On 13/10/2006 14:25, Marc Perkel wrote: > Been getting a fair number of these. Maybe one every 20 minutes or so.
> Running 64 bit FC5. Anyone seen this? What is Error 4?
>
> Oct 13 00:24:57 pascal kernel: exim[19299]: segfault at 0000000000000000
> rip 0000003845f73590 rsp 00007fff428aad68 error 4
We've been here before, and the basic answer was "no, nobody else has
seen this with Exim". That may be because you're running your own kernel
on esoteric hardware, it could be that your server is overheating, or it
could be phases of the moon. Either way, nobody had seen it then.
You've also posted to LKML and got a few very interesting replies, one
of which pointed to threads and the NPTL invocations thereof. I don't
know if anything came of it, as I frankly can't be bothered following
the entire set of threads, but it would be useful if - when reporting a
problem - you include *all* the relevant information.
What kernel are you running?
Are you patched up-to-date?
What CPU, mobo etc?
Does this occur with many different kernels?
Have you compiled Exim yourself, or is it a Fedora version?
Either way, is it threaded?
Can you downgrade to a 32-bit kernel and see if that helps?
Can you downgrade to a non-SMP kernel and see if that helps?
Do you have HT turned on (or whatever the AMD analogue is)? Have you
tried with it turned off?
Do you only see this error with Exim, or do other applications manifest
it too?
So many things you can do to help yourself here, Marc. Then once you
have an idea of what's going wrong, others might be able to help.
