I have two node Group replication. After two or three days slave becomes sluggish to catch up with master. status shows online but slave kinda hangs. On restarting the slave mysql instance the slave catches up and works fine.
The doubt that arise is.
If slave can catchup after restart, why could not it catch up earlier in the first place.
My wild guess is some flow control that is trying to do some middleman job between master and slave but which is missing during slave startup and catchup. or some lock contention
system : MySQL 5.7 / RHEL 7.2
Regards
- Surinder
The doubt that arise is.
If slave can catchup after restart, why could not it catch up earlier in the first place.
My wild guess is some flow control that is trying to do some middleman job between master and slave but which is missing during slave startup and catchup. or some lock contention
system : MySQL 5.7 / RHEL 7.2
Regards
- Surinder