I'd like to monitor a few processes for statistical purposes only. If they are down, it's OK - they aren't "critical" or "warning" or anything like that. E.g. whether a certain application was active on our server, and for how long.
What could be simpler, right? Set up a monitor, and figure out a way to exclude it from generic alerts we already have, so they don't get triggered by that monitor?
That's exactly the challenge:
- edit alerts one-by-one, add a condition to "trigger conditions" or "suppression" to exclude that monitor, or a class of monitors
- test each one to make sure it still works as it should
- remember to exclude that condition in each new alert we set up
Is there a simpler, more straightforward way?
Also, component monitors don't have custom properties (as far as I could tell) which I could configure so that my alerting knows if the component is critical or not. Does it mean I have to rename a component monitor to include a certain phrase in its name (e.g. "Word UI (statistics only)" or "Word UI (non-critical)" and then configure my alerting to ignore those?
Thanks!