MariaDB 10.0.14 Changelog

The most recent release in the MariaDB 10.0 series is:MariaDB 10.0.38 Download Now

DownloadRelease NotesChangelogOverview of 10.0

Release date: 26 Sep 2014

For the highlights of this release, see the release notes.

The revision number links will take you to the revision's page on Launchpad. On Launchpad you can view more details of the revision and view diffs of the code modified in that revision.

• Revision #4416 Wed 2014-09-24 15:41:42 +0200

  • crash in GROUP_CONCAT(IF () ORDER BY 1)

• Wed 2014-09-24 17:27:00 +0400

  • ujis and eucjmps erroneously accept 0x8EA0 as a valid byte sequence

• Wed 2014-09-24 13:08:47 +0200

  • remove the bug fix for "crash in GROUP_CONCAT(IF () ORDER BY 1)"

• [merge] Tue 2014-09-23 23:55:29 +0200

  • 5.5 merge

  • [merge] Tue 2014-09-23 23:37:35 +0200

    • merge
• Tue 2014-09-23 15:58:54 +0400

  • Adding tests for handling 0x5C as the second byte in a multi-byte sequence, and as a escape character when SET NAMES xxx, character_set_connection=binary; for cp932,big5,gbk,sjis

• Mon 2014-09-22 23:25:56 +0300

  • Fixed test failures Added comments Ensure that tokudb test works even if jemalloc is not installed Removed not referenced function Item::remove_fixed()

• [merge] Thu 2014-09-18 21:54:45 +0200

  • 5.5 merge

  • Thu 2014-09-18 17:00:44 +0200

    • support statically linked jemalloc. use that for release builds

• Thu 2014-09-18 15:24:30 +0200

  • print binlog unsafe errors at log_warnings level 1, not 2.

• Thu 2014-09-18 12:40:55 +0400

  • Trailing incomplete characters are not replaced to question marks on conversion

• Thu 2014-09-18 09:26:30 +0900

  • Merge Spider 3.2.11

• Wed 2014-09-17 19:38:42 +0200

  • fixes for valgrind failures

• Wed 2014-09-17 15:11:24 +0200

  • fix intermittent failures of main.create_or_replace test in buildbot

• [merge] Tue 2014-09-16 14:08:05 +0200

  • merge

  • Tue 2014-09-16 14:04:50 +0200

    • fixes for test cases
• Tue 2014-09-16 07:37:00 +0300

  • Avoid using log_sys mutex when printing out show engine innodb status, instead peek (maybe) old data.

• [merge] Tue 2014-09-16 00:06:05 +0300

  • Auto merge

  • Tue 2014-09-16 00:00:47 +0300

    • Don't give warning if there are two unique keys used with INSERT .. ON DUPLICATE KEY UPDATE. We should assume that the store engine will report the first duplicate key for this case.
• Mon 2014-09-15 20:33:11 +0400

  • Changes in storage_engine test suite:

• Sat 2014-09-13 17:15:11 +0400

  • Adding big5, cp932, gbk, sjis tests covering characters that can have 0x5C as the second byte in a multi-byte character.

  • Adding big5 tests covering an unassigned character 0xC840 being stored into char/varchar/text/enum columns.

• [merge] Fri 2014-09-12 16:44:27 +0200

  • 10.0-connect merge

  • Sat 2014-09-06 18:08:28 +0200

    • Fix (indexing on void table)
• Fri 2014-09-12 16:06:18 +0400

  • Stored routines do now work with swe7: "The table mysql.proc is missing, corrupt, or contains bad data" Fixed the bug itself. Also, added "SET NAMES swe7" which was forgotten in the previous commit, so latin1 was actually tested lati1 instead of swe7 in a mistake. Now it tests swe7.

• Fri 2014-09-12 14:49:24 +0300

  • Fixed compiler warnings

• Fri 2014-09-12 14:49:13 +0300

  • Don't use LOCK_status for the duration of SHOW STATUS because of possible lookups. Instead we use LOCK_status only to protect summary of thread statistics and use a new mutex, LOCK_show_status to protect concurrent SHOW STATUS.

