Hi,

Good guess Andrew ... but it seems not sufficient, I've tested :
I've set one nodeid different on each node in corosync.conf
(HA cluster with 4 nodes) ,with two interfaces in corosync.conf ,
but only one interface available (one not seeable with ip addr) ,
the result is :
now I can effectively start corosync on nodes (whereas I can't
start it on more than one node without nodeid set in file)
but after 60s, I have with crm_mon :
============
Last updated: Mon Apr 19 09:07:37 2010
Stack: openais
Current DC: node3 - partition WITHOUT quorum
Version: 1.0.8-3225fc0d98c8fcd0f7b24f0134e89967136a9b00
2 Nodes configured, 2 expected votes
5 Resources configured.
============

Online: [ node3 ]
OFFLINE: [ node6 ]

whereas status of corosync is "running" on both nodes.

Then, on node6, a /etc/init.d/corosync stop remains stalled
"Waiting for corosync services to unload:..."

I  guess it is a corosync problem , not a Pacemaker one, so it is not
the right ML, but it was just to give a try and an answer to your 
suggestion.

Thanks
Regards
Alain



_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to