Hi,

I'm having a bit of an issue settings up and dependency rollup for a process monitor that I have targeted at 2 of my servers. I'll give you a brief description of what I need to achieve and then details the issues I'm encountering.

Requirement

We're running OpsMgr 2012 R2. I have one bespoke process named master.exe that's running on 2 of the my windows 2008 R2 servers. The process needs to be running on at least one of the servers for the application to function correctly. We only require an alert if the process is not running on both servers.

Setup So Far

1. Created a group named "master group" that is populated with the windows object of the 2 servers "server1" and "server2".
2. Created a process monitor using the process monitoring template that is targeted at the above mentioned group. This monitor works correctly and changes the state of the object to critical when not running.
3. Disabled alerting on the process monitor.
4. Created a group (master group dependency) that is populated with the monitor for server1 and server2.
5. Created a dependency rollup monitor with the following settings -
- Target: master group dependency mentioned in step 4
- Parent Monitor: Availability
- Monitor Dependency: Object (Contains Entities) > Entity Health > Availability
- Heath Rollup Policy: Best state of any member
- Alerting: Enabled

Issue Encountered

At first the setup seemed to work well. There were no alerts with a single process not running but when both stopped the alert appeared, happy days. The problem is the alert generated by the rollup is randomly appearing. I check the health of the process within the monitoring pane under the "windows service and process monitoring" folder "process state" view and can see that the health of one of the server processes health state has not changed to critical since the monitor was configured. I also confirmed that the process has not stopped on the server.

I'm not sure why the alert is generating if the process is running on at least one of the servers. I'm also unsure if the above setup is correct or do I need an aggregate rollup.

Thanks in advance,

Simon

• Edited by Sibow Wednesday, May 20, 2015 9:15 AM

Hi,

Found a blog which seemed helpful for our question, please refer to the link below.

Intelligent Service Monitoring Part II: By Example

http://thoughtsonopsmgr.blogspot.com/2011/09/intelligent-service-monitoring-part-ii.html