Today we had an issue with our Master-Master replication. Master2 was throwing the below error and the replication was stopped
Last_SQL_Errno: 145
Last_SQL_Error: Error 'Table './exampledb/exampletable' is marked as crashed and should be repaired' on query. Default database: 'livewbmy_db'. Query: 'INSERT INTO `exampletable` (`challenge`, `response`, `hashkey`, `expiration`) VALUES ('XBUL', 'xbul', 'a9e0f4e3dxxxxxccccccc0c8f8f1', '2016-09-29 21:10:03')'
Since its a Master-Master replication, we need to be careful before repairing the table on Master2.
Solution
- Master1 –
mysql -e "stop slave;" - Master2 –
mysql -e "stop slave;" - Master2 –
service mysqld stop - run the below now to repair the table.
[root@dr ~]# myisamchk /var/lib/mysql/exampledb/exampletable.MYI -r - recovering (with sort) MyISAM-table '/var/lib/mysql/exampledb/exampletable.MYI' Data records: 35564 - Fixing index 1 - Fixing index 2 Data records: 35522
- Master2 – Now start the slave with mysql -e “start slave;” and observe the show slave status \G and make sure Seconds_Behind_Master: has reached 0
- Master1 – Start the slave with mysql -e “start slave;” and observe show slave status \G; and make sure it is in sync.
Please comment if found helpful.
