I have a networking device whose SNMP process is notoriously unstable and will die occasionally. This problem resolves itself after about 20 minutes or so but until then I get alarms that I don't need to pay attention to. Is there a way to either:

1. Monitor SNMP itself on the device and use that as a dependency

or

2. Differentiate between "interface down" and "error 2003" such that the latter does not notify as if the interface is down.

Article Comments

The uptime value of the device in case is probably not available when the SNMP process dies. Our suggestion would be to add a SNMP Uptime sensor and use the same as a dependancy for the entire device (this would pause the respective sensors without triggering notifications).


Jul, 2014 - Permalink