I have not seen any headers so far with BAYES_40, BAYES_44, BAYES_50, or BAYES_56. Is this because they are all score 0.001?
In SA 2.54, they're scored at 0, not 0.001. Have you already changed the scores?
Assuming they're at 0, then yes, this is the reason why they've never shown up. In this range it's a 50/50 chance, as far as Bayes is concerned (with some margin for error), so there's no justification to change the score in either direction.
If so, what is the threshold to set the score beyond in order to make Bayes results show on every message?
I believe I've heard people refer to using a score 0.001 as the best way to get something to show up in the report without altering the score significantly, so if you've already changed the scores on those tests, I would guess one of three things is happening:
1. Your Bayes database distribution is very weighted toward the extremes, such that there really isn't anything that will hit the 40-59% range. Extremely unlikely, but probably possible.
2. There's a typo in your config file. I've occasionally forgotten to write "score" in front of a rule that I was re-scoring. spamassassin --lint should catch this.
3. Both of us misunderstand the threshold for reporting.
As for increasing the scores of Bayes hits above 50%, I'd recommend doing it for 60% and up instead, just to keep that 10% margin of error. Chances are pretty good that you have non-spam hitting in that unscored 40-59% range.
Kelson Vibber
SpeedGate Communications <www.speed.net>
------------------------------------------------------- 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
