- 17 Apr, 2013 40 commits
-
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22581 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22552 c7de825b-a66e-492c-adef-691d508d4ae1
-
Yoni Fogel authored
git-svn-id: file:///svn/toku/tokudb@22550 c7de825b-a66e-492c-adef-691d508d4ae1
-
Yoni Fogel authored
git-svn-id: file:///svn/toku/tokudb@22547 c7de825b-a66e-492c-adef-691d508d4ae1
-
Yoni Fogel authored
Cleaning up of rollback logs in toku_rollback_txn_close is indicative of a problem that should cause a crash right away, not the next time you start up mysql. This might be the direct cause of global's problem, but requires another bug to exist (rollback nodes must not have been cleaned up during a 'successful' abort or commit first in order for the 'cleanup' to cause a problem) git-svn-id: file:///svn/toku/tokudb@22526 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22418 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22416 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22413 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@22354 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
{{{ svn merge -r22020:22070 https://svn.tokutek.com/tokudb/toku/tokudb.2755c }}} . git-svn-id: file:///svn/toku/tokudb@22160 c7de825b-a66e-492c-adef-691d508d4ae1
-
Yoni Fogel authored
git-svn-id: file:///svn/toku/tokudb@22014 c7de825b-a66e-492c-adef-691d508d4ae1
-
Dave Wells authored
git-svn-id: file:///svn/toku/tokudb@21915 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21899 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
When processing suppress_rollback log entry, associate brt with transaction. git-svn-id: file:///svn/toku/tokudb@21894 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21893 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21878 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21856 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
git-svn-id: file:///svn/toku/tokudb@21828 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
Fixes #2782. The brtloader uses temp dir. Note however that #2783 isn't fixed yet. close[t:2782]. git-svn-id: file:///svn/toku/tokudb@21827 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
git-svn-id: file:///svn/toku/tokudb@21769 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
git-svn-id: file:///svn/toku/tokudb@21573 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
* That's 2/3 of 3/4, which is calculated as 1/4 is unreservable. The loader asks for 2/3 of what's reservable, getting one half. The second loader gets 1/6 of cachetable memory. * That's 2/3 of 1/4: The loader asks for 2/3 of what's reservable. 1/2 has been reserved. 1/4 is unreservable. 1/4 is reservable. 2/3 of 1/4 is 1/6. The third loader gets 1/18 of cachetable memory. * That's 2/3 or 1/12. (1/4 is unreservable. 1/2 was reserved by the first loader. 1/6 by the second. That leaves 1/12) Generally the next one gets 1/3 of what the previous one got. Refs #2759. [t:2759] git-svn-id: file:///svn/toku/tokudb@21564 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
{{{ svn merge -r21499:21502 https://svn.tokutek.com/tokudb/toku/tokudb.2755 }}} . git-svn-id: file:///svn/toku/tokudb@21503 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21309 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21289 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21260 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21217 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21103 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21084 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
git-svn-id: file:///svn/toku/tokudb@21083 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
Candidate fix for #2730 (the race in the loader and the io-thread where they glare at each other). Refs #2730. [t:2730] git-svn-id: file:///svn/toku/tokudb@21082 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
[t:2721] Add -k option to loader-cleanup-test to test only abort-via-poll, but test it exhaustively for any size database. git-svn-id: file:///svn/toku/tokudb@21072 c7de825b-a66e-492c-adef-691d508d4ae1
-
Barry Perlman authored
[t:2721] Improve loader-cleanup-test to return non-zero from polling function for different cases (first time, second time, etc., just like error injection). git-svn-id: file:///svn/toku/tokudb@21070 c7de825b-a66e-492c-adef-691d508d4ae1
-
Yoni Fogel authored
dbt.size should be length of string, not size of pointer git-svn-id: file:///svn/toku/tokudb@21052 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21045 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21043 c7de825b-a66e-492c-adef-691d508d4ae1
-
Rich Prohaska authored
git-svn-id: file:///svn/toku/tokudb@21040 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
git-svn-id: file:///svn/toku/tokudb@21039 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
Need an extra test for {{{loader-stress-test.loader}}} in the {{{Makefile}}} to manifest #2717. Refs #2717. [t:2717] git-svn-id: file:///svn/toku/tokudb@21038 c7de825b-a66e-492c-adef-691d508d4ae1
-
Bradley C. Kuszmaul authored
git-svn-id: file:///svn/toku/tokudb@21037 c7de825b-a66e-492c-adef-691d508d4ae1
-