @enumerate a @item The modified work must itself be a software library. @item You must cause the files modified to carry prominent notices stating that you changed the files and the date of any change. @item You must cause the whole of the work to be licensed at no charge to all third parties under the terms of this License. @item If a facility in the modified Library refers to a function or a table of data to be supplied by an application program that uses the facility, other than as an argument passed when the facility is invoked, then you must make a good faith effort to ensure that, in the event an application does not supply such function or table, the facility still operates, and performs whatever part of its purpose remains meaningful. (For example, a function in a library to compute square roots has a purpose that is entirely well-defined independent of the application. Therefore, Subsection 2d requires that any application-supplied function or table used by this function must be optional: if the application does not supply it, the square root function must still compute square roots.) @end enumerate These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Library, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Library, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Library. In addition, mere aggregation of another work not based on the Library with the Library (or with a work based on the Library) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. @item You may opt to apply the terms of the ordinary GNU General Public License instead of this License to a given copy of the Library. To do this, you must alter all the notices that refer to this License, so that they refer to the ordinary GNU General Public License, version 2, instead of to this License. (If a newer version than version 2 of the ordinary GNU General Public License has appeared, then you can specify that version instead if you wish.) Do not make any other change in these notices. Once this change is made in a given copy, it is irreversible for that copy, so the ordinary GNU General Public License applies to all subsequent copies and derivative works made from that copy. This option is useful when you wish to copy part of the code of the Library into a program that is not a library. @item You may copy and distribute the Library (or a portion or derivative of it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange. If distribution of object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place satisfies the requirement to distribute the source code, even though third parties are not compelled to copy the source along with the object code. @item A program that contains no derivative of any portion of the Library, but is designed to work with the Library by being compiled or linked with it, is called a ``work that uses the Library''. Such a work, in isolation, is not a derivative work of the Library, and therefore falls outside the scope of this License. However, linking a ``work that uses the Library'' with the Library creates an executable that is a derivative of the Library (because it contains portions of the Library), rather than a ``work that uses the library''. The executable is therefore covered by this License. Section 6 states terms for distribution of such executables. When a ``work that uses the Library'' uses material from a header file that is part of the Library, the object code for the work may be a derivative work of the Library even though the source code is not. Whether this is true is especially significant if the work can be linked without the Library, or if the work is itself a library. The threshold for this to be true is not precisely defined by law. If such an object file uses only numerical parameters, data structure layouts and accessors, and small macros and small inline functions (ten lines or less in length), then the use of the object file is unrestricted, regardless of whether it is legally a derivative work. (Executables containing this object code plus portions of the Library will still fall under Section 6.) Otherwise, if the work is a derivative of the Library, you may distribute the object code for the work under the terms of Section 6. Any executables containing that work also fall under Section 6, whether or not they are linked directly with the Library itself. @item As an exception to the Sections above, you may also combine or link a ``work that uses the Library'' with the Library to produce a work containing portions of the Library, and distribute that work under terms of your choice, provided that the terms permit modification of the work for the customer's own use and reverse engineering for debugging such modifications. You must give prominent notice with each copy of the work that the Library is used in it and that the Library