Johan Hoeke wrote: > I observed the following behavior while testing the cluster. I stopped > the service foo (blackboard actually) manually using the lsb init script > on julia, the active node. The cluster notices this because op the > monitor operation. The resource foo (blackboard_init) fails on julia. > The cluster stonith-s node julia, but the non-failed resources keep the > status "running on julia" until julia is done rebooting. After that's > done the resources fail over to node koch. (I monitor the progress of > the failover with the gui - sorry ;) ) > > Obviously I want the resource to start on the other node right away, not > have them wait until the first node recovers. Why does my cluster act > this way? It _seems_ to have happend since upgrading to 2.1.3 but I'm > not sure. > > t.i.a. for the help, > > regards, > > Johan > > > > note on the hb_report's: koch contains the problem discription etc >
LS, Didn't run hb_report properly at first. Have done so now and fixed some issues noted in analysis.txt. Will close this thread and post a proper hb_report with hopefully some clearer questions. regards, Johan
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
