Hi Guys, We have a master / master setup running and I was wondering if ISPConfig had any script or option to email admin if replication has failed ?
Replication issues are reported as loglevel 'ERROR', so when you have chosen to get an email in case that an error of that level occurs for this slave (see system > server config of this slave server), then you should get an email notification.
Not had a chance to look at this all week. Yes Error level is set for both master nodes but I'm not getting any replication email errors on either primary or secondary servers. We had an issue with the secondary server at one point with a complete crash under Azure with no information as to why. Then the primary a week after that had the machine run out of memory, again not figured out why. On both occasions due to the duration of the outage (happened early morning) I stopped the slaves and changed logs before restarting. So slave is working fine on primary but the secondary seems to get stuck now and then overnight. I guess the best was is to stop both slaves and copy the primary to the secondary and restart maybe ?
I guess I misunderstood your first email. You are talking about MySQL replication while I thought you were talking about ISPConfig replication. ISPConfig reports errors that happen in it's own database replication mechanism by email, but it does not report errors that happen in MySQL master/master replication as it is not aware of these errors. If MySQL master/master replication broke, then it's indeed the best option to set it up again by copying dbispconfig and dbispconfig2 from first to second server and reinitiate the sync.