• Fri 2014-09-12 14:45:11 +0300

  • Cleanup of my_hash_sort patch

• Fri 2014-09-12 12:57:27 +0400

  • Adding thorough tests covering what happens with escaped sequences in the SQL parser.

• Fri 2014-09-12 03:21:54 +0400

  • Fixes in storage_engine test suite

• [merge] Thu 2014-09-11 23:50:31 +0300

  • Automatic merge

  • Thu 2014-09-11 22:42:35 +0300

    • DUPLICATE KEY Errors on SELECT .. GROUP BY that uses temporary and filesort.
• [merge] Thu 2014-09-11 16:44:16 +0200

  • XtraDB 5.6.20-68.0

  • Thu 2014-09-11 10:15:01 +0200

    • percona-server-5.6.20-68.0

• [merge] Thu 2014-09-11 16:42:54 +0200

  • InnoDB 5.6.20

  • Thu 2014-09-11 10:13:35 +0200

    • 5.6.20

• Thu 2014-09-11 15:41:30 +0300

  • : InnoDB: Failing assertion: state == TRX_STATE_NOT_STARTED in file trx0trx.ic line 60

• Wed 2014-09-10 17:45:09 +0200

  • MariaDB CLI client doesnt show CREATE INDEX progress

• Wed 2014-09-10 13:22:44 +0200

  • max_relay_log_size and sql_slave_skip_counter misbehave on PPC64

• Wed 2014-09-10 13:22:20 +0200

  • CONNECT temporary table created

• Wed 2014-09-10 13:22:01 +0200

  • search order for my.cnf inconsistent in docs/use, and global override with build-time -DDEFAULT_SYSCONFDIR is ignored

• [merge] Wed 2014-09-10 13:10:24 +0200

  • IMPORT_EXECUTABLES feature broken by addition of gen_pfs_lex_token

  • Fri 2014-08-15 22:54:07 +0100

    • : Properly handle gen_pfs_lex_token when cross-compiling

• Mon 2014-09-08 12:59:57 +0200

  • compilation fixes for WITH_ATOMIC_OPS=rwlocks

• Sun 2014-09-07 22:04:29 +0200

  • typos in cmake -DWITH_ATOMIC_OPS=xxx

• Sun 2014-09-07 20:40:36 +0200

  • mysql_options4 symbol missing from libmysqlclient.so.18

• Sun 2014-09-07 20:19:29 +0200

  • compilation failure on x86

• Sun 2014-09-07 20:19:12 +0200

  • Assertion `thd' failed in my_malloc_size_cb_func upon writing status report into error log

• [merge] Sun 2014-09-07 20:18:17 +0200

  • breaks WITHOUT_SERVER build

  • Mon 2014-08-11 22:43:29 +0100

    • : Fix WITHOUT_SERVER build

• Sat 2014-09-06 09:46:41 +0200

  • Assertion `thd->is_error() || thd->killed' failed in mysql_execute_command on executing an SP with repeated CREATE TABLE .. SELECT

• Sat 2014-09-06 08:33:56 +0200

  • Server crashes in my_hash_first if shutdown is performed when FLUSH LOGS is running

• Thu 2014-09-04 21:58:48 +0400

  • Bad column name for UCS2 string literals The Item_string constructors called set_name() on the source string, which was wrong because in case of UCS2/UTF16/UTF32 the source value might be a not well formed string (e.g. have incomplete leftmost character). Now set_name() is called on str_value after its copied (with optionally left zero padding) from the source string.

  • Illegal mix of collation with a PS parameter Item_param::convert_str_value() did not set repertoire. Introducing a new structure MY_STRING_METADATA to collect character length and repertoire of a string in a single loop, to avoid two separate loops. Adding a new class Item_basic_value::Metadata as a convenience wrapper around MY_STRING_METADATA, to reuse the code between Item_string and Item_param.

• Thu 2014-09-04 12:43:41 +0400

  • Creating a new class in_string::Item_string_for_in_vector and moving set_value() from Item_string to Item_string_for_in_vector, as set_value() updates the members incompletely (e.g. does not update max_length), so it was dangerous to have set_value() available in Item_string.

• Thu 2014-09-04 12:15:05 +0400

  • Using more Item_string_sys

• Thu 2014-09-04 08:50:06 +0400

  • MySQL BUG#12735829 - SPACE() FUNCTION WARNING REFERS TO REPEAT() IN ER_WARN_ALLOWED_PACKET_OVERFLOWED Merged from 5.6

• Wed 2014-09-03 18:24:31 +0400

  • Adding Item_string_sys and Item_string_ascii to reduce duplicate code

• Wed 2014-09-03 16:31:47 +0400

  • Illegal mix of collation with bit string B'01100001'

• Tue 2014-09-02 14:07:01 +0200

  • : Slave replicating using GTID doesn't recover correctly when master crashes in the middle of transaction

• Wed 2014-09-03 01:56:21 +0400

  • Moving Item::str_value from public to protected.

• Wed 2014-09-03 01:47:39 +0400

  • A parameter and a string literal with the same values are not recognized as equal by the optimizer

• Tue 2014-09-02 22:04:48 +0400

  • Different optimizer plan for "a BETWEEN 'string' AND ?" and "a BETWEEN ? AND 'string'" Item_string::eq() and Item_param::eq() in string context behaved differently. Introducing a new class Item_basic_value to share the eq() code between literals (Item_int, Item_double, Item_string, Item_null) and Item_param.

• Tue 2014-09-02 17:50:09 +0300

  • innodb.innodb_simulate_comp_failures_small is too slow if it's run on a real disk

• Tue 2014-09-02 17:34:29 +0400

  • A clean-up for the previous patch

• Mon 2014-09-01 20:57:32 +0400

  • PI() does not work well in UCS2/UTF16/UTF32 context Malformed result for CONCAT(utf8_column, binary_string)

• Mon 2014-08-25 13:35:33 +0300

  • : InnoDB recover ends in signal 11

• Thu 2014-08-21 21:25:22 +0200

  • SHOW GRANTS for current_user_name@wrong_host_name

• Sun 2014-08-10 14:36:17 +0200

  • sanity

• Wed 2014-08-20 15:02:10 +0200

  • Fix test case that requires dbug to not fail in release build.

• Wed 2014-08-20 10:59:39 +0200

  • : close_temporary_tables() in format description event not serialised correctly

• Tue 2014-08-19 14:26:42 +0200

  • : close_temporary_tables() in format description event not serialised correctly

• Wed 2014-07-02 12:51:45 +0200

  • : close_temporary_tables() in format description event not serialised correctly

• [merge] Tue 2014-08-19 21:35:14 +0300

  • Automatic merge from 5.5 Fixed 2 failing tests by replacing result files

  • Tue 2014-08-19 19:28:35 +0300

    • - MariaDB crash on Power8 when built with advance tool chain
• Fri 2014-08-15 11:31:13 +0200

  • : Some replication errors are ignored if slave_parallel_threads > 0

• Thu 2014-08-14 15:38:08 +0300

  • If one uses 3 --verbose options to mysql_upgrade or mysqlcheck one will now get on stdout all ALTER, RENAME and CHECK commands that mysqlcheck executes. If one uses 4 --verbose to mysql_upgrade it will also write out all mysqlcheck commands invoked.

• Thu 2014-08-14 15:36:48 +0300

  • Added Innobase .ic and errmsg-utf8.txt to tagged files Fixed compiler warning

• Wed 2014-08-13 13:34:28 +0200

  • , failing to update gtid_slave_pos for a transaction that was retried.

• Wed 2014-08-13 09:10:56 +0300

  • : innodb.innodb_simulate_comp_failures_small fails sporadically

• Tue 2014-08-12 19:16:44 +0400

  • Increased the version number

• Tue 2014-08-12 07:52:19 +0400

  • Recoding feedback_plugin_send.result (forgotten in the previous commit).

• Mon 2014-08-11 05:45:45 +0400

  • Collation usage statistics Adding collation usage statistics into the feedback plugin I_S table.