No bayesian for me?

[Mauricio Tavares]

Martin Hepworth wrote:
> you prob haven't for a valid bayes DB yet (one wiith 200 spam and 200 ham).
> 
	I am not sure. Last week I fed sa-learn some 3K pieces of nice spam and 
ham. It seemed then to be working; here is an excerpt for a properly 
identified spam from the 9th:

Date: Sun, 9 Aug 2009 20:38:33 +0300
X-MailScanner: Found to be clean
X-MailScanner-SpamCheck: spam, SpamAssassin (not cached, score=9.892,
         required 4.7, BAYES_95 3.00, HTML_IMAGE_ONLY_28 1.56,
         HTML_MESSAGE 0.00, MIME_HTML_ONLY 1.46, RAZOR2_CF_RANGE_51_100 
0.50,
         RAZOR2_CF_RANGE_E8_51_100 1.50, RAZOR2_CHECK 0.50, RCVD_IN_PBL 
0.91,
         RDNS_NONE 0.10, URI_HEX 0.37)
X-MailScanner-SpamScore: *********
X-MailScanner-Envelope-From: gnawenoc at tulimpimpim.com.br
X-Spam-Status: Yes

As you can see, the bayesian filter is there in all of its glory. But 
yesterday and today it has gone missing.

> also the autowhitelist isn't turned off using the "use_auto_whitelist 0" 
> line anymore - remove the pluging from the V310.pre file
> 
	Thanks!
> -- 
> Martin Hepworth
> Oxford, UK
> 
> 2009/8/11 Mauricio Tavares <raubvogel at gmail.com 
> <mailto:raubvogel at gmail.com>>
> 
>            Ok, I thought I knew how to setup bayesian under mailscanner.
>      I went to spam.assassin.prefs.conf and added
> 
>     # Bayesian filter
>     use_bayes 1
>     use_bayes_rules 1
>     bayes_path /var/spool/MailScanner/bayes/bayes
>     bayes_auto_learn 1
>     bayes_file_mode 0640
>     bayes_auto_learn_threshold_nonspam 0.1
>     bayes_auto_learn_threshold_spam 12.0
>     bayes_ignore_header X-monetra-com-MailScanner
>     bayes_ignore_header X-monetra-com-MailScanner-SpamCheck
>     bayes_ignore_header X-monetra-com-MailScanner-SpamScore
>     bayes_ignore_header X-monetra-com-MailScanner-Information
> 
>     lock_method flock
>     use_auto_whitelist 0
> 
>     But I am getting emails whose header seem to indicate the bayesian
>     filter is not present:
> 
>     X-MailScanner: Found to be clean
>     X-MailScanner-SpamCheck: not spam, SpamAssassin (not cached,
>     score=4.464,
>            required 4.7, HTML_MESSAGE 0.90, INVALID_DATE 0.50,
>            MIME_HTML_MOSTLY 0.00, MIME_QP_LONG_LINE 1.82,
>     MISSING_MIMEOLE 0.00,
>            MPART_ALT_DIFF 1.14, RDNS_DYNAMIC 0.10)
>     X-MailScanner-SpamScore: ****
>     X-MailScanner-Envelope-From: promolx1 at tutopia.com
>     <mailto:promolx1 at tutopia.com>
>     Subject: -Soluciones para tiempo y asistencia-
>     X-Spam-Status: No
> 
>     Anything I could be missing? I believe there is a way to ask
>     mailscanner if it has bayesian but I can't remember it.
>     -- 
>     MailScanner mailing list
>     mailscanner at lists.mailscanner.info
>     <mailto:mailscanner at lists.mailscanner.info>
>     http://lists.mailscanner.info/mailman/listinfo/mailscanner
> 
>     Before posting, read http://wiki.mailscanner.info/posting
> 
>     Support MailScanner development - buy the book off the website!
> 
> 
> 
>