SA Bayes Questoin

Nathan Johanson nathan at TCPNETWORKS.NET
Fri Apr 25 21:22:22 IST 2003

Thanks so much for the quick response.

-----Original Message-----
From: Julian Field [mailto:mailscanner at ECS.SOTON.AC.UK] 
Sent: Thursday, April 24, 2003 1:21 PM
Subject: Re: SA Bayes Questoin

At 21:03 24/04/2003, you wrote:
>I set up a test box with SpamAssassin 2.53 and the latest MailScanner
>build. (I wanted to make sure I get all of the potential wrinkles
>out before going live).
>I fired some spam messages through the test system. They do get scored
>as spam by Spam Assassin, but the bayes databases have not been created
>yet (bayes_msgcount, bayes_seen, and bayes_toks).

You won't see the databases created until the first message is
auto-learned. The auto-learning thresholds are (I think) -2 and +25. A
message has to score outside that range in order to be auto-learned.
it a message that is extremely spammy (scoring > 25) and all the files
should spring into existence. It can take a while before your mail
gets one that spammy.

>  At this point, I'm
>just sticking with the default location of /root/.spamassassin/.

That's fine. Keep an occasional eye on the free space in your root

>  Am I
>falsely assuming that these databases should be created immediately
>the auto-learning functionality, after only processing a few spam
>messages? Or does it take a much larger mass of email before these
>databases are created and subsequently populated with information? I
>assumed that the first processed spam message would be enough to at
>least trigger the creation of the db files.

The other fact to know is that the bayes databases won't actually be
to help score a message until 200 messages have been auto-learned.

Basically, set it up, leave it chewing mail for a few days and come back
it, when you should find all the files have been created and it is
scoring messages.
Julian Field
Professional Support Services at
MailScanner thanks transtec Computers for their support

More information about the MailScanner mailing list