[Mimedefang] Re: OT: New Attack/Poor SPAMming programming?

[Jan-Pieter Cornet]

On Fri, Jan 12, 2007 at 01:13:04PM -0600, Ben Kamen wrote:
> >Here is a great article about sendmail time-outs (and other antispam 
> >tricks such as greet pause):
> >
> >http://www.acme.com/mail_filtering/sendmail_config_frameset.html

The information is a bit outdated. Also I don't agree with the "DNSBLs
are bad" recommendation (we're blocking over 50% of the mail using
DNSBLs here).

> My original email asked if anyone else was seeing this... not so much
> as "what can I do about it". I've been using sendmail for about 11yrs now.

I just did some stats, I'm seeing about 80 disconnects per minute
due to command timeout... out of about 6000 simultaneous sendmail
connections outstanding.

I used to have the command timeout at 10 minutes, so that meant
800 wasted sendmail processes (about 13%) so I reduced it to 2
minutes (your peers have the option to use that timeout to
cache outgoing sendmail connections. It's polite to leave it
at the maximum recommended MCI_CACHE_TIMEOUT, which is 10 minutes.
(But in over 80% of the cases you'd be polite to spammers...)

-- 
Jan-Pieter Cornet <johnpc at xs4all.nl>
!! Disclamer: The addressee of this email is not the intended recipient. !!
!! This is only a test of the echelon and data retention systems. Please !!
!! archive this message indefinitely to allow verification of the logs.  !!