[Mimedefang] using sa-learn in conjunction with mimedefang
Alexander Dalloz
alexander.dalloz at uni-bielefeld.de
Tue Dec 23 14:05:15 EST 2003
Am Di, den 23.12.2003 schrieb Douglas, Jason um 18:47:
> Hello Everyone,
>
> I am wondering if I am teaching my spamassassing incorrectly. What I do
> is keep the spam messages in a certain spam folder. Then I run sa-learn
> as root like so:
>
> sa-learn --spam -C /etc/mail/spamassassin --showdots --dir SPAM/
>
> In /etc/mail/spamassassin/ there is a file called local.cf which
> contains:
>
> auto_whitelist_path /var/spool/spamassassin/auto-whitelist
> auto_whitelist_file_mode 0666
> bayes_path /etc/mail/spamassassin/bayes
>
> Sa-learn created the bayes files (bayes_msgcount, bayes_seen,
> bayes_toks) in /etc/mail/spamassassin/ the first time I ran it.
>
> My MIMEDefang runs as the user defang. Is there something else I have to
> do in order to get the sa-learn to work correctly with MIMEDefang? The
> types of spam I teach it still seem to get through (although they had
> very low scores to begin with, like 2.7). Am I wasting my time trying to
> teach it these very low scoring messages?
>
> Thanks!
>
>
> Jason Douglas
> Network Support Technician
See the default values:
# Mail which scores outside this range will be fed back into
SpamAssassin's
# learning system automatically, to train the Bayesian scanner.
bayes_auto_learn_threshold_nonspam 0.1
bayes_auto_learn_threshold_spam 12.0
in /usr/share/spamassassin/10_misc.cf
You can customize these values in your sa-mimedefang.cf, but a SPAM
value of 2.7 is not likely to be auto-learned. It should be sure that
the mail is really SPAM.
I also noticed that you run sa-learn as root. So be sure that all bayes
files have chown defang which is your MIMEDefang user.
Alexander
--
Alexander Dalloz | Enger, Germany
PGP key valid: made 13.07.1999
PGP fingerprint: 2307 88FD 2D41 038E 7416 14CD E197 6E88 ED69 5653
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Dies ist ein digital signierter Nachrichtenteil
URL: <https://lists.mimedefang.org/pipermail/mimedefang_lists.mimedefang.org/attachments/20031223/00d17e22/attachment.sig>
More information about the MIMEDefang
mailing list