and its use are covered by this License. You must supply a copy of this License. If the work during execution displays copyright notices, you must include the copyright notice for the Library among them, as well as a reference directing the user to the copy of this License. Also, you must do one of these things: @enumerate a @item Accompany the work with the complete corresponding machine-readable source code for the Library including whatever changes were used in the work (which must be distributed under Sections 1 and 2 above); and, if the work is an executable linked with the Library, with the complete machine-readable ``work that uses the Library'', as object code and/or source code, so that the user can modify the Library and then relink to produce a modified executable containing the modified Library. (It is understood that the user who changes the contents of definitions files in the Library will not necessarily be able to recompile the application to use the modified definitions.) @item Use a suitable shared library mechanism for linking with the Library. A suitable mechanism is one that (1) uses at run time a copy of the library already present on the user's computer system, rather than copying library functions into the executable, and (2) will operate properly with a modified version of the library, if the user installs one, as long as the modified version is interface-compatible with the version that the work was made with. @item Accompany the work with a written offer, valid for at least three years, to give the same user the materials specified in Subsection 6a, above, for a charge no more than the cost of performing this distribution. @item If distribution of the work is made by offering access to copy from a designated place, offer equivalent access to copy the above specified materials from the same place. @item Verify that the user has already received a copy of these materials or that you have already sent this user a copy. @end enumerate For an executable, the required form of the ``work that uses the Library'' must include any data and utility programs needed for reproducing the executable from it. However, as a special exception, the materials to be distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. It may happen that this requirement contradicts the license restrictions of other proprietary libraries that do not normally accompany the operating system. Such a contradiction means you cannot use both them and the Library together in an executable that you distribute. @item You may place library facilities that are a work based on the Library side-by-side in a single library together with other library facilities not covered by this License, and distribute such a combined library, provided that the separate distribution of the work based on the Library and of the other library facilities is otherwise permitted, and provided that you do these two things: @enumerate a @item Accompany the combined library with a copy of the same work based on the Library, uncombined with any other library facilities. This must be distributed under the terms of the Sections above. @item Give prominent notice with the combined library of the fact that part of it is a work based on the Library, and explaining where to find the accompanying uncombined form of the same work. @end enumerate @item You may not copy, modify, sublicense, link with, or distribute the Library except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense, link with, or distribute the Library is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. @item You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Library or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Library (or any work based on the Library), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Library or works based on it. @item Each time you redistribute the Library (or any work based on the Library), the recipient automatically receives a license from the original licensor to copy, distribute, link with or modify the Library subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties with this License. @item If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Library at all. For example, if a patent license would not permit royalty-free redistribution of the Library by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Library. