This is currently not easy to do with our alerting system, which is all geared to act on values returned from SNMP. We're not /too/ enamored of the idea of running shell scripts in the poller, as that can cause timing problems for the poller processes.

The old way to do this is via the services system, which was deprecated some time ago and is currently not linked to the alerting system (and probably doesn't work). If/When we rewrite and re-enable this, that's probably the place you'd do it.

Alerting for us is quite a recent addition, so is still very much a work in progress in terms of featureset!

Greetings,

My office recently purchased Observium and we really enjoy the server stats on CPU, memory and storage. It appears that I can set up other health checks as well but I was not able to find any examples in the HTML help files.

What I would like to do is write a simple shell script that returns one of three values: information, warning, and critical and have Observium act on that value with an alert (on warnings and critical error codes).

Does anyone have a brief example they can share? Any assistance would be greatly appreciated. Thanks a lot.

Regards,

David Gietka

NGS Webmaster