I have been engaged on an analysis for a one of our customers, the anlysis was made after the problem ocurred, so I was not able to make memory dumps or thread dumps. I was informed that the transaction impacted by the slowness wass the SalvarProposta.
Checking on the Response Times Hotspots and Remoting dashlets I was able to identify high syncronization methods (wait) and and also a WCF class that is thas a long total class, but very fast execution and inter tier times. Where can I chek this extra time was spent?
On the discussion with the developer he keeps telling me that the high .Net remoting times is not the problem, something else is causing this long times. I have digged a lot on this but the furtherest I was able to get was on this WCF methods. Is there something else I could do get more information?
This post is a further investigation based on this one:
Please see the attached session for more details on the problem.
Thanks a lot,
If you watch the elapsed time column and scroll down until you see a huge jump, you will find "Win32Exception: The wait operation timed out". I think this may be the root cause and the "WaitOneNative" is a common red herring in WCF where time goes to die (meaning there is nothing else going on).