[Pacemaker] colocation/order problem

Thomas ThomasCaspari at t-online.de
Wed May 11 09:33:18 UTC 2011


Hello all,

I'm having a little problem with my cluster config. Since I seem to be unable to
find the solution in the howtos/wiki, I hope someone here can help me.

Communication between nodes of a two-node-cluster (pacemaker/corosync) died
after accidentially changing one's hostname. The two nodes now seem to have
their own cib and refuse to communicate with each other. It seems I now have two
one-node-clusters instead of a two-node one. The cluster is currently being
built, it's not in productive use yet.

How can I reset one node to reestablish communication? Deleting cib on both
nodes is possible because I have saved my config. I tried to delete
/var/lib/heartbeat/crm/* on one node, but that doesn't suffice. The
communication doesn't return, because both nodes seem to think they manage their
cluster on their own. 

How can I get back to the point after fresh installation of pacemaker/corosync?

regards

Thomas





More information about the Pacemaker mailing list