[Pacemaker] Resource Stickiness

frank frank at dead-link.org
Tue Mar 23 20:58:43 UTC 2010


On Tue, 2010-03-23 at 16:26 +0100, Andrew Beekhof wrote:

> > Killed Corosync on data01, the node goes down as expected and the
> > resource fails over to data02. After data01 is up again the failover-ip
> > moves back to data01.
> >
> > Any ideas?
> 
> yes, you told it to:
> 
> > location cli-prefer-failover-ip failover-ip \
> >        rule $id="cli-prefer-rule-failover-ip" inf: #uname eq data01
> 
> 100000000 == inf

Got it... works!

Frank






More information about the Pacemaker mailing list