Thk, and i do reset the value after the action juste to make sure, maybe the user will do some mod that need to be logged in the same session, we never know!
Alain
>>> a question remains, is the context info reseted when a session is created?
>
>I did a rough test. Filled the value with 'Hello World', checked it, got the spid
>
>closed the connection and reopened it. verify that I had the same connection.
>the value was cleared.
>
>now, you really need to do a more through test and remember that there are options for the connection string that determine whether an explicit reset is made when the connection is pulled from the connection pool.
>
>Another option would be to reset context_info after explicitly setting it.
>
>-Mike
>
>>I've found a workaround using a mixture of different sugestions.
>>
>>first set the context_info of the session
>>
>>declare @BinarySessionId varbinary(128)
>>select @BinarySessionId = convert(varbinary(128), @@spid)
>>Set context_info @sessionid
>>
>>Then add in the Trigger :
>>Declare @BinarySessionIdToSkip varbinary(128)
>>SELECT @BinarySessionIdToSkip = context_info
>> FROM master.dbo.sysprocesses
>> WHERE SPID=@@SPID
>>
>>If @SessionIdToSkip <> convert(varbinary(128), @@spid)
>>begin
>> --do logging process
>>
>>end
>>return
>>
>>a question remains, is the context info reseted when a session is created?
>>
>>>Thk for the reply i'll look into it next week
>>>
>>>>What about using an INSTEADOF trigger. Within it you can query the sysprocess table looking for a flag in the context_info section. You can use SET CONTEXT_INFO to put a value in there.
>>>>
>>>>As another thought, if there are situations where the behavior needs to be circumvented, maybe a trigger is the wrong tool.
>>>>
>>>>-Mike
>>>>
>>>>
>>>>>Hi,
>>>>>
>>>>>is there a way to do a update command without firing the update trigger for a session?
>>>>>
>>>>>we can't use the "alter table table1 disable trigger UpdTrig1" because it will affect all session.
>>>>>
>>>>>
>>>>>I was thinking of using the not for replication option of the trigger.
>>>>>Is there a way to "fool" the trigger to believe that i'm in a replication update?
>>>>>
>>>>>Alain
