Emergency

Percona Server 5.1.53-12.4

Percona is glad to announce the release of Percona Server 5.1.53-12.4 on December 29, 2010 (Downloads are available here and from the Percona Software Repositories).

Percona Server 5.1.53-12.4 is now the current stable release version.

Functionality Added or Changed

  • Percona Server 5.1.53-12.4 is based on MySQL 5.1.53.
  • New Features Added:
    • Precompiled UDFs for Maatkit (FNV and MurmurHash hash functions to provide faster checksums) are now included in distributions. Fixes feature request #689992. (Aleksandr Kuzminsky)
  • Other Changes:
    • percona_innodb_doublewrite_path - It’s no longer necessary to recreate your database and InnoDB system files when a dedicated file to contain the doublewrite buffer is specified. (Yasufumi Kinoshita)
    • Added system variable have_response_time_distribution and compile option --without-response_time_distribution in Response Time Distribution. (Oleg Tsarev)

Bugs Fixed

  • Bug #643149 - Slow query log entries were not written in the usual parsing format. (Alexey Kopytov)
  • Bug #671764 - innochecksum wasn’t distributed with RPM and .DEB packages. (Aleksandr Kuzminsky)
  • Bug #673426 - Use of these system variables as command-line options could cause a crash or undefined behavior: log_slow_timestamp_every, log_slow_sp_statements, slow_query_log_microseconds_timestamp, use_global_long_query_time. (Oleg Tsarev)
  • Bug #673567 - Compiler could produce spurious warnings when building on non-Linux platforms. A check is now made to see if clock_gettime() is present in librt at the configure stage. If yes, `` -lrt`` is added to LIBS. (Alexey Kopytov)
  • Bug #673929 - Query cache misses were being reported for some queries when hits were actually occurring.
  • Bug #676146 - The development environment test of log_slow_verbosity=innodb on a slave for row-based replication was not working correctly. (Oleg Tsarev)
  • Bug #676147, #676148 - The development environment tests of options log_slow_slave_statements and use_global_long_query_time work only on statement-based replication. They were failing when row-based replication was attempted. A check is now made for the replication type to test. (Oleg Tsarev)
  • Bug #676158 - Setting the query cache size to 512M caused test failure on low memory systems (Aleksandr Kuzminsky)
  • Bug #677407 - The innodb_information_schema test could fail sporadically due to flawed logic in the INFORMATION_SCHEMA.INNODB_LOCKS caching mechanism. (contributed by Kristian Nielsen) (Alexey Kopytov)
  • Bug #681486 - A dependency between Debian packages libmysqlclient16 and percona-server-common was removed. (Aleksandr Kuzminsky)
  • Bug #693815 - The test percona_innodb_buffer_pool_shm was failing. It should be run with the --big-test option. As the buffer pool size used in the test is 128M, the shared memory segment should be increased appropriately in order to run the test successfully.
  • Bug #693814, #693815, #693816, #693817, #693819 - Tests in the test environment were updated to reflect past INFORMATION_SCHEMA changes. (Aleksandr Kuzminsky)
  • Bug #693818 - Warning and error messages for stored routines could incorrectly report row numbers due to a change in the slow_extended patch. (Alexey Kopytov)

Table Of Contents

Previous topic

Percona Server 5.1.54-12.5

Next topic

Percona Server 5.1.53-11.7