At 11:17 PM 9/2/03, you wrote: >On Wed, 2003-09-03 at 01:17, Brian W. Antoine wrote: >> I'm on a Mandrake 9.1 system using clamav-milter with sendmail and about >> once a day I find the following: ><snip> > >I had the exact same problems myself, with RH 8.0 and Sendmail, using >clamav-milter. I asked a few times about what a solution would be to >this, and never got much of a response. It just seems that >clamav-milter is not ready for a production server and still needs some >work, I guess.
Except it's clamd dieing, not the milter, unless I'm reading this ps output wrong. 24655 ? S 0:00 /usr/local/sbin/clamd LESSKEY=/etc/.less LC_MONETARY= 24659 ? S 0:00 \_ /usr/local/sbin/clamd LESSKEY=/etc/.less LC_MONET 24667 ? S 0:00 \_ /usr/local/sbin/clamd LESSKEY=/etc/.less LC_M 24669 pts/5 S 0:00 /usr/local/sbin/clamav-milter -blo local:/var/run/clm 24670 pts/5 S 0:00 \_ /usr/local/sbin/clamav-milter -blo local:/var/run 24671 pts/5 S 0:00 \_ /usr/local/sbin/clamav-milter -blo local:/var 24684 ? S 0:00 sendmail: accepting connections ons It's usually that second clamd process that falls over. >My solution was to use Amavis-New/amavis-milter. Since getting it >installed and running, I have had no problems whatsoever, and it's been >running now for about 5 days. The nice thing about amavis as well, as >there is built in spam assassin scanning, and it's all controlled by the >amavisd.conf file, which you can tweak a bit. Yeah, I'd noticed that sourceforge itself used amavis instead of clamav-milter, that should have been a red flag. I assume that amavis calls clamscan instead of clamdscan, so even if something pukes, it only effects a single email instead of taking out your mail server when clamd dies. ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Clamav-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/clamav-users
