bayes problem

[Billy A. Pumphrey]

> -----Original Message-----
> From: mailscanner-bounces at lists.mailscanner.info [mailto:mailscanner-
> bounces at lists.mailscanner.info] On Behalf Of Philippe Delodder
> Sent: Friday, September 29, 2006 10:02 AM
> To: MailScanner discussion
> Subject: bayes problem
> 
> Hi,
> 
> when i run spamassassin -D --lint i see that bayes is used but when i
> check the header of an email that is spam i don't see use of bayes in
> MailScanner-SpamCheck. is that normal?
> 
> I'm using MailScanner version 4.54.6 with postfix
> 
> Philippe Delodder
> 

I would start by making sure you do the correct lint.  Make sure you
specify the config file.  It was mentioned that you shouldn't have to
anymore, either I misunderstood it or it was wrong.  Anyway, do this:
spamassassin -D --lint -p /etc/MailScanner/spam.assassin.conf.prefs

Here is the different in the lint:
(without the -p argument)
[1712] dbg: config: using "/root/.spamassassin/user_prefs" for user
prefs file

(with the -p argument)
[1712] dbg: config: using "/root/.spamassassin/user_prefs" for user
prefs file

I am not sure about the normal bayes in header question.  I checked my
headers of random emails and bayes was in all of them.  Seems like I
remember something normal about it sometimes not being in the header,
but wait for other input.

Also here is a good command to run.
sa-learn --dump magic

Example of mine:
[root at WoodenMS2 ~]# sa-learn --dump magic
0.000          0          3          0  non-token data: bayes db version
0.000          0     164042          0  non-token data: nspam
0.000          0     328180          0  non-token data: nham
0.000          0     174468          0  non-token data: ntokens
0.000          0 1159604893          0  non-token data: oldest atime
0.000          0 1159797182          0  non-token data: newest atime
0.000          0 1159796609          0  non-token data: last journal
sync atime
0.000          0 1159777716          0  non-token data: last expiry
atime
0.000          0     172800          0  non-token data: last expire
atime delta
0.000          0      31252          0  non-token data: last expire
reduction count

Shows you that things are going on and bayes is populating.  I am also
assuming that you are running this all in root and using config defaults
in most places, like the bayes configuration in
/etc/MailScanner/spam.assassin.conf.prefs