Archives of the TeradataForum

Message Posted: Fri, 25 Apr 2003 @ 13:25:35 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Dave,

I didn't see anyone reply on this one, so I'll throw in my two cents.

Are you certain that that you are getting 2631's because of a block on database objects? The best way to determine that in real time would be through PMON. The best way after the fact would be through the use of the DumpLockLog utility. This utility will identify what sessions are being blocked, what session is blocking, what object is being blocked, and for how long. That, combined with the 3111 messages that, I believe, show up in the Software_Event_Log, should give you good insight into the problem.

However, a 2631 error can also occur when the system is blocked on other resources that you have little direct control over. For example, when NCR first implemented intra node message passing in V2R3 (think of it as the ByNet implemented in memory for vprocs on the same node), there was a problem that caused lots of 2631 errors when the system was very busy. I believe this only occurred on very large systems with "less than optimal" workloads. However, the point is that a 2631 can be generated for reasons other than the typical "timeout waiting for a database table".

I would suggest pursuing the first path with PMON, DumpLockLog, etc. If that is not conclusive, I would then start to investigate other possible causes.

Regards,

Thomas F. Stanek
TFS Consulting, Inc.
www.tfsconsulting.com

	<Prev	Next>		<<First	<Prev	Next>	Last>>