Cross Platform Cumulative Update 3 for Microsoft System Center Operations Manager 2007 R2 is now available!

Release notes for this update

  • This update is only for cross platforms including the Linux operating system and the UNIX operating system.
  • This update is released as built-in components that are included in Cumulative Update 3 for System Center Operations Manager 2007 R2 instead of as a separate update.
  • Cross platform updates are language neutral. Therefore, this update is released in English only.
For more information about Cumulative Update 3 for System Center Operations Manager 2007 R2, click the following article number to view the article in the Microsoft Knowledge Base:

2251525 (http://support.microsoft.com/kb/2251525/ ) System Center Operations Manager 2007 R2 Cumulative Update 3

Issues that this update resolves

The following issues are resolved in Cross Platform Cumulative Update 3 for Systems Center Operations Manager 2007 R2.

Issues

Before you install this update

After you install this update
The default polling interval may cause excessive CPU usage in monitors that are created from the service template. The default polling interval is 30 seconds. This interval may cause excessive CPU usage on Solaris-based computers that run lots of processes. The default polling interval is changed to 300 seconds.

Note If you have an override in an existing template, you may have to override it again after you import the new template.
The default discovery interval for Linux objects or for UNIX objects may cause excessive CPU usage. The default discovery interval for Linux objects or for UNIX objects is 3600 seconds (1 hour). This interval may cause excessive CPU usage on monitored computers. The default discovery interval is upgraded to 14400 seconds (4 hours).
Linux management packs and UNIX management packs report errors in Management Pack Best Practices Analyzer. These errors have no operational impact. However, the errors are unnecessarily displayed when you run Management Pack Best Practices Analyzer. These errors are not displayed when you run the Management Pack Best Practices Analyzer.
Monitors in Linux management packs and in UNIX management packs incorrectly generate alerts when performance counters return NULL values. Performance counters do not return performance values if a multi-segment logical disk is monitored on a Linux operating system or on a UNIX operating system. Instead, the performance counters return NULL values. These NULL values cause an alert to be incorrectly generated. This issue is resolved.
The agent file for monitoring the HP-UX operating system uses excessive memory when HP-UX enumerates lots of processes. The internal design of the agent file that is in HP-UX causes excessive memory to be used when HP-UX enumerates lots of processes.  The design is changed to significantly reduce memory usage.
The agent file for monitoring the Solaris 10 operating system leaks handles. If an error occurs while the /proc as file is read in a Solaris 10 operating system, a file handle leak may occur. After a while, the agent file on Solaris 10 becomes unresponsive. This issue is resolved.

Additionally, the following feature addition is added in this update to control the rules that generate log file alerts in Linux management packs or in UNIX management packs:

  • The default behavior combines duplicate log file entries into a single alert. After you install this update, you can generate a separate alert for each log file entry.