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply, and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. @item If the distribution and/or use of the Library is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Library under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. @item The Free Software Foundation may publish revised and/or new versions of the Lesser General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Library specifies a version number of this License which applies to it and ``any later version'', you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Library does not specify a license version number, you may choose any version ever published by the Free Software Foundation. @item If you wish to incorporate parts of the Library into other free programs whose distribution conditions are incompatible with these, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. @iftex @heading NO WARRANTY @end iftex @ifinfo @center NO WARRANTY @end ifinfo @item BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE LIBRARY ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. @item IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. @end enumerate @iftex @heading END OF TERMS AND CONDITIONS @end iftex @ifinfo @center END OF TERMS AND CONDITIONS @end ifinfo @page @appendixsec How to Apply These Terms to Your New Libraries If you develop a new library, and you want it to be of the greatest possible use to the public, we recommend making it free software that everyone can redistribute and change. You can do so by permitting redistribution under these terms (or, alternatively, under the terms of the ordinary General Public License). To apply these terms, attach the following notices to the library. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the ``copyright'' line and a pointer to where the full notice is found. @smallexample @var{one line to give the library's name and an idea of what it does.} Copyright (C) @var{year} @var{name of author} This library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 of the License, or (at your option) any later version. This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. You should have received a copy of the GNU Lesser General Public License along with this library; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA. @end smallexample Also add information on how to contact you by electronic and paper mail. You should also get your employer (if you work as a programmer) or your school, if any, to sign a ``copyright disclaimer'' for the library, if necessary. Here is a sample; alter the names: @smallexample Yoyodyne, Inc., hereby disclaims all copyright interest in the library `Frob' (a library for tweaking knobs) written by James Random Hacker. @var{signature of Ty Coon}, 1 April 1990 Ty Coon, President of Vice @end smallexample That's all there is to it! @node Placeholder, Function Index, LGPL license, Top @appendix Pieces of the manual in transit @menu * Installing binary:: Installing a MySQL Binary Distribution * Perl support:: Perl Installation Comments * Group by functions:: Functions for Use with @code{GROUP BY} Clauses @end menu @node Installing binary, Perl support, Placeholder, Placeholder @appendixsec Installing a MySQL Binary Distribution @cindex installing, binary distribution @cindex binary distributions, installing @menu * Linux-RPM:: Linux RPM files * Building clients:: Building client programs @end menu You need the following tools to install a MySQL binary distribution: @itemize @bullet @item GNU @code{gunzip} to uncompress the distribution. @item A reasonable @code{tar} to unpack the distribution. GNU @code{tar} is known to work. Sun @code{tar} is known to have problems. @end itemize @cindex RPM, defined @cindex RedHat Package Manager An alternative installation method under Linux is to use RPM (RedHat Package Manager) distributions. @xref{Linux-RPM}. @c texi2html fails to split chapters if I use strong for all of this. If you run into problems, @strong{PLEASE ALWAYS USE} @code{mysqlbug} when posting questions to @email{mysql@@lists.mysql.com}. Even if the problem isn't a bug, @code{mysqlbug} gathers system information that will help others solve your problem. By not using @code{mysqlbug}, you lessen the likelihood of getting a solution to your problem! You will find @code{mysqlbug} in the @file{bin} directory after you unpack the distribution. @xref{Bug reports}. @cindex commands, for binary distribution The basic commands you must execute to install and use a MySQL binary distribution are: @example shell> groupadd mysql shell> useradd -g mysql mysql shell> cd /usr/local shell> gunzip < /path/to/mysql-VERSION-OS.tar.gz | tar xvf - shell> ln -s mysql-VERSION-OS mysql shell> cd mysql shell> scripts/mysql_install_db shell> chown -R root /usr/local/mysql shell> chown -R mysql /usr/local/mysql/data shell> chgrp -R mysql /usr/local/mysql shell> chown -R root /usr/local/mysql/bin shell> bin/safe_mysqld --user=mysql & @end example @cindex adding, new users @cindex new users, adding @cindex users, adding You can add new users using the @code{bin/mysql_setpermission} script if you install the @code{DBI} and @code{Msql-Mysql-modules} Perl modules. A more detailed description follows. To install a binary distribution, follow the steps below, then