So the way Zenoss handles topology-awareness and alerting is that zenping attempts to reach devices and maintains a topology in zodb. When a daemon cannot reach a device, it is supposed to check the zodb before alerting. Unfortunately there are gaps in which daemons check in to see what is currently marked as 'down' by zenping. zencommand is currently being fixed to become topology-aware, I'm guessing that's the offender here. http://dev.zenoss.org/trac/ticket/3445 There are tickets opened against some of the other daemons already. http://dev.zenoss.org/trac/ticket/4783 http://dev.zenoss.org/trac/ticket/4784
Thanks, Matt Ray Zenoss Community Manager community.zenoss.com [email protected] On Apr 2, 2009, at 2:24 PM, troyaden wrote: > Hello all, I am hoping someone can provide me with some clear > instructions on how to configure Zenoss to not send alerts from > devices downstream from the device that had the original failure. To > put it in more practical terms, if Zenoss loses comunication with > router A but there is a Router B behind (downstream from) Router A > and that router has over 100 nodes behind it that Zenoss is also > monitoring why is my obviously misconfigured zenoss sending alerts > for all the devices downstream as well as the original failing device? > > And, the most important part, how can I go about fixing this so I do > not get over 700 msgs next time this happens? > > Any help, (hopefully step by step) would be most helpful. > > Thanks in advance!!!!!! > > > > > -------------------- m2f -------------------- > > Read this topic online here: > http://forums.zenoss.com/viewtopic.php?p=32963#32963 > > -------------------- m2f -------------------- > > > > _______________________________________________ > 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
