I've just upgraded from SpamAssassin version 2.43 to 2.55. I've been using sa-learn to build up the Bayesian database. Still, SpamAssassin doesn't seem to be using the Bayes test. All of the spams it catches fail due to other rules, and it is letting a good number of other spam get through. Is there a switch I need to turn on to get the Bayes test running?
How many emails have you trained? SA requires 200 nonspam AND 200 spam messages for learning input before it will even consider using bayes. In some cases it may take more training, as it also requires a minimum number of "good" tokens to have been added as a result of the training..
There is a use_bayes config option, but it defaults to 1 (on).
You can see SA's decision making process if you look at the debug output:
spamassassin -tD <sample-spam.txt
<bunch of output> debug: debug: Only 1 spam(s) in Bayes DB < 200 <bunch more output>
------------------------------------------------------- This SF.net email is sponsored by: Etnus, makers of TotalView, The best thread debugger on the planet. Designed with thread debugging features you've never dreamed of, try TotalView 6 free at www.etnus.com. _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
