April 26, 2007 at 11:29 am
I have a nightly job that copies a production database to a reporting server. A simple backup/copy/restore.
For the past two nights, the job has failed with the following mesage in the restore step, but the SQL Server Job reports success.
Executed as user: ALPHAPROD\SQLAdam. OLE DB provider "SQLNCLI" for linked server "Pricing" returned message "Query timeout expired". [SQLSTATE 01000] (Message 7412). The step succeeded.
Anyone know why the job fails, but SQL Server reports it as succeeding?
Eric
April 26, 2007 at 12:52 pm
Longshot- is there a chance that the job is configured to report success even though it failed? i.e., is it possible that the "On failure action" is "Quit the job reporting success"?
April 26, 2007 at 1:17 pm
It was the first thing that I checked.
Great minds think alike....
Eric
April 26, 2007 at 1:27 pm
I saw your replies in other threads and was sure that my proposed solution was way too basic Was worth the shot...
The only other thing I can think of is that timeouts (especially over linked servers) are not interpreted as errors. In other words, the timeout may not be returned as an error to the driver at the server where the job runs. I'm not sure if this is the case in practice, but I wouldn't be surprised.
It's very easy to test- NCLI implies that you're using SQL 2K5 so you can change the config option for remote query timeouts on a remote server to be small, and then run a WAIT command as part of a job and let it timeout and check the error report (if any).
In the meantime I'll scratch my head and try to think of other possible issues...
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply