On Tue, 2008-06-10 at 17:24 +0200, Maurizio Rottin wrote: > hi, > i had a problem with a cluster node (i call it nodeA for now on). > It suddently stopped it's service without notifing the other nodes (or > maybe it did notify, but the other nodes did not get the packets). As > a result, no one was able to get a lock on a gfs filesystem shared > across 5 nodes. (and no one was trying to fence it). Could you get cman_tool services output ? > Perhaps it was a network problem, because the nodeA was still working, > or a ccsd problem. > > Anyway, since if it happen once it can happen again, is there any way > to get the cluster send emails or snmp messages on nodes critical > events? You can set syslog to do email, I think. -- Lon -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
