Improving base SA

In the hopes this might spark a thread of "here's stuff I did to get
SA-exim to work more effectively" here is a list of some recent changes
I've been experimenting with:

**

FWIW, I've found the stock SA 2.63 misses sufficiently much spam now
that the Bayes classifier is autolearning spam, which then feeds back
and lets even more through. To help against that I used,

  bayes_auto_learn_threshold_nonspam -0.5

..rather than the default of 0.1.

**

The extra rules at the SA Rules Emporium have helped a lot too:
http://www.rulesemporium.com/

The ones on http://www.rulesemporium.com/rules.htm I think are pretty
much implementable right away. There are some others that need a little
care before firing up (e.g. the antidrug.cf stuff if you're a meds
provider).

**

The SpamCopURI patch to query SURBL is well worth adding as the
URL/domain is sometimes about the only thing that identifies a
mail as spam. http://sourceforge.net/projects/spamcopuri

**

The SA-exim greylisting patch is really great - benefits:
 * rejecting spam
 * reducing resource usage:
   * it's reducing the amount of spam that gets quarantined here so
     /var/lib/sa-exim requires less space and end-user hassle
   * false positives are less drowned out, reducing mistakes

IMO, if you've been putting off implementing this on your system,
definitely consider pushing it up the sysadmin TODO list.

http://marc.merlins.org/linux/exim/files/sa-exim-cvs/README.greylisting

HTH,
Paul

-- 
Paul Makepeace ................................  http://paulm.com/ecademy

"If we park in driveways and drive on parkways, then that must make me a
 goddess."
   -- http://paulm.com/toys/surrealism/