Geeks With Blogs
Hornet's Nest A few of Mike Hoerner's Favorite Topics

Recently, we built a new two-node Exchange 2003 Cluster.  Both servers run AMD processors.  After building the Exchange Cluster, we ran the Microsoft Exchange Best Practices Analyzer tool.  We noticed several Domain controller server response time and Global catalog server response time critical issues listed in the Best Practices Report.


After reading the following Microsoft Exchange Server Analyzer articles, we started troubleshooting the reported critical issues.

Round-trip times to Active Directory server are taking more than 100 ms

Round-trip times to global catalog server are taking more than 100 ms

We ran DCDiag.exe /s:<domain controller> and NetDiag.exe on each of the cluster nodes.  We checked the Event logs for any network-related and other error messages on the cluster nodes and Active Directory Domain controllers.  We checked the network configurations on each of the cluster nodes.  Everything checked out and we did not find any reported problems.

When we ran the command: ping –l 4096 –n 6 <domain controller> on each of the cluster nodes to emulate the Analyzer’s ping behavior manually, we noticed the response times were over 200ms (it should be less than 10ms).

After researching further, we determined the pings were reporting inaccurate information because the servers were running AMD processors.

With that in mind, we added the /usepmtimer switch in the boot.ini file on both cluster nodes to prevent the Exchange Best Practices Analyzer tool from red-flagging this as an issue and to allow the pings on each cluster node to report accurate information.

For more information, check out or search for /usepmtimer on Google.

Posted on Monday, February 8, 2010 5:15 PM | Back to top

Comments on this post: Exchange 2003 Server Round-trip times to Active Directory > 100 ms

No comments posted yet.
Your comment:
 (will show your gravatar)

Copyright © BWCA | Powered by: