Running SA 1.5 / SPAMD -u nobody -d on a machine that accepts 88k mails per day.
Today was the big test, and there was a huge failure. The issue is that I get so many procmail processes opening that the CPU average on the machine pegs and the load average then causes sendmail to die. I get on an average 350 open procmail processes at any one time. 'ps ax |grep proc | wc -l" I need to figure out how to scale this. The machine is a PIII with 512 megs in Raid 5 configuration. The load averages used to be 0.8 and now I am up to 3.59. What works more effectively? One of the error messages is "timeout waiting for input from local during Draining Input" "http://www.sendmail.org/~ca/email/smenhanced.html"; I will upgrade sendmail tonite to fix this. More CPU? More memory? Faster disks? I also would consider load balancing the configuration. Should I balance the incoming mail servers or should I set up a couple machines to run spamd? Any suggestions would be appreciated (other than switching from sendmail to something else) I can expend hardware to fix this problem I just need to know were the best place to invest would be for a scalable spam processing system. Net:DNS is not working and I am using lookups for razor / SA, could this contribute to my issue? _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
