June 11, 2003 at 8:40 am
I've been getting a Timeout expired message off and on from our enterprise app. The error comes from our data layer, in that it doesn't receive a response within 30 seconds, the default amount of time that it will wait.
All of our pages come back in a second or less, so any page taking longer than 10 or 15 seconds means there is a problem. I know of the error because we have the site automatically email errors to us. When I immediately go to the site after receiving the error, everything is fine.
I recently employed a .bat file that called a SQL file via osql and ran the following query "sp_lock" and redirected the output to a text file every 10 minutes to see if any tables were being locked. When it happens to execute at the same time as this problem occurs, I see only "Timeout expired" in the text file, meaning I'm assuming that it isn't just a table locking problem since it is affecting it as well.
Any ideas?? I'm pretty much at the end of my rope on this one and all the customers are proably getting sick of seeing this error.
June 11, 2003 at 9:19 am
I would look at a profiler dump to see what is going on. There are several scripts that can capture profiler traces based on an error here. It could be a number of issues other than just locking. It could be a time out waiting for resources at the page level you could have some kind of disk contention issues that keep SQL from filling request. Without more information it makes it difficult to troubleshoot the problem.
Wes
June 11, 2003 at 9:29 am
What information do I need to give in order to make it easier to diagnose?
June 12, 2003 at 11:26 am
We've gotten this error about 30 times today, it's getting pretty bad. If anyone can help out I'd appreciate it. I ran SQL Profiler and it appears that when this is happening there are no stored procedures executing at all and no connections can be made to the server so I don't think it is a table locking issue.. Any ideas why the server would just stop accepting connections for a brief amount of time and then suddenly recover and everything be fine?
June 12, 2003 at 12:24 pm
Can you pull server versions MDAC version and service pack levels? It may not be the machine have you had your network guys take a look at thier end? Can you verify its not a web server problem? If you log into QA can you execute code against the db when it is having this problem?
Wes
June 12, 2003 at 12:30 pm
The error appears for such a brief period of time I can't ever do any troubleshooting while it exists. Service packs and MDAC are up to date and 98% of the time the site is very fast and works very well.
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply