UPDATE - checked the other servers and getting the same segfaults there
too. But they aren't locking up. S - it is Exim.
Running on FC5 with 64 bit kernel. What can I do to help track this down?
Marc Perkel wrote: > I don't know if this is an Exim problem, hardware, or a kernel problem.
> But Exim is what's crashing, has locked up the server twice, and it's
> always at the same location. Riunning Exim 4.63. I tried cocompiling and
> reinstalled. Running on Fedora Core 5 with the latest upgrades, except
> for a custom kernel to fix a motherboard related problem.
>
> Just put a new server online trying out the new AMD AM2 processor. I
> compiled a custom kernel because the regular Fedora Core kernels aren't
> yet compatible with my Asus M2NPV-VM motherboard using the nVidia chipset.
>
> I compiled 2.6.18rc6 and got segfault errors. So I tried the 2.6.17.13
> kernel and same thing. About every 20 minutes or so I get one of these.
>
> Sep 11 12:05:18 pascal kernel: exim[19840]: segfault at 0000000000000000
> rip 0000003f53e73ee0 rsp 00007fff9e561d18 error 4
>
> At one point the server locked up. Before I put it online I did several
> days of memory testing with no errors. I believe the Exim code is solid
> as it has been running flawlessly on all my other servers.
>
> It's probably hardware or some incompatibility but I'm not sure where to
> start looking. Trying to understand what this error means. What is Error
> 4? How do I track this down?
>
> Thanks in advance.
>
>
