NSEs are not being processed and the Altiris Console is too slow


Article ID: 180711


Updated On:


Management Platform (Formerly known as Notification Server)





The Altiris Console seems too slow and NSEs are taking a long time to be processed.

I started noticing the Altiris Console running slow around a certain time in the day. No modifications or updates were done during the day.

The EvtQFast and EvtQueue are not processing any of the coming events. NSEs already in the Process folder stay there. If I delete them from the Process folder, no new ones come in. If I place them in the Process folder they stay there.

The Altiris Services (Altiris Service and Altiris Client Message Dispatcher) are running. The Notification Server was rebooted but still the same issue. The only error that was found occurring frequently is the EventRouter one (see article 20282, "EventRouter::DeliverPostData caught unexpected exception").

The logs don't show that the NSEs are bad or that the Notification Server is busy. The SQL server (separate from the Notification Server) is not reporting deadlocks or lost connectivity, nor does it show connection problems.

If I check the PauseActivities key (HKEY_LOCAL_MACHINE\SOFTWARE\Altiris\eXpress\Notification Server), it is set to 0 as expected. As well the EvtQFastFull and PausedNSMessaging keys are set to 0.

If I check the MaxConcurrentFastMsgs and MaxConcurrentSlowMsgs those are already set to 5 and 2 respectively as article 19750, "Notification Server Maintenance," suggests.

New NSEs are still coming in but no process in the queues is done.

Notification Server 6.0.6074 SP3
SQL Server 2000* (on a separate server)
Windows Server 2000*

The SQL Server was not processing requests from the Notification Server, causing the NSEs not to be processed properly. Even though the Notification Server didn't show any complaints regarding this in the logs, we assume that the SQL Server was busy with other internal tasks causing the requests from the Notification Server to be queued and taking longer in being processed.

After rebooting the SQL Server, the queues (EvtQFast and EvtQueue primarily) started processing the NSEs.

Note: If the SQL Server seems to be working as usual, but you notice that the Altiris Console is acting too slow or slower than before, it is recommended to evaluate the SQL Server performance and how it is maintained (referring to if there is a SQL Server Maintenance Plan put in place as part of the best practices for most companies).

The following are articles suggested by Microsoft regarding the SQL Server performance:

  • "How to configure SQL Server to use more than 2 GB of physical memory" (MS-KB274750).
  • "Enabling AWE Memory for SQL Server" (MSDN reference).
  • "SQL Server 2000 Backup and Restore" (TechNet article).
  • "Microsoft SQL Server 2000 Index Defragmentation Best Practices" (TechNet article).
The following are articles created with basic concepts for the SQL Server Performance improvements that were suggested to our customers in the past:
  • Notification Server Maintenance (19750)
  • SQL and large memory-external (3977)
  • How To: SQL Tuning and Maintenance for Notification Server (17079)
  • SQL Tuning and Performance (28122)

The following steps are basic suggestions to help the SQL Server to be in good shape and improve the processing/request time with the Notification Server:

  • Enable the AWE switch (if sufficient physical memory is available)
  • Purge the event tables for the Notification Server database (See article 21310).
  • Defrag the database indexes periodically.