#14109: mariadb-10.5.6
-------------------------+-----------------------
 Reporter:  bdubbs       |       Owner:  bdubbs
     Type:  enhancement  |      Status:  assigned
 Priority:  normal       |   Milestone:  10.1
Component:  BOOK         |     Version:  SVN
 Severity:  normal       |  Resolution:
 Keywords:               |
-------------------------+-----------------------

Comment (by bdubbs):

 MariaDB 10.5.5 Release Notes

 Notable Changes

 Deprecated variables

 * innodb_thread_concurrency
 * innodb_commit_concurrency
 * innodb_replication_delay
 * innodb_concurrency_tickets
 * innodb_thread_sleep_delay
 * innodb_adaptive_max_sleep_delay

 InnoDB

 * Fixed corruption in delete buffering
 * Fixed a deadlock in FLUSH TABLES…FOR EXPORT
 * InnoDB data file extension is not crash-safe
 * Minor fixes related to encryption and FULLTEXT INDEX
 * Dropping the adaptive hash index may cause DDL to lock up InnoDB
 * innodb_log_optimize_ddl=OFF is not crash safe
 * Mariadb service won't shutdown when it's running and the OS datetime
 updated backwards
 * Doublewrite recovery can corrupt data pages
 * Fixed race conditions related to buffer pool resizing
 * ALTER TABLE fixes
 * Slow InnoDB shutdown on large instance
 * Fixes to performance regressions introduced in MariaDB 10.5.4
 * Performance improvements
 * Correctly implemented the scrubbing of freed pages
 * Crash recovery fixes

 Replication

 * Make the binlog dump thread to log into errorlog a requested GTID
 position
 * Fix stop of the optimistic parallel slave at requested START-SLAVE-UNTIL
 position
 * Properly handle RESET MASTER TO value, when the value exceeds the max
 allowed 2147483647
 * Correct 'relay-log.info' updating by concurrent parallel workers
 * Eliminate deadlock involving parallel workers, STOP SLAVE and FLUSH
 TABLES WITH READ LOCK
 * Correct master-slave automatic reconnection by slave to always pass
 through all steps
   of the initial connect. Specifically, do not skip master notification
 about slave
   binlog checksum awareness
 * Refine mysqlbinlog output to print out START TRANSACTION at
 Gtid_log_event
   processing which satisfies clients that submit the output with
 sql_mode=oracle
 * Replication aborts with ER_SLAVE_CONVERSION_FAILED upon CREATE ...
 SELECT in ORACLE mode

 Optimizer

 * ALTER TABLE ... ANALYZE PARTITION ... with EITS reads and locks all rows
 ...
 * Print ranges in the optimizer trace created for non-indexed columns when
   optimizer_use_condition_selectivity >2
   Now the optimizer trace shows the ranges constructed while getting
 estimates from EITS
 * LATERAL DERIVED is not clearly visible in EXPLAIN FORMAT=JSON, make
   LATERAL DERIVED tables visible in EXPLAIN FORMAT=JSON output
 * Crash on WITH RECURSIVE large query
 * Crash with Prepared Statement with a '?' parameter inside a re-used CTE

 Other

 * MariaDB could crash after changing the query_cache size
 * Errors and SIGSEGV on CREATE TABLE w/ various charsets
 * Crash in CREATE TABLE AS SELECT when the precision of returning type = 0
 * XA: Reject DDL operations between PREPARE and COMMIT
 * Stop mariabackup --prepare on errors during innodb redo log applying
 * Server crashes in mysql_alter_table upon adding a non-null date column
 under
   NO_ZERO_DATE with ALGORITHM=INPLACE
 * Can't uninstall plugin if the library file doesn't exist
 * Mariabackup parameter cleanup
 * Rounding functions return wrong datatype
 * Create mariadb.sys user on each update even is the user is not needed
 * INFORMATION_SCHEMA.INNODB_TABLESPACES_ENCRYPTION required SUPER instead
 PROCESS privilege
 * Reinforce DDL operation rejection after XA PREPARE
 * As per the MariaDB Deprecation Policy, this will be the last release of
   MariaDB 10.5 for Ubuntu 19.10 Eoan and Fedora 30

--
Ticket URL: <http://wiki.linuxfromscratch.org/blfs/ticket/14109#comment:2>
BLFS Trac <http://wiki.linuxfromscratch.org/blfs>
Beyond Linux From Scratch
-- 
http://lists.linuxfromscratch.org/listinfo/blfs-book
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to