April 7, 2009 at 4:11 pm
About 4 weeks ago, we did an upgrade from sql 2000 to 2005 on the cluster. We decided to do a complete wipe/install process from the OS on.
The web application (classic asp, not asp.net) is now encountering random timeout errors. These were not happening prior to the upgrade.
The app uses a vb6 com-based object to do data access. It creates an ADO recordset, establishes a connection, executes the command, closes the connection and frees the recordset.
The timeout is set to 60 seconds.
From the application logs that we've put in place on this object, we can see that there is plenty of successful connections going on immediately prior to and after the timeout (and even during the same period from other users).
The command being run normally runs in under 1 sec. The logs also have the command being run, so we can manually run them to verify. Other calls to the same procedure (different parameters) were executed just prior to the error. I can run the command with the parameters that failed, and get results in < 1 sec.
I'm confident that this is not a sql issue... yet others are confident that it is. So, where to start looking? I've looked over the forum here at many of the timeout expired issues, and everything that I've read points to application issues.
Since we rebuilt the cluster, is there anything there that should be looked at to ensure everything is correct there?
Thanks,
Wayne
Wayne
Microsoft Certified Master: SQL Server 2008
Author - SQL Server T-SQL Recipes
April 7, 2009 at 4:24 pm
I think it would be good for you to change your DAL(data access layer) because ADO.NET uses in memory database and uses connection pooling so many Command related issues are cleaned up. A quick fix you could make minor changes to the ADO layer of the application.
Kind regards,
Gift Peddie
Viewing 2 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply