Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Not a Bug
-
22.08.4
-
None
-
MXS-SPRINT-195, MXS-SPRINT-196, MXS-SPRINT-201
Description
The switch master breaks the replication if server has currently running events with super-privilege.
During a switch call, the monitor disable the EVENTS on the actual master before proceeding to the switch, unfortunately this does not prevent that there are event running, and we will end up with a switch succeed but the Ex-master, get more GTID and the replication will be out of sync.
after disable all the events,
select * from information_Schema.processlist should tell if there are event running.
based on that we or refuse the switch, or wait, or handle the event kill