Slow Query In Profiler Shows 0 CPU, Reads, or Writes

  • I've been looking at a long running procedure in Profiler and can't quite see why it is showing such poor duration values. I have a profiler trace set up to just look for procedures taking longer than 3 seconds to complete. I won't see anything for a while and then all of a sudden I see 10 or 15 calls to this one procedure show up, each taking a little bit less time that the previous to complete. As if they were all launched one after another, got blocked by some thing, the resource freed up and they all completed at the same time. What's weird is that each one shows 0 CPU, Reads, or Writes. So I guess my newbie question would be, is this just some blocking problem that I'm not detecting? I never see this procedure when I run dbcc opentran, and I do not see blocking processes returned from sp_who2

  • I found the cause of the problem, an incorrect web service was installed on a server that was making the calls. Still weird that I never saw a block or an open transaction though.

  • Did you have any filters applied besides the time? It's possible that something held you up, but it didn't take a long time itself to complete.

  • Duration was the only filter. I wish I knew a little more about the web service making the call, but as soon as I asked a developer a question about it they had replaced it with the current one an gave the old "move along, nothing to see here...". Oh well, it's all working properly now.

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply