Greetings,
I recently upgraded 4 FreeBSD 3.4-STABLE systems from Exim 2.12 to
3.13. On one of the machines, the one that sends and receives by far the
most mail, exim seems to crash regularly with the following message coming
up in /var/log/messages:
Mar 2 09:13:33 calico /kernel: pid 9603 (exim), uid 300: exited on signal 11
Every time it happens, I see no corresponding time entry in any of the
Exim log files indicating that a delivery attempt was made (I've been
running a continuous tail -f on /var/log/messages, and Exim's mainlog,
paniclog, and rejectlog files).
Also, it should be noticed that when I said "upgraded" above, it's more
like "replaced". I killed the old exim process, renamed its directory,
compiled and installed the new one, created a new configuration file, and
then copied my virtual domain files back over from the old directory. I
did, of course, re-generate the dbm files for the local and virtual
domains.
Nothing I could find in the mailing list archives or FAQ helped. The FAQ
suggested using Exim's internal DB function rather than dbm; that didn't
seem to make any difference. I also tried backing down one version, from
3.13 to 3.12... no change.
If anyone has any suggestions on what to check for, or if this has
happened before, I'd love to hear them. Thanks in advance!
*********************************************************************
* Bryce Newall * Email: data@??? * ICQ: 461599 *
* SysAdmin / Hostmaster - DreamHaven Internet Services, LLC *
* www.dreamhaven.net * www.dreamhaven.net/~data *
*********************************************************************
