bayes_auto_learn not working???

[Odhiambo Washington <wa...@wananchi.com>]

Hello Assassins,

Someone can give me clues as to why autolearn does not work?
I am using  bayes system-wide with the following configurations:

# Bayes Setting (man sa-learn)
use_bayes                       1
bayes_path                      /etc/mail/spamassassin/bayes
bayes_file_mode                 0666
bayes_use_hapaxes               1
bayes_expiry_max_db_size        2000000
bayes_auto_learn                1

bayes_auto_learn_threshold_nonspam      -10.0
bayes_auto_learn_threshold_spam         4.9


So why did it NOT autolearn as ham, the message whose scores are below?

X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on ns2.wananchi.com
X-Spam-Level:
X-Spam-Status: No, hits=-100.7 required=4.0 tests=BAYES_01,NO_REAL_NAME,
        TO_MALFORMED,USER_IN_WHITELIST autolearn=no version=2.63



Or is it because of the following note in Mail::SpamAssassin::Conf?

<cut>
 Note: SpamAssassin requires at least 3 points from the header, and
       3 points from the body to auto‐learn as spam.  Therefore,
       the minimum working value for this option is 6.
</cut>


Thanks in advance for any insights.

-Wash

http://www.netmeister.org/news/learn2quote.html

--
+======================================================================+
    |\      _,,,---,,_     | Odhiambo Washington    <wa...@wananchi.com>
Zzz /,`.-'`'    -.  ;-;;,_ | Wananchi Online Ltd.   www.wananchi.com
   |,4-  ) )-,_. ,\ (  `'-'| Tel: +254 20 313985-9  +254 20 313922
  '---''(_/--'  `-'\_)     | GSM: +254 722 743223   +254 733 744121
+======================================================================+
On a paper submitted by a physicist colleague:

"This isn't right.  This isn't even wrong."
		-- Wolfgang Pauli

Re: bayes_auto_learn not working???

[Theo Van Dinter <fe...@kluge.net>]

On Tue, Mar 23, 2004 at 01:37:39AM -0500, Theo Van Dinter wrote:
> No, it probably has more to do with this snippet of docs right above
> that one:

Actually, I was close ...

           Note that certain tests are ignored when determining whether a mes-
           sage should be trained upon:
            - auto-whitelist (AWL)
            - rules with tflags set to 'learn' (the Bayesian rules)
            - rules with tflags set to 'userconf' (user white/black-listing
           rules, etc)


your message got -100.7, -100 of which was a userconf WHITELIST rule.



-- 
Randomly Generated Tagline:
"Was Jimi Hendrix's modem, a Purple Hayes?"

Re: bayes_auto_learn not working???

[Theo Van Dinter <fe...@kluge.net>]

On Tue, Mar 23, 2004 at 09:25:15AM +0300, Odhiambo Washington wrote:
> Or is it because of the following note in Mail::SpamAssassin::Conf?
> 
> <cut>
>  Note: SpamAssassin requires at least 3 points from the header, and
>        3 points from the body to auto???learn as spam.  Therefore,
>        the minimum working value for this option is 6.
> </cut>

No, it probably has more to do with this snippet of docs right above
that one:

     Also note that auto-training occurs using scores from either score-
     set 0 or 1, depending on what scoreset is used during message
     check.  It is likely that the message check and auto-train scores
     will be different.

-- 
Randomly Generated Tagline:
: I could understand principles of Perl source in 2-3 days [. . .]
 
 Gee, it took me about eleven years.  :-)
              -- Larry Wall in <19...@wall.org>