Hi,
I'm using lock:timeout in SQL Profiler and traces and would like to identify moments where the user application stops because a lock:timeout.
There is lock:timeout and lock:timeout (>0). Following explanations in the web, I think I should use lock:timeout (>0) to get real timeouts that stops the application.
The problem is that lock:timeout(>0) never gets anything, even when the application stops by timeout. I noticed that even when my applications (and I'm talking about event a simple demo web app only to reproduce this behavior) stops by lock:timeout the duration field is set to 0 and, probably because this, lock:timeout (>0) doesn't get the application timeout.
I identified that if I try to simulate the timeout in SSMS using SET LOCK_TIMEOUT the duration field is filled, but from the application the duration field is always 0.
So, first question is that I couldn't understand this behavior and because that I couldn't create a trace to get only the timeouts that stops the application. Sometimes I get too much, sometimes I get nothing. This happens event if I use lock:timeout (instead of lock:timeout (>0) ) filtering textdata field (ignoring null values) and isSystem (ignoring 1).
Second question : What textdata means ?
I noticed there was two kinds of textdata, one in the form of 1:9239339 and the other in the form of (230390223). What this means ?
Third question : Is there a more in deep explanation of lock:timeouts and all these scenarios, explaining also what are the lock:timeouts with duration=0 ?
Thank you !
Dennes
Dennes - Se resolveu, classifique a mensagem, por favor - [http://www.bufaloinfo.com.br] NOVO DVD Segurança no ASP.NET : http://www.bufaloinfo.com.br/LearingSeriesSegurancaASPNET2.asp