On 12/04/2013, at 6:52 AM, Keith Ouellette <keith.ouelle...@airgas.com> wrote:
> We am using Nagios for network monitoring and for Pacemaker, we are using > check_crm to provide status for nagios. This works very well if the failure > is still seen by Pacemaker at the time nagios polls. If the failure > condidtion causes a switchover, but the recovery is before the nest Nagios > poll, Nagios does not report any issue, because there isn't one at the time > of the poll, even if the transition of the resource changed to another node. > > Is there a way to know that the transistion occured without scraping the > messages log in a set interval and looking for only new events that state a > transition or "Members Joined"/"Members Left". It appears that failcount only > increments on a failure, but not on a transition imposed by someone manually > puttintg a node into standby. > > I am just trying to figure out a way to trigger an alarm for Nagios any time > resources get moved between nodes, so the cause can be investigated. > > Any ideas? Run crm_mon in daemon mode and use --external-agent to write the notifications to disk somewhere (or push them into nagios)? _______________________________________________ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org
