[Mono-bugs] [Bug 472732] mod-mono crashes after timeout

[bugzilla_noreply at novell.com]

https://bugzilla.novell.com/show_bug.cgi?id=472732

User mhabersack at novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=472732#c3





--- Comment #3 from Marek Habersack <mhabersack at novell.com>  2009-02-05 13:02:32 MST ---
(In reply to comment #2)
> Thanks Marek, I check on the file descriptors:
> 
> ~# cat /proc/sys/fs/file-max
> 169203
This is the maximum files that can be opened system-wide, it's not the per
process limit. To check the latter look at /proc/PID/limits (or issue ulimit -a
as root if you use bash). Also check /etc/security/limits.conf (that's the
usual location of the file) for per-login or per-group limits. Remember that if
you log in as root and restart apache, then the root shell's limits will apply
to apache. Additionally, apache may limit resource usage even more.

> 
> and the machine has plenty of memory. Is there anything else I can do to help
> debug this? I rebooted the server and it works fine now. 
As above - there's plenty of memory system-wide, individual applications may
have limits (you can check for them in the same file as above). One of the out
of memory causes might be limited RSS (Resident Set Size) memory for
application. Depending on the kernel version and distribution it is possible
that other means are used to limit application resource usage.

> Also, do you think the socket timeout is related? It happened about the same
> time and was the last entry in our log.
Yes, it might be related. It is possible that one end succesfully creates the
socket, connects to it, the other end receives the connection but, for
instance, cannot open a file to read and crashes (because, e.g., it hasn't
checked open (2) return error code and uses the invalid fd)

> FWIW, I'm running Microsoft's MVC binaries on mono with good success.
That's fine, but we still can't support MVC at this point.

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