On 4/13/07, sam <[EMAIL PROTECTED]> wrote:
I am evaluating Zenoss for our organization. I have a question that any exist
Zenoss user may already know the answer to -
If a switch or router dies, will Zenoss report errors for servers behind that
router? Will alerts be generated for those servers? I suppose that the
ideal case is for alerts NOT to be generated for those servers.
Is there official documentation that explains what happens in such a case?
Sam,
I can't find any documentation on the subject, but Zenoss does
currently support this behavior. When a system fails to respond to the
ping check, Zenoss walks your route paths back to the Zenoss server to
see if anything higher up the chain is also unreachable. Once it find
the ping failure closest to the Zenoss server it generates an event
with a state of "New." All of the dependent failed events get their
state set to "Suppressed." It is advisable to craft your alerting
rules to only alert on "New" states for this reason.
This functionality is dependent upon Zenoss being able to model your
routers' routing tables properly. If you take a look in your
$ZENHOME/log/zenping.log file and see any messages like, "... not in
topology," then the device mentioned won't have support for event
suppression.
--
Chet Luther
[EMAIL PROTECTED]
_______________________________________________
zenoss-users mailing list
[EMAIL PROTECTED]
http://lists.zenoss.org/mailman/listinfo/zenoss-users
