by 'crashes', that includes the clamd process just silently aborting, so mail starts to pile up on my MX gateways, or other times clamd is "running", using lots of cpu time, but non-responsive - messages again start piling up on the MX gateways, unable to be accepted by the spam/virus proxy that clamd is running on.
I've seen this happen on FreeBSD 4.10 on a only a couple of our servers, it doesn't happen very often though. We scan a few hundred thousand messages a day and it seems to chug along ok.
By "silently" aborting is the kernel not reporting an exit reason? (check dmesg)
When ours would silently quit it would be due to a signal 6, bus error. I've not seen that happen for a few weeks now. The funny thing is, even with sysctl kern.sugid_coredump=1 set, clamd was not producing a core file anywhere.
-- Robert Blayzor, BOFH INOC, LLC rblayzor\@(inoc.net|gmail.com) PGP: http://www.inoc.net/~dev/ Key fingerprint = 1E02 DABE F989 BC03 3DF5 0E93 8D02 9D0B CB1A A7B0
Hardware: The parts of a computer system that can be kicked. _______________________________________________ http://lists.clamav.net/cgi-bin/mailman/listinfo/clamav-users
