[Mono-bugs] [Bug 415608] Intermittent hang when calling WaitHandle. WaitOne with a timeout specified

bugzilla_noreply at novell.com bugzilla_noreply at novell.com
Fri Aug 8 20:07:11 EDT 2008


User sharris at nvidia.com added comment

--- Comment #6 from Sean Harris <sharris at nvidia.com>  2008-08-08 18:07:10 MDT ---
(In reply to comment #5 from Zoltan Varga)
> I can repro this consistently with 1.9, but not with 2.0 or snv head. So its
> probably fixed by the above mentioned changes.

Unfortunately, I was just able to reproduce this using Mono 2.0 on Windows
Vista x64.  What OS did you use for your repro test?  Is it possible it's fixed
in some OS's but not others?

By the way, as additional data, the hang processes display evidence of waiting
on some cross-process semaphore.  Our admins found that if a number of
processes are hung, killing the oldest process manually would allow the others
to complete.  There is no code in this application that creates any
cross-process synchronization primivites (at least not on purpose).  Is it
possible the web service client code does this?

Finally, Miguel mentioned there's some way to send the process a signal that's
return a stack trace to see where it's hung.  I couldn't figure out how to do
that, at least not on windows.  The MinGW version of GDB wouldn't do much on my
system.  I also didn't get whether Miguel said I should try to send a "Quick
Signal" or a "Quit Signal" and how that translates into one of the documented
signal tokens.

Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.

More information about the mono-bugs mailing list