The OBM Management Pack for Infrastructure does not document the logic of its policies. This makes it very difficult to:
- Know what the policies actually monitor?
- Which policy is relevant for which type of server?
- How to tune the parameters for the specific requirements of a business?
- How to test the policies?
For instance, for memory monitoring, one may wonder what's the difference between the policies Sys_MemoryUsageAndPerformance and Sys_MemoryBottleneckDiagnosis. I have often seen customers prefer Sys_MemoryUsageAndPerformance because it has a simpler set of thresholds, however this policy is rarely relevant and the combination of Sys_MemoryBottleneckDiagnosis and Sys_SwapCapacityMonitor is a far better choice in the majority of cases.
In order to come to this conclusion, one needs to carefully read through the policy scripts and have in-depth knowledge of how an OS performs memory management, and also clearly understand the difference between capacity monitoring and performance bottleneck monitoring. In other words, while the OBM Management Pack for Infrastructure should provide an out-of-the box solution that encapsulates the expertise, it actually requires to be an expert to know how to use it correctly.
I attach a document that provides detailed information for a few key policies. This type of information should be documented in the manual for all policies.