DIfferent Incidents with different thresholds for similar measures.
This is most likely a very simple issue however I cannot seem to figure out the right combination.
We have a very simple measure, a copy of the CPU total time that we have an incident based on. This works great. However the problem is that I would like to have separate incidents or separate alerting for different groups of servers at different levels. For example, ave. utilization of 75% on server group A warrants an alert however on group B, I want an alert at only 50%.
So I have another copy of the same measure, basically the same however named differently, with different warning & severe thresholds, but the only way I see to limit this measure to certain servers is in the details tab and "calculate by agent group", so I select the appropriate agent group, and now I don't see anything when graphing this measure, no measurements available. If I deselect that, then I get results, but for ALL agents in my profile.
Either I am doing something simple wrong (very likely), or perhaps there is a better way of alerting on the same type of measure however with different thresholds for different agents. Thanks!!
Pretty much except that the first measure isn't limited to any particular agent group, and it behaves as expected, for all agents.
If I undstand you right you created 2 CPU total time measures with different thresholds (the first one restricted to the first agent group, the other one restricted to a second agent group). After that you created 2 different incidents each based on one of the measures. From my perspecive I cannot see any mistake and I would have done it like you. The interesting thing is that the measures show up without restriction but not if restricted for a specific agent group.
