Andreas Metzler wrote:
> On 2006-04-26 Michel Meyers <steltek@???> wrote:
>> Philip Hazel wrote:
>>> On Tue, 25 Apr 2006, Michel Meyers wrote:
>
>>>> If there's anything I could run to debug it, please let me know (as I
>>>> can reproduce the problem pretty easily here).
>
>>> Perhaps compiling Exim with -g and taking a core dump of a stuck process
>>> would help. If it is in a loop, that should indicate where the loop is.
>>> You may have to do special things in order to get a dump of a setuid
>>> process such as Exim - seek advice from Debian experts.
>
>> OK, I'm not knowledgeable with debugging stuff at all so I will
>> definitely need help there.
> [...]
>
> Hello,
> I can provide you with packages with debuuging symbols (built with -g
> and unstrippped).
>
> http://downhill.aus.cc/debian/misc/exim4-daemon-heavy_4.61-1.0.0.debug_i386.deb
> (md5: 1adecd15e58afe4f06d35b41bbda8871)
>
> I thought this should work but it does not:
> -----
> Afaik
> kill -ABRT process-id
> should produce a core file in /var/spool/exim4. You'll probably need
> to add
> ulimit -c 50000
> to the top of the init-script as core file generation seems to be
> disabled by default.
> -----
I tried that but it didn't work. Even adding core 50000 to
/etc/security/limits.conf didn't seem to help.
Then I tried gcore:
gcore -o /var/core 30611
That dumped a file which I gzipped. I don't know whether my previous
mail went through (might've been rejected with the big file), so just to
be sure I uploaded it to Rapidshare (I don't have any fast webspace
available myself). You can get it here:
http://rapidshare.de/files/19037900/core.30611.gz.html
Greetings,
Michel