proceed to @ref{Post-installation}, for post-installation setup and testing: @enumerate @item Pick the directory under which you want to unpack the distribution, and move into it. In the example below, we unpack the distribution under @file{/usr/local} and create a directory @file{/usr/local/mysql} into which MySQL is installed. (The following instructions therefore assume you have permission to create files in @file{/usr/local}. If that directory is protected, you will need to perform the installation as @code{root}.) @item Obtain a distribution file from one of the sites listed in @ref{Getting MySQL, , Getting MySQL}. MySQL binary distributions are provided as compressed @code{tar} archives and have names like @file{mysql-VERSION-OS.tar.gz}, where @code{VERSION} is a number (for example, @code{3.21.15}), and @code{OS} indicates the type of operating system for which the distribution is intended (for example, @code{pc-linux-gnu-i586}). @item If you see a binary distribution marked with the @code{-max} prefix, this means that the binary has support for transaction-safe tables and other features. @xref{mysqld-max, , @code{mysqld-max}}. Note that all binaries are built from the same MySQL source distribution. @item Add a user and group for @code{mysqld} to run as: @example shell> groupadd mysql shell> useradd -g mysql mysql @end example These commands add the @code{mysql} group and the @code{mysql} user. The syntax for @code{useradd} and @code{groupadd} may differ slightly on different versions of Unix. They may also be called @code{adduser} and @code{addgroup}. You may wish to call the user and group something else instead of @code{mysql}. @item Change into the intended installation directory: @example shell> cd /usr/local @end example @item Unpack the distribution and create the installation directory: @example shell> gunzip < /path/to/mysql-VERSION-OS.tar.gz | tar xvf - shell> ln -s mysql-VERSION-OS mysql @end example The first command creates a directory named @file{mysql-VERSION-OS}. The second command makes a symbolic link to that directory. This lets you refer more easily to the installation directory as @file{/usr/local/mysql}. @item Change into the installation directory: @example shell> cd mysql @end example You will find several files and subdirectories in the @code{mysql} directory. The most important for installation purposes are the @file{bin} and @file{scripts} subdirectories. @table @file @item bin @tindex PATH environment variable @tindex environment variable, PATH This directory contains client programs and the server You should add the full pathname of this directory to your @code{PATH} environment variable so that your shell finds the MySQL programs properly. @xref{Environment variables}. @item scripts This directory contains the @code{mysql_install_db} script used to initialize the @code{mysql} database containing the grant tables that store the server access permissions. @end table @item If you would like to use @code{mysqlaccess} and have the MySQL distribution in some non-standard place, you must change the location where @code{mysqlaccess} expects to find the @code{mysql} client. Edit the @file{bin/mysqlaccess} script at approximately line 18. Search for a line that looks like this: @example $MYSQL = '/usr/local/bin/mysql'; # path to mysql executable @end example Change the path to reflect the location where @code{mysql} actually is stored on your system. If you do not do this, you will get a @code{Broken pipe} error when you run @code{mysqlaccess}. @item Create the MySQL grant tables (necessary only if you haven't installed MySQL before): @example shell> scripts/mysql_install_db @end example Note that MySQL versions older than Version 3.22.10 started the MySQL server when you run @code{mysql_install_db}. This is no longer true! @item Change ownership of binaries to @code{root} and ownership of the data directory to the user that you will run @code{mysqld} as: @example shell> chown -R root /usr/local/mysql shell> chown -R mysql /usr/local/mysql/data shell> chgrp -R mysql /usr/local/mysql @end example The first command changes the @code{owner} attribute of the files to the @code{root} user, the second one changes the @code{owner} attribute of the data directory to the @code{mysql} user, and the third one changes the @code{group} attribute to the @code{mysql} group. @item If you want to install support for the Perl @code{DBI}/@code{DBD} interface, see @ref{Perl support}. @item If you would like MySQL to start automatically when you boot your machine, you can copy @code{support-files/mysql.server} to the location where your system has its startup files. More information can be found in the @code{support-files/mysql.server} script itself and in @ref{Automatic start}. @end enumerate After everything has been unpacked and installed, you