What steps will reproduce the problem?
1. Switch back and forth between two servers
2. relay_log_purge will be set to OFF on both
I think this is expected and by design, but I want to be able to avoid setting
relay_log_purge to OFF. Our use case only involves a Master and a Standby. The
reason for using MHA is to have tooling performing the failover/switchover and
avoid human error etc. We do not rely on being able to read relays from other
slaves (since we do not have them). Because of this, we do not want to
introduce new components for handling relay-log rotation etc. We want to use
the built in purge.
I cannot find a way to have MHA avoid disabling relay_log_purge, is this
possible?
Original issue reported on code.google.com by noj.nils...@gmail.com on 5 May 2014 at 8:00
Original issue reported on code.google.com by
noj.nils...@gmail.com
on 5 May 2014 at 8:00