Re: Command.CommandTimeout bug
Posted by:
Dr. No
Date: October 26, 2006 01:50AM
Reggie,
I think my initial formulation regarding ExecuteReader was not correct. Let me try to explain correctly:
1. If the CommandTimeout is set to 0 I would expect that the timeout is disabled and the command will take whatever time it takes to complete. I believe this is implemented this way in other data providers (MSSQL, Oracle). However this is nowhere specified of course and therefore is not a must - but it would be better to my humble opinion. Otherwise it would not be possible at all to specify a infinite time.
2. The current implementation is not done this way. When I interpret the source code correctly then a thread for command cancellation is created in any case for an ExecuteReader. Therefore if one sets the command timeout to 0 then the execute reader will very quickly terminate.
Do you agree with my opinion?
Subject
Written By
Posted
Re: Command.CommandTimeout bug
October 26, 2006 01:50AM
Sorry, you can't reply to this topic. It has been closed.
Content reproduced on this site is the property of the respective copyright holders.
It is not reviewed in advance by Oracle and does not necessarily represent the opinion
of Oracle or any other party.