On 12/3/2009 at 10:47 AM, Shravan Mishra <shravan.mis...@gmail.com> wrote: > Hi guys, > > We are testing Active passive scenario. We pulled out the power cord from > primary and expected to see secondary takeover as primary but this is what > we see: > > ... > > We are using ipmi fencing devices. >
When one node disappears, the other one will (or should) try to STONITH it to make sure it's really, truly dead. Only when STONITH succeeds will that node's resources be taken over by the remaining node. I suspect the problem in your case is that IPMI STONITH can't succeed when run against a node that has no power, because it is unable to respond to IPMI requests. Regards, Tim -- Tim Serong <tser...@novell.com> Senior Clustering Engineer, Novell Inc. _______________________________________________ Pacemaker mailing list Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker
