MySQL 5.6.19 Community Release Notes

Thank you to the MySQL Community, on behalf of the MySQL team @ Oracle. Your bug reports, testcases and patches have helped create a better MySQL 5.6.19.

In particular:

  • Thank you to Daniël van Eeden for reporting that the utility charset2html was unmaintained and obsolete. We followed Daniël’s suggestion and removed it from MySQL distributions. Bug #71897.
  • Thank you to Dario Kampkaspar for reporting that upgrading from 5.6.10 to a more recent 5.6 release would fail. Bug #72079.
  • Thank you to Fangxin Flou for reporting that InnoDB would call memset too often. Fangxin also provided a suggested patch to fix the issue. Bug #71014.
  • Thank you to Daniël van Eeden for reporting an assertion when enabling the innodb_table_monitor. Bug #69641.
  • Thank you to Domas Mituzas for reporting an off by one error in the handling of innodb_max_dirty_pages_pct. Thanks also to Valeriy Kravchuk for suggesting that the value be a float, and Mark Callaghan for providing a sample patch. Bug #62534.
  • Thank you to Guangpu Feng for reporting an issue where an aborted TRUNCATE TABLE command could still be written to the binary log. Guangpu also provided detailed code-analysis, which helped with our investigation. Bug #71070.
  • Thank you to Justin Swanhart for reporting an issue where the server did not always correctly handle file permissions on the auto.cnf file. Bug #70891.
  • Thank you to Frédéric Condette for reporting an issue when using replication with GTIDs and replicate-ignore-db. Bug #71376.
  • Thank you to Michael Gmelin for reporting compilation errors when building in C++11 mode. Michael also provided a patch with a suggested fix. Bug #66803.
  • Thank you to Hartmut Holzgraefe for reporting that CMake produced not-useful warnings. Bug #71089.
  • Thank you to John Curtusan for reporting that on Windows REPAIR TABLE and OPTIMIZE TABLE failed on MyISAM tables with .MYD files larger than 4GB. Bug #69683.
  • Thank you to Elena Stepanova for reporting an issue that could break row-based replication. Elena also managed to reduce the issue down to a simple test case. Bug #71179.
  • Thank you to Yu Hongyu for reporting that an UPDATE statement could fail to update all applicable rows. Bug #69684.
  • Thank you to Tom Lane for his comments on why the test “outfile_loaddata” could be failing. His suggestion proved to be correct, and helped tremendously in tracking down the root cause. Bug #46895.

In Addition:

Thank you again to all the community contributors listed above. If I missed a name here, please let me know!

Please also note:
There was no MySQL Community Server 5.6.18. That version number was used for an out-of-schedule release of the Enterprise Edition to address the OpenSSL “Heartbleed” issue. This issue did not affect the Community Edition because it uses yaSSL, not OpenSSL, so a new release of the Community Server was not needed, and 5.6.17 is followed by 5.6.19.

– Morgan