should initialize and test your distribution. You can start the MySQL server with the following command: @example shell> bin/safe_mysqld --user=mysql & @end example @xref{safe_mysqld, , @code{safe_mysqld}}. @xref{Post-installation}. @node Perl support, Group by functions, Installing binary, Placeholder @appendixsec Perl Installation Comments @cindex Perl, installing @cindex installing, Perl @menu * Perl installation:: Installing Perl on Unix * ActiveState Perl:: Installing ActiveState Perl on Windows * Windows Perl:: Installing the MySQL Perl distribution on Windows * Perl support problems:: Problems using the Perl @code{DBI}/@code{DBD} interface @end menu @node Perl installation, ActiveState Perl, Perl support, Perl support @appendixsubsec Installing Perl on Unix Perl support for MySQL is provided by means of the @code{DBI}/@code{DBD} client interface. @xref{Perl}. The Perl @code{DBD}/@code{DBI} client code requires Perl Version 5.004 or later. The interface @strong{will not work} if you have an older version of Perl. MySQL Perl support also requires that you've installed MySQL client programming support. If you installed MySQL from RPM files, client programs are in the client RPM, but client programming support is in the developer RPM. Make sure you've installed the latter RPM. As of Version 3.22.8, Perl support is distributed separately from the main MySQL distribution. If you want to install Perl support, the files you will need can be obtained from @uref{http://www.mysql.com/Downloads/Contrib/}. The Perl distributions are provided as compressed @code{tar} archives and have names like @file{MODULE-VERSION.tar.gz}, where @code{MODULE} is the module name and @code{VERSION} is the version number. You should get the @code{Data-Dumper}, @code{DBI}, and @code{Msql-Mysql-modules} distributions and install them in that order. The installation procedure is shown below. The example shown is for the @code{Data-Dumper} module, but the procedure is the same for all three distributions: @enumerate @item Unpack the distribution into the current directory: @example shell> gunzip < Data-Dumper-VERSION.tar.gz | tar xvf - @end example This command creates a directory named @file{Data-Dumper-VERSION}. @item Change into the top-level directory of the unpacked distribution: @example shell> cd Data-Dumper-VERSION @end example @item Build the distribution and compile everything: @example shell> perl Makefile.PL shell> make shell> make test shell> make install @end example @end enumerate The @code{make test} command is important because it verifies that the module is working. Note that when you run that command during the @code{Msql-Mysql-modules} installation to exercise the interface code, the MySQL server must be running or the test will fail. It is a good idea to rebuild and reinstall the @code{Msql-Mysql-modules} distribution whenever you install a new release of MySQL, particularly if you notice symptoms such as all your @code{DBI} scripts dumping core after you upgrade MySQL. If you don't have the right to install Perl modules in the system directory or if you to install local Perl modules, the following reference may help you: @example @uref{http://www.iserver.com/support/contrib/perl5/modules.html} @end example Look under the heading @code{Installing New Modules that Require Locally Installed Modules}. @node ActiveState Perl, Windows Perl, Perl installation, Perl support @appendixsubsec Installing ActiveState Perl on Windows @cindex installing, Perl on Windows @cindex Perl, installing on Windows @cindex ActiveState Perl To install the MySQL @code{DBD} module with ActiveState Perl on Windows, you should do the following: @itemize @bullet @item Get ActiveState Perl from @uref{http://www.activestate.com/Products/ActivePerl/index.html} and install it. @item Open a DOS shell. @item If required, set the HTTP_proxy variable. For example, you might try: @example set HTTP_proxy=my.proxy.com:3128 @end example @item Start the PPM program: @example C:\> c:\perl\bin\ppm.pl @end example @item If you have not already done so, install @code{DBI}: @example ppm> install DBI @end example @item If this succeeds, run the following command: @example install ftp://ftp.de.uu.net/pub/CPAN/authors/id/JWIED/DBD-mysql-1.2212.x86.ppd @end example @end itemize The above should work at least with ActiveState Perl Version 5.6. If you can't get the above to work, you should instead install the @strong{MyODBC} driver and connect to MySQL server through ODBC: @example use DBI; $dbh= DBI->connect("DBI:ODBC:$dsn","$user","$password") || die "Got error $DBI::errstr when connecting to $dsn\n"; @end example @node Windows Perl, Perl support problems, ActiveState Perl, Perl support @appendixsubsec Installing the MySQL Perl Distribution on Windows The MySQL Perl distribution contains @code{DBI}, @code{DBD:MySQL} and @code{DBD:ODBC}. @itemize @bullet @item Get the Perl distribution for Windows from @uref{http://www.mysql.com/download.html}. @item Unzip the distribution in @code{C:} so that you get a @file{C:\PERL} directory. @item Add the directory @file{C:\PERL\BIN} to your path. @item Add the directory @file{C:\PERL\BIN\MSWIN32-x86-thread} or @file{C:\PERL\BIN\MSWIN32-x86} to your path. @item Test that @code{perl} works by executing @code{perl -v} in a DOS shell. @end itemize @node Perl support problems, , Windows Perl, Perl support @appendixsubsec Problems Using the Perl @code{DBI}/@code{DBD} Interface @cindex problems, installing Perl @cindex Perl DBI/DBD, installation problems If Perl reports that it can't find the @file{../mysql/mysql.so} module, then the problem is probably that Perl can't locate the shared library @file{libmysqlclient.so}. You can fix this by any of the following methods: @itemize @bullet @item Compile the @code{Msql-Mysql-modules} distribution with @code{perl Makefile.PL -static -config} rather than @code{perl Makefile.PL}. @item Copy @code{libmysqlclient.so} to the directory where your other shared libraries are located (probably @file{/usr/lib} or @file{/lib}). @item On Linux you can add the pathname of the directory where @file{libmysqlclient.so} is located to the @file{/etc/ld.so.conf} file. @tindex LD_RUN_PATH environment variable @tindex Environment variable, LD_RUN_PATH @item Add the pathname of the directory where @file{libmysqlclient.so} is located to the @code{LD_RUN_PATH} environment variable. @end itemize If you get the following errors from @code{DBD-mysql}, you are probably using @code{gcc} (or using an old binary compiled with @code{gcc}): @example /usr/bin/perl: can't resolve symbol '__moddi3' /usr/bin/perl: can't resolve symbol '__divdi3' @end example Add @code{-L/usr/lib/gcc-lib/... -lgcc} to the link command when the @file{mysql.so} library gets built (check the output from @code{make} for @file{mysql.so} when you compile the Perl client). The @code{-L} option should specify the pathname of the directory where @file{libgcc.a} is located on your system. Another cause of this problem may be that Perl and MySQL aren't both compiled with @code{gcc}. In this case, you can solve the mismatch by compiling both with @code{gcc}. If you get the following error from @code{Msql-Mysql-modules} when you run the tests: @example t/00base............install_driver(mysql) failed: Can't load '../blib/arch/auto/DBD/mysql/mysql.so' for module DBD::mysql: ../blib/arch/auto/DBD/mysql/mysql.so: undefined symbol: uncompress at /usr/lib/perl5/5.00503/i586-linux/DynaLoader.pm line 169. @end example it means that you need to include the compression library, -lz, to the link line. This can be doing the following change in the file @file{lib/DBD/mysql/Install.pm}: @example $sysliblist .= " -lm"; to $sysliblist .= " -lm -lz"; @end example After this, you MUST run 'make realclean' and then proceed with the installation from the beginning. If you want to use the Perl module on a system that doesn't support dynamic linking (like SCO) you can generate a static version of Perl that includes @code{DBI} and @code{DBD-mysql}. The way this works is that you generate a version of Perl with the @code{DBI} code linked in and install it on top of your current Perl. Then you use that to build a version of Perl that additionally has the @code{DBD} code linked in, and install that. On SCO, you must have the following environment variables set: @example shell> LD_LIBRARY_PATH=/lib:/usr/lib:/usr/local/lib:/usr/progressive/lib or shell> LD_LIBRARY_PATH=/usr/lib:/lib:/usr/local/lib:/usr/ccs/lib:/usr/progressive/lib:/usr/skunk/lib shell> LIBPATH=/usr/lib:/lib:/usr/local/lib:/usr/ccs/lib:/usr/progressive/lib:/usr/skunk/lib shell> MANPATH=scohelp:/usr/man:/usr/local1/man:/usr/local/man:/usr/skunk/man: @end example First, create a Perl that includes a statically linked @code{DBI} by running these commands in the directory where your @code{DBI} distribution is located: @example shell> perl Makefile.PL -static -config shell> make shell> make install shell> make perl @end example Then you must install the new Perl. The output of @code{make perl} will indicate the exact @code{make} command you will need to execute to perform the installation. On SCO, this is @code{make -f Makefile.aperl inst_perl MAP_TARGET=perl}. Next, use the just-created Perl to create another Perl that also includes a statically-linked @code{DBD::mysql} by running these commands in the directory where your @code{Msql-Mysql-modules} distribution is located: @example shell> perl Makefile.PL -static -config shell> make shell> make install shell> make perl @end example Finally, you should install this new Perl. Again, the output of @code{make perl} indicates the command to use. @node Group by functions, , Perl