Thanks for your reply Dejan, > > > I've tried /usr/lib64/heartbeat/hb_delnode > > ghostnode to remove it, but it doesn't want to go away. When looking > > however with cibadmin -Q in the cib, the node doesn't exist. I can't > > oversee what the implications for my cluster really are, but I would be > > delighted to get rid of this ghost node anyway. > > Not sure what's going on. Can you post the hb_report generated > report. > Too late, it "automagically" disappeared by itself after creating and testing a new resource in the cluster. I don't think that I'll be able to find any traces of what exactly the situation was.
Anyway, I've seen it before when teaching Heartbeat in one of my classes. Just by not paying attention suddenly two clusters are present on the same network (and yes, shame on me, using the same port as well ;-). If that happens, the clusters mix up with each other, which is, you CAN see the other nodes when using "crm_mon -1", but if you do a dump of the CIB using "cibadmin -Q", the nodes are not listed. BTW: crm_mon still shows them, even if the entire other cluster has been removed completely from the network. Since these kinds of accidents happen too easily in the classroom / pilotlabs environments that I do most of my heartbeat work in, it would help to know how to get rid of them. > You can definitely get rid of it by restarting the cluster. You > can keep the services running by setting the unmanaged mode in > the cluster options. Check first if any resources are set to the > managed mode---an attribute on the resource takes precedence over > global options. > I do like this tip about the managed/unmanaged mode. Thanks, Sander _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
