Nagios Support Forum

I am monitoring a few network devices, which have a large number of interfaces. I have the device modeled such that each interface is a separate service (since we only want to monitor interfaces that are in the admin up and ifoperstatus up state).

sporadically I'll get a flood of alerts, which I think are just due to switch load, where the switch stops answering snmp for a few minutes (since snmp is a lower priority task for the switch, if it spikes load, snmp will get dropped).

is there a way for me to rate limit alerts for a specific host, so that I can say something like "dont have this host send more than X service alerts within a specific period of time"?

thanks,

-Luke

Unfortunately that functionality isn't available. I would probably go the route of increasing the retry interval or max check attempts value in this case to avoid false positives. Another option would be to use a different check that actually sends traffic through the port instead of querying the lower priority snmp process.