[Pacemaker] MySQL Master-Master replication with Corosync and Pacemaker

Yves Trudeau y.trudeau at videotron.ca
Thu Feb 9 23:18:35 UTC 2012


Hi Mark,
    nothing happens to the VIP?  As it is, the agent is not monitoring 
mysqld itself.  In my case, I call mysqld_safe so the kill -9 would be 
handled.  I am out for a few days, answers will be delayed until end of 
next week.

Regards,

Yves

On 12-02-09 06:04 PM, Mark Grennan wrote:
> Yves
>
> I've been testing the Master-Master / Pacemaker / MHA failover with some good results.
>
> Oddly, the one thing that is not happening is, if I just kill -9 mysql Pacemaker doesn't see it.
>
> Pulling the network cable works good. This is a really the worst thing to do.  Pacemaker sees it and turns the slave into the master quickly!!! I waited two minutes let the "old master" get good and behind and plugged the cable back in. SPLIT BRAIN!  Pacemaker saw it very quickly and restarted mysql and brought the "old master" online as the new slave. After some time a checked the Tables with pt-table-checksum and all was good.  Just amazing.
>
> Mark G.
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>




More information about the Pacemaker mailing list