innodb - sync_binlog parameter mysql -


i going through documentation sync_binlog parameter , found discrepancy in sync_binlog parameter documentation.

the documentation here http://dev.mysql.com/doc/refman/5.6/en/replication-options-binary-log.html#sysvar_sync_binlog says:

a value of 1 safest choice because in event of crash lose @ 1 commit group binary log.

which means data updated might not there in binlogs.

however,in binary log documentation here http://dev.mysql.com/doc/refman/5.6/en/binary-log.html says:

for example, if using innodb tables , mysql server processes commit statement, writes many prepared transactions binary log in sequence, synchronizes binary log, , commits transaction innodb. if server crashes between 2 operations, transaction rolled innodb @ restart still exists in binary log.

which means transaction first written in binlog , committed innodb, there chance in case of crash row there in binlog doesn;t exist in database.

i have asked question in mysql forum , waiting response, appreciate if some-one has used parameter can give details on of following 2 behaviour correct?

appreciate help.

syncing in server different syncing slaves in replication setup.

the binlog (hence, not sync_binlog , "group commit") not used in integrity of innodb tables on master. however, point out, involved.

the crash mention happen after logs written local machine, hence local machine (if comes up) can recover. "group commit" may not make slaves. if gets binlog, not assure slave (much less, slaves) has gotten value yet.

see "semi-sync" replication, wherein @ least 1 slave must acknowledge before commit ack'd.

you quoted implies transaction slave, yet rolled on master. think there open bug on case.

gtids confuse issue. of quotes made predate gtids , need updating.

consider reporting @ http://bugs.mysql.com .


Comments

Popular posts from this blog

java - Jasper subreport showing only one entry from the JSON data source when embedded in the Title band -

mapreduce - Resource manager does not transit to active state from standby -

serialization - Convert Any type in scala to Array[Byte] and back -