Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
sureshot
Yak Posting Veteran
72 Posts |
 Posted - 2006-05-25 : 11:43:39
|
| Running SQL2005 and on occasion I see durations of 928809 or even 2830562 in the trace for commands that usually run in the 7000 range. Reads are still in the 1200's even during the huge duration spikes. These are calls from ASP.NET ADO.NET and there should be a 30 second timeout on them by default to boot. I see a few other times that we've logged timeouts so I know the timeout is working at times but not at others. Any ideas what's going on both with the strange serious outliers and the fact that these queries aren't timed out properly? |
|
|
sureshot
Yak Posting Veteran
72 Posts |
Posted - 2006-05-25 : 11:53:53
|
| From the very handy latest article here...Duration. This is the actual duration of the batch. In SQL Server 2000 traces it is in milliseconds. In SQL Server 2005 this is reported in microseconds (one millionth of a second) however Profiler 2005 automatically translates it back into microseconds.Oops, I got surprised by a SQL2005 change in behavior. |
 |
|
|
|
|
|
Huge durations in trace