On Wed, 18 May 2005 [EMAIL PROTECTED] wrote:
Tue May 17 15:35:10 2005 -> Reading databases from /usr/local/share/clamav Tue May 17 15:35:10 2005 -> Database correctly reloaded (34417 viruses) LibClamAV Warning: Not reloading database until idle - waiting for 2 children LibClamAV Warning: Waiting for 1 children until databae reload LibClamAV Warning: Not accepting inputs at the moment LibClamAV Warning: Waiting for 0 children until databae reload LibClamAV Warning: Accepting inputs again Tue May 17 17:13:05 2005 -> SelfCheck: Database status OK. Tue May 17 17:13:06 2005 -> /tmp/clamav-135326ee7c681aaa/msg.yzNKKB: Worm.SomeFool.P FOUND
Wait, so the problem fixed itself after 1.75 hours? That's *very* interesting. (I don't think anyone has waited it out before, so it wasn't known that it could fix itself.)
Could you tell us how you're running clamav-milter? Specifically, I'd like to know if you're using --external and your --max-children setting. Also, do you often hit the --max-children setting? I suspect that the bug occurs when reloading the database when max_children is hit, though that hasn't been proven yet. Perhaps increasing its value will help?
If you're using --external, also post the uncommented lines of your clamd.conf, since it's possible the problem could be with clamd itself.
This behavior is new as of 0.85.1
Actually it's new as of 0.84. Check the archives. Unfortunately nobody has been able to spot the bug in the code. (I've personally spent several hours looking for it.)
Until this bug gets squashed, you might want to run my clmilter_watch program [0], which can verify that clamav-milter is accepting inputs and restart it if necessary. You might also consider changing how sendmail treats milter failures (you could have it pass the mail through unscanned, for example), though that depends on local conditions (number of windows machines, gullibility of users, etc).
Certainly *do* post if it happens again, and if you get any more information. There are at least 5 people on this list who are very concerned about this issue.
[0] http://www.itg.uiuc.edu/itg_software/clmilter_watch/
Damian Menscher -- -=#| Physics Grad Student & SysAdmin @ U Illinois Urbana-Champaign |#=- -=#| 488 LLP, 1110 W. Green St, Urbana, IL 61801 Ofc:(217)333-0038 |#=- -=#| 4602 Beckman, VMIL/MS, Imaging Technology Group:(217)244-3074 |#=- -=#| <[EMAIL PROTECTED]> www.uiuc.edu/~menscher/ Fax:(217)333-9819 |#=- -=#| The above opinions are not necessarily those of my employers. |#=- _______________________________________________ http://lurker.clamav.net/list/clamav-users.html
