Why SA don't use bayes for some e-mails?

[spamis <sp...@pobladores.com>]

(sorry my poor english)

Hi,

I have a machine with qmail+qmail-scanner+spamassassin 3.0.0. Some time ago,
I have saw that some spam are passing through filter with 0 score. Then, I
activated sa_debug='1' and sa_hdr_report='1' in qmail-scanner for see how SA
are giving the score to e-mails and I saw that many e-mails don't have any
XX_BAYES score. I think that all e-mails should be this score, with more or
less score, but all should be marked with a XX_BAYES.

Why happen this? I want use bayes for all messages.

Any fix?

Thank you very much
-- 
View this message in context: http://www.nabble.com/Why-SA-don%27t-use-bayes-for-some-e-mails--tp15585584p15585584.html
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.

RE: Why SA don't use bayes for some e-mails?

[spamis <sp...@pobladores.com>]

Robert - elists-2 wrote:
> 
> Good question. Tough one without debugging your machine personally.
> 
> Did you do any web searching for this?
> 
> Have you considered upgrading to current SA 3.2.4 ?
> 
>  - rh
> 
> 
> 

Yes, I have been finding any info relative to this, but I can't found
nothing. It's very weird. All messages are checked, but some of them doesn't
have any mark of type XX_BAYES. I think that in the best case, a message
should be mark, at least, with a 00_BAYES.

I'm considering upgrading all my server, since it have and older version of
SA and other software... but it's a heavy job. I need some time for plan the
migration.

I have asked about this because I thought that could be a problem of bad
configuration.
-- 
View this message in context: http://www.nabble.com/Why-SA-don%27t-use-bayes-for-some-e-mails--tp15585584p15607224.html
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.

RE: Why SA don't use bayes for some e-mails?

[Robert - elists <li...@abbacomm.net>]

> 
> 
> (sorry my poor english)
> 
> Hi,
> 
> I have a machine with qmail+qmail-scanner+spamassassin 3.0.0. Some time
> ago,
> I have saw that some spam are passing through filter with 0 score. Then, I
> activated sa_debug='1' and sa_hdr_report='1' in qmail-scanner for see how
> SA
> are giving the score to e-mails and I saw that many e-mails don't have any
> XX_BAYES score. I think that all e-mails should be this score, with more
> or
> less score, but all should be marked with a XX_BAYES.
> 
> Why happen this? I want use bayes for all messages.
> 
> Any fix?
> 
> Thank you very much
> --

Spamis

Good question. Tough one without debugging your machine personally.

Did you do any web searching for this?

Have you considered upgrading to current SA 3.2.4 ?

 - rh