Hello,
Let's assume I have a "deep" network topology (with many parent hosts above
the Nagios box), such as:
Nagios Server => Switch 01 => Switch 02 => Firewall => Packet Shaper =>
Router => ...
When the 'parents' directive is properly setup and "Switch 01" goes down,
check-host-alive tells me it went down, and supresses notifications for all
the other hosts.
Now, if "Switch 01" if heavely loaded and my fping service finds out it has
a 500ms rta, all other hosts above it (switch 02, firewall etc) will have
response times equal or bigger than 500ms.
My question is, in this scenario, how can I supress critical messages from
the fping service regarding all other hosts which are slow because of
switch 01 ? (having fping figure out that, since the host above it is slow,
it doesn't need to notify me about all other hosts)
Perhaps this could be implemented in a future Nagios version (an option in
nagios.cfg such as "apply_parents_directive_to_network_latency")
PS: This could be accomplished with service dependency (maybe?), but it
wouldn't be practical since I would have to create several dependency
checks for each parent host, which would make it really hard to maintain.
Regards,
Rainer Alves
Unisys Brazil
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]