Hi Thanks but I never asked "to run monitoring on an unmanaged resource " ... ? ! I ask for the opposite : a way to set one resource in a state near to "umanage", meaning "umanaged and wo monitoring", and wo to be forced to set all the cluster-management umanaged with maitenance-mode=true.
I think that, with regards to the responses, this function does not exist ... Alain Le 27/03/2013 08:24, Ulrich Windl a écrit : > Hi! > > I see little sense to run monitoring on an unmanaged resource, specifically as > some _monitoring_ operations are not strict read-only, but do change the state > of a resource (which may be quite unexpected). One example is the RAID RA, > which tries to re-add missing devices. > > > Regards, > Ulrich > >>>> Moullé Alain<[email protected]> schrieb am 27.03.2013 um 07:56 in > Nachricht > <[email protected]>: >> Hi >> OK thanks, but sorry it was not quite the response I was expected as I >> already know >> all that about cleanup, reprobe, etc. So more clearly my question was : >> Is there a way by crm to invalidate the monitoring temorarily for one >> specific resource ? >> >> Thanks >> Alain >>> Hi, >>> >>> On Mon, 25 Mar 2013 16:25:54 +0100 Moullé Alain <[email protected]> >> wrote: >>>> I've tested two things : >>>> >>>> 1/ if we set maintenance-mode=true : >>>> >>>> all the configured ressources become 'unmanaged' , as displayed >>>> with crm_mon >>>> ok start stop are no more accepted >>>> and it seems that ressources are no more monitored any more by >>>> pacemaker >>> Probably maintainance-mode also tells the cluster-manager to completely >>> stop monitoring. >>> >>>> 2/ if we target only one resource via the crm resource umanage >>>> <resname> : >>>> >>>> it is also displayed unmanage with crm_mon >>>> ok start stop are no more accepted >>>> BUT pacemaker always monitors the resource >>>> >>>> Is there a reason for this difference ? >>> Its un-managed, not un-monitored ;-) >>> Actually this is not a problem, it will monitor as long as the service >>> is up. As the first monitor-action fails, the resource is marked as >>> "failed" and no more monitor action is run. Until you explicitely ask >>> for it with "cleanup <resource>" or "reprobe <node>". >>> >>> Have fun, >>> >>> Arnold >>> >>> >>> _______________________________________________ >>> Linux-HA mailing list >>> [email protected] >>> http://lists.linux-ha.org/mailman/listinfo/linux-ha >>> See also: http://linux-ha.org/ReportingProblems >> _______________________________________________ >> Linux-HA mailing list >> [email protected] >> http://lists.linux-ha.org/mailman/listinfo/linux-ha >> See also: http://linux-ha.org/ReportingProblems >> > > > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
