[Mimedefang] Training SA when mail is not stored locally?

[Joe Arnstein]

Michael,

Thanks for the reply. 
 
Is it OK to turn on bayes_auto_learn without first training SA manually?
My thinking is that the server might learn the wrong thing out of the
gate which would be bad since it is difficult to see what is being
tagged. Is that a misguided notion?   

If SA learns what is spam/ham based on the spam/nonspam threshold from
messages it's read on its own (no intervention from me), don't we
encounter a chicken and egg problem?

How do you keep the server from learning the wrong thing?  

Joe



-----Original Message-----
From: mimedefang-bounces at lists.roaringpenguin.com
[mailto:mimedefang-bounces at lists.roaringpenguin.com] On Behalf Of
Michael Faurot
Sent: Wednesday, February 04, 2004 2:22 PM
To: mimedefang at lists.roaringpenguin.com
Subject: Re: [Mimedefang] Training SA when mail is not stored locally?

In article <00c701c3eb50$1762d550$ceaa2799 at mcilink.com> you wrote:

> I'm running an MD/SA gateway for a customer where mail is scanned,
> tagged, and forwarded directly to their servers (nothing is stored
> locally), but I need to train SpamAssassin and beef up its bayes db.
> How do people typically gather ham and spam to train the box under
these
> conditions?  Is it possible to do it without too much intervention on
> the customer's part? 

Yes, just use the bayes_auto_learn option in SA.  If there's a good
amount
of traffic going through the box, it should build up a corpus fairly
quickly.  You may also want to tweak bayes_auto_learn_threshold_nonspam
and bayes_auto_learn_threshold_spam if you don't like the defaults.
I wound up leaving bayes_auto_learn_threshold_nonspam at its default
but adjusted bayes_auto_learn_threshold_spam to 8.0.
_______________________________________________
Visit http://www.mimedefang.org and http://www.canit.ca
MIMEDefang mailing list
MIMEDefang at lists.roaringpenguin.com
http://lists.roaringpenguin.com/mailman/listinfo/mimedefang