Ex: SP ODBC RetryOnError

Paul Henson henson at signet.id
Thu Dec 8 20:54:26 UTC 2022


On 12/8/2022 12:47 PM, Paul Henson wrote:

> ODBC Error: 40001:1:1205:[Microsoft][ODBC Driver 17 for SQL Server][SQL
> Server]Transaction (Process ID 79) was deadlocked on lock resources with
> another process and has been chosen as the deadlock victim. Rerun the
> transaction.
> ODBC Error: 01000:2:3621:[Microsoft][ODBC Driver 17 for SQL Server][SQL
> Server]The statement has been terminated.
Hmm, I might have been too quick on the draw on this one; I just got an 
update from the client saying that he sent me the log excerpts from 
their log analysis platform, not directly from the source shibd log, and 
the 3621 might have actually been logged before the 1205, which would 
make more sense and work fine with the current implementation 8-/. He's 
going to go back directly to the source log and verify.

Sorry for the possible noise, although I am still curious from an 
abstract perspective as to whether or not it is a valid assumption that 
the last error returned if multiple errors come back from an ODBC call 
is the only one significant for the purposes of determining whether or 
not to retry.

Thanks again…


-- 
Signet - The Art of Access
https://www.signet.id/




More information about the users mailing list