Monitoring Applications, not server downtime

I’ve been configuring application subgroups. However I come into trouble with my alarms and what they actually mean. For example, the health of my servers is always in a warning state because it reports warnings for alarms such as this:

Health of CDM Index Servers is warning.
Root Cause :

CPU Utilization is now back to normal, current value is 72%: Top 3 Processes:-Idle - 19.871%;PCDOIServer - 19.736%;System - 13.383%

Which should be a clear alarm as far as I can see....

Additionally I find that the availability figure is not configurable. Say I have a particular service that, if it is down, means the availability of my monitor is critical. It seems to have automatically decided that if that service or process is down, this affects health and sets it to warning. I see no way to change this and it's a real deal breaker for my environment. I was expecting this to be an application monitor, not just a server downtime monitor.

I should add that I've been using the OpManager addon for this.