If we have a 2+ node cluster which is replicating from an async master
and the binlog_format is set to STATEMENT and multi-row inserts are executed
on a table with an auto_increment column such that values are automatically
generated by MySQL, then the server node generates wrong auto_increment
values, which are different from what was generated on the async master.

In the title of the MDEV-9519 it was proposed to ban start slave on a Galera
if master binlog_format = statement and wsrep_auto_increment_control = 1,
but the problem can be solved without such a restriction.

The causes and fixes:

1. We need to improve processing of changing the auto-increment values
after changing the cluster size.

2. If wsrep auto_increment_control switched on during operation of
the node, then we should immediately update the auto_increment_increment
and auto_increment_offset global variables, without waiting of the next
invocation of the wsrep_view_handler_cb() callback. In the current version
these variables retain its initial values if wsrep_auto_increment_control
is switched on during operation of the node, which leads to inconsistent
results on the different nodes in some scenarios.

3. If wsrep auto_increment_control switched off during operation of the node,
then we must return the original values of the auto_increment_increment and
auto_increment_offset global variables, as the user has set. To make this
possible, we need to add a "shadow copies" of these variables (which stores
the latest values set by the user).

https://jira.mariadb.org/browse/MDEV-9519

ib_wqueue_is_empty(): protect ib_list_is_empty() call

Closes #1202

Since MariaDB 10.1.17 the new default values for wsrep_max_ws_rows and wsrep_max_ws_size were set:

wsrep_max_ws_rows

Default Value:
0 (>= MariaDB Galera 10.0.27, MariaDB 10.1.17)
131072 (<= MariaDB Galera 10.0.26, MariaDB 10.1.16)
wsrep_max_ws_size

Default Value:
2147483647 (2GB, >= MariaDB Galera 10.0.27, MariaDB 10.1.17)
1073741824 (1GB, <= MariaDB Galera 10.0.26, MariaDB 10.1.16)

This reverts commit 32629670.
It was checked in by mistake

Failed CREATE OR REPLACE for existing user removes that user
from acl_users array. Thus dependend structures (roles, check_host) must
be rebuilt.

but even if this script called as /bin/mysql_install_db
it is still standard install and scripts are in /usr/share/
(but not in the /share/)
2. fix of bindir path

By pure chance the macro worked in the cases it was used, but
better to get this fixed!

dict_create_foreign_constraints_low(): Clean up the way in
which the error messages are initialized, and ensure that
the table name is always initialized.

Most of the mtr tests in the galera_3nodes suite fail
for a variety of reasons with a variety of errors.

This patch fixes several substantial flaws
in the galera_3nodes suite tests and in the mtr framework
service files, adapting the tests from galera_3nodes
for the current version of MariaDB.

This patch also synchronizes some galera_3nodes-related
files with the latest changes made for MDEV-17835 (v2 patch)
and for MDEV-18379 in other branches (10.2 and 10.3).

Closes #1161

The code path where the table was not being rebuilt during ALTER TABLE
was not covered by the test. Add coverage, and remove the debug assertion
that could fail in this case.

i_s_innodb_mutexes_fill_table(): Use the C++ RAII pattern
to ensure that the mutexes are released if an OK()
macro returns from the function prematurely.

buf_page_is_corrupted(): Read the global variable srv_checksum_algorithm
only once in order to avoid a race condition when
SET GLOBAL innodb_checksum_algorithm=...;
is being executed concurrently with this function.

This is joint work with Oleksandr Byelkin.

wsrep_certification_rules: Define as a weak global symbol.
While there are separate _embedded.a for statically
linked storage engine plugins, there is only one ha_innodb.so
which is supposed to work with both values of WITH_WSREP.

The merge from 10.0-galera introduced a reference to a global
variable that is only defined when the server is built WITH_WSREP.
We must define that symbol as weak global, so that when
a dynamically linked InnoDB or XtraDB is used with the embedded
server (which never includes write-set replication patches),
the variable will be read as 0, instead of causing a failure to
load the InnoDB or XtraDB plugin.

Only starting with MariaDB 10.2, the .result file will echo
"connect" and "connection" statements. There is no way how
the test could have passed on debug builds after
commit 1037edcb
(which looks like an untested backport from a later version).

ha_innobase::commit_inplace_alter_table(): Do not crash if
innobase_update_foreign_cache() returns an error. It can return
an error on ALTER TABLE if an inconsistent FOREIGN KEY constraint
was created earlier when SET foreign_key_checks=0 was in effect.
Instead, report a warning to the client that constraints cannot
be loaded.

ha_innobase::prepare_inplace_alter_table(): Filter out duplicates
from ha_alter_info->alter_info->drop_list.elements.

innobase_rename_column_try(): Declare fk_evict as std::set
instead of std::list, in order to filter out duplicates.

@@open_files_limit now behaves differenly and cannot
be used to skip the test anymore.

Reverting part of da34c7de that was already fixed by MDEV-17531 by Marko