I'm still having timeout for 2 minutes (and high CPU usage) when zenhub starts between this two messages.
2009-04-26 00:59:34 INFO zen.plugins: Loading collector plugins from: /opt/zenoss/ZenPacks/ZenPacks.BlakeDrager.fping-1.0-py2.4.egg/ZenPacks/BlakeDrager/fping/modeler/plugins 2009-04-26 01:01:29 INFO zen.plugins: Loading collector plugins from: /opt/zenoss/ZenPacks/ZenPacks.Nova.Cisco.Mibs-1.0-py2.4.egg/ZenPacks/Nova/Cisco/Mibs/modeler/plugins Chet Luther wrote: > On Apr 25, 2009, at 8:06 AM, Anton Menshutin wrote: > >> I also found that after starting other daemons this periods of high >> CPU >> Utilizations also happen. >> In zenhub.log a have >> DEBUG:zen.hub:Servicing getThresholdClasses in PingConfig >> DEBUG:zen.hub:Time in getThresholdClasses: 103.36 >> This 103 seconds hang was one of those. >> >> And the other problem, may be related, is the following error. >> CRITICAL:zen.hub:ZenPing 'zenoss-hostname' not found, ignoring network >> topology. >> >> And then comes all device listing, saying that they all are not in >> topology. How can I fix this? >> > > That 103 seconds in getThresholdClasses was recently discovered to be > a major time wasted in zenhub. There is a fix for it that you should > be able to patch in. Run the following commands as your zenoss user to > apply the patch. > > zenpatch 12983 > zenhub restart > > Ticket #4808: remote_getThresholdClasses needs to be cached > http://dev.zenoss.org/trac/ticket/4808 > _______________________________________________ > zenoss-users mailing list > [email protected] > http://lists.zenoss.org/mailman/listinfo/zenoss-users > _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
