Keith, would you mind providing more details about "When we get CPU alarms from a Windows host, we ask the processes probe to give us the process list and grab the top three by CPU utilization." I would love to implement something similar. (Actually, I would really love to see it built into the CDM!)
Using an AO profile, or calling nexec remotely (I'd definitely like to know how to do this!) requires connectivity to the NAS, correct? That means that my script responses can only be triggered if the monitored server can connect to the NAS and the NAS is working properly.
Ideally, I would like the ability to configure a response within each probe. Perhaps this could be associated with the Message Pool so that when a particular message is generated the response can be triggered, but better than that what I would really like to see is "problem detected/trigger response to attempt recovery/if successful send Information message/if unsuccessful send critical message".
Alternatively, why wouldn't I just run a NAS (or better yet, a NAS-lite) on all of my servers that can listen for alarms and trigger a response?
Retrieving data ...