support, Placeholder @appendixsec Functions for Use with @code{GROUP BY} Clauses @findex GROUP BY functions @findex functions, GROUP BY If you use a group function in a statement containing no @code{GROUP BY} clause, it is equivalent to grouping on all rows. @table @code @findex COUNT() @item COUNT(expr) Returns a count of the number of non-@code{NULL} values in the rows retrieved by a @code{SELECT} statement: @example mysql> select student.student_name,COUNT(*) from student,course where student.student_id=course.student_id GROUP BY student_name; @end example @code{COUNT(*)} is somewhat different in that it returns a count of the number of rows retrieved, whether or not they contain @code{NULL} values. @code{COUNT(*)} is optimized to return very quickly if the @code{SELECT} retrieves from one table, no other columns are retrieved, and there is no @code{WHERE} clause. For example: @example mysql> select COUNT(*) from student; @end example @findex COUNT(DISTINCT) @findex DISTINCT @item COUNT(DISTINCT expr,[expr...]) Returns a count of the number of different non-@code{NULL} values: @example mysql> select COUNT(DISTINCT results) from student; @end example In MySQL you can get the number of distinct expression combinations that don't contain NULL by giving a list of expressions. In ANSI SQL you would have to do a concatenation of all expressions inside @code{CODE(DISTINCT ..)}. @findex AVG() @item AVG(expr) Returns the average value of @code{expr}: @example mysql> select student_name, AVG(test_score) from student GROUP BY student_name; @end example @findex MIN() @findex MAX() @item MIN(expr) @itemx MAX(expr) Returns the minimum or maximum value of @code{expr}. @code{MIN()} and @code{MAX()} may take a string argument; in such cases they return the minimum or maximum string value. @xref{MySQL indexes}. @example mysql> select student_name, MIN(test_score), MAX(test_score) from student GROUP BY student_name; @end example @findex SUM() @item SUM(expr) Returns the sum of @code{expr}. Note that if the return set has no rows, it returns NULL! @findex STD() @findex STDDEV() @cindex Oracle compatibility @cindex compatibility, with Oracle @item STD(expr) @itemx STDDEV(expr) Returns the standard deviation of @code{expr}. This is an extension to ANSI SQL. The @code{STDDEV()} form of this function is provided for Oracle compatibility. @findex BIT_OR() @item BIT_OR(expr) Returns the bitwise @code{OR} of all bits in @code{expr}. The calculation is performed with 64-bit (@code{BIGINT}) precision. @findex BIT_AND() @item BIT_AND(expr) Returns the bitwise @code{AND} of all bits in @code{expr}. The calculation is performed with 64-bit (@code{BIGINT}) precision. @end table @cindex @code{GROUP BY}, extensions to ANSI SQL MySQL has extended the use of @code{GROUP BY}. You can use columns or calculations in the @code{SELECT} expressions that don't appear in the @code{GROUP BY} part. This stands for @emph{any possible value for this group}. You can use this to get better performance by avoiding sorting and grouping on unnecessary items. For example, you don't need to group on @code{customer.name} in the following query: @example mysql> select order.custid,customer.name,max(payments) from order,customer where order.custid = customer.custid GROUP BY order.custid; @end example In ANSI SQL, you would have to add @code{customer.name} to the @code{GROUP BY} clause. In MySQL, the name is redundant if you don't run in ANSI mode. @strong{Don't use this feature} if the columns you omit from the @code{GROUP BY} part aren't unique in the group! You will get unpredictable results. In some cases, you can use @code{MIN()} and @code{MAX()} to obtain a specific column value even if it isn't unique. The following gives the value of @code{column} from the row containing the smallest value in the @code{sort} column: @example substr(MIN(concat(rpad(sort,6,' '),column)),7) @end example @xref{example-Maximum-column-group-row}. @cindex @code{ORDER BY}, aliases in @cindex aliases, in @code{ORDER BY} clauses @cindex @code{GROUP BY}, aliases in @cindex aliases, in @code{GROUP BY} clauses @cindex expression aliases @cindex aliases, for expressions Note that if you are using MySQL Version 3.22 (or earlier) or if you are trying to follow ANSI SQL, you can't use expressions in @code{GROUP BY} or @code{ORDER BY} clauses. You can work around this limitation by using an alias for the expression: @example mysql> SELECT id,FLOOR(value/100) AS val FROM tbl_name GROUP BY id,val ORDER BY val; @end example In MySQL Version 3.23 you can do: @example mysql> SELECT id,FLOOR(value/100) FROM tbl_name ORDER BY RAND(); @end example @node Function Index, Concept Index, Placeholder, Top @unnumbered SQL command, type and function index @printindex fn @page @node Concept Index, , Function Index, Top @unnumbered Concept Index @printindex cp @bye