Hi:

I am using Spamassassin version 2.63 (spamd and spamc invoked by
procmail) installed from 10CE cd and it is only catching around 20% of
spams, which in my case means that is missing between 60-70 spams a day.
Since missing that large percent of spams is something that is only
happening to me, my conclusion is that I must be doing something awfully
wrong (or missing something extremely obvious).

I have read that bayesian filters start doing their job after been
trained with 2000 or more spams. I already have fed spamassassin with
more than 2300 messages using the command:
sa-learn --mbox --spam /home/adolfobello/evolution/local/sa-spam/mbox

Less than 2 weeks ago I also started training it with hams.
sa-learn --mbox --ham /home/adolfobello/evolution/local/sa-ham/mbox

Every time that I run those commands I also delete the spam and ham
mailboxes. (Is this the right thing to do?)

My /etc/mail/spamassassin/local.cf looks like:
######
rewrite_subject 0
report_header 0
use_terse_report 1
always_add_report 0
auto_whitelist_path        /var/spool/spamassassin/auto-whitelist
auto_whitelist_file_mode   0666
report_safe 0
use_bayes 1
#####

Every line in ~/.spamassassin/user_prefs is commented out.

Command to invoke spamc from procmailrc
:0fw
| /usr/bin/spamc -s 256000

Command to start spamd:
/usr/bin/spamd -d -c -a -m5 -H

Mailing lists, mail from trusted domains and some other virus catching
procmail filters are called before spamc.

What am I missing or doing wrong?

Any help will be truly appreciated. I am drowning in junk mails.

Adolfo



____________________________________________________
Want to buy your Pack or Services from MandrakeSoft? 
Go to http://www.mandrakestore.com
Join the Club : http://www.mandrakeclub.com
____________________________________________________

Reply via email to