Bayes score question - sort of a rehash

Kai Schaetzl maillists at CONACTIVE.COM
Wed Dec 28 19:32:07 GMT 2005

    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

Steve Campbell wrote on         Wed, 28 Dec 2005 11:49:42 -0500:

> At what point do the Bayes_00, etc start showing up again. I have enough 
> spam emails, but not enough ham emails, in my new Bayes DBs. Should I see 
> something before I reach 200 of both, or should I see something with just 
> the 200+ spam entries. I did not use FSL's starter DB files. My DB files are 
> growing, so I know they are being used.

There will only be Bayes scores once you have at least those mentioned 200 ham 
*and* spam. However, there is a setting which allows you to change that limit 
in sa. In the past, this option didn't allow to go under the 200, but from 
newer documentation it seems you can do this now, f.i. for testing (not 
recommended for production). Also running a spamassassin -D on a message will 
show you if Bayes seems to be at least usable, although not getting used yet. 
You should actively learn the first 200 (or more) of ham and spam unless you 
want to wait for a long time before it kicks in.


Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services:

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at with the words:
'leave mailscanner' in the body of the email.
Before posting, read the Wiki ( and
the archives (

Support MailScanner development - buy the book off the website!

More information about the MailScanner mailing list