SCTOTD: Tips for Deep Analysis of Community Management Packs66766

SCLogo System Center Tip-of-the-Day for May 15th, 2010

Community-developed management packs are great, but require special consideration before implementation since they are not tested and supported by Microsoft. But how can you test a management pack to uncover any risks lie within? Here’s a process that will allow you to understand how an MP works under the hood. While we’re talking about community MPs here, this process would work equally well for Microsoft and third party MPs.

MP Review Process

Here are the high level steps in the process with some resource links to help with the details.

  1. Import the MP in a test environment.
  2. Export the MP to unsealed format with the OpsMgr Command Shell
  3. Open the MP in the MP Viewer, export to Excel. This gives you a sortable report you can use to uncover interesting info for MP tuning (see the “4 tips” article below for details)
  4. Install the MP Authoring Console and MP Resource Kit Tools
  5. Run Cookdown Analysis Tool to see which workflows cook down and where you might break cook down with instance-specific overrides. (for more on cook down, see the “more information” section below)
  6. Run the MP Best Practices Analyzer to see what violations may exist and their criticality
  7. Import and configure the MP in a test environment and see what errors result.
  8. Have a look at the objects discovered by the MP on a single host to get an idea of how the MP might scale. If you’re introducing an MP that discovers thousands of objects that are targeted for monitoring, performance impact could be significant and “config churn” a definite possibility in a larger production environment. (see “What is config churn” link for more info.
  9. Look for changes in CPU and memory utilization on RMS and agents to see if the MP is causing CPU or memory pressure. (see the “which scripts are consuming resources” link below for more info)
  10. After a day or two in test, use the Top Generators Reports to see if the MP is generating a lot of data in the OpsMgr DBs, and if so, which workflows are responsible for it (see “Database Hygiene Part 6” for more info)
  11. Tune discovery, monitoring thresholds, event generation and performance collection and observe behavior for improvements
  12. Use the Top Generators Reports again to see if your changes have improved the situation.

At this point, you should have a feel for the potential impact of the management pack on your environment and how you can tune settings to suit your environment.

More Information (Links)

Database Hygiene Part 6: Identifying and Tuning Top Generators with the Operations Manager Core MP Reports

OpsMgr and SCE: 4 Tips for Proactive Management Pack Tuning

OpsMgr MP Authoring and Clarification on Cook Down and OnDemand Detection

http://blogs.technet.com/kevinholman/archive/2009/10/05/what-is-config-churn.aspx

MP Authoring Reskit: How to Verify MP Best Practices in the OpsMgr 2007 R2 MP Authoring Console

MP Authoring Reskit: Analyze and Troubleshoot Workflows More Quickly with the Workflow Analyzer

How to identify which script is consuming resources in OpsMgr

Previous Installments

Follow System Center Central

RSS | TWITTER

Sponsors and Supporters

Quest | Veaam | System Center Influencers

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.