Ryan Tracey wrote:
> Exim version 4.32 #1 built 26-Apr-2004 09:57:22
4.34 is current, and IIRC there is also an debian package for it.
There were 2 crashes fixed in 4.33 and 4.34, maybe they are related to
your problem:
4.32/19. If the log selector return_path_on_delivery was set, Exim
crashed when bouncing a message because it had too many Received: header
lines.
(this one also applies when you set +all in your log_selector)
4.33/2. If verify=header_syntax was set, and a header line with an
unqualified address (no domain) and a large number of spaces between the
end of the name and the colon was received, the reception process
suffered a buffer overflow, and (when I tested it) crashed. This was
caused by some obsolete code that should have been removed. The fix is
to remove it!
> However, every week or so Exim will inexplicably die. I understand
> that there was a problem in earlier versions of Exim4 where roughly one
> in a million emails could potentially cause it to crash (unless I dreamt
> that -- and with my current jetlag I wouldn't be surprised if I did.)
> As I understand it a bugfix for this should have worked its way into
> Debian Sarge by now. If this is a case, then perhaps I have done
> something wrong in configuring Exim.
Exim should not crash on bad configuration.
> Is there anything I could check right now to help troubleshoot this.
> I'll sanitise my config in anticipation of being asked show it.
Does the exim DAEMON die? That should never happen....
Do you have logentries when the crash happens?
Is there any data lying around in your spool directory?
Do you have some kind of "special" system configuration?
