In spamassassin 3 you could debug bayes points running command line spamassassin with "-D bayes" parameter. I think you can in version 4 too.
In the log all the tokens extracted from the message are listed with the points assigned, so you can see exactly how the score is calculated. I had a similar experience in the past: the messages are exactly the same content, word by word but the sender hosts were different. So, some of the extracted tokens used by the score system were different, and they contributed to the final score. On Tue, Nov 5, 2024 at 3:55 PM Grega via users < [email protected]> wrote: > > Hi all. > > > I`m posting this here as well. > > Now I sorted out this score and disabled rules thing, but I still have > bayes issues even after I re-trained the whole thing by hand AND disabled > autolearn. > > Here is post: https://forum.efa-project.org/viewtopic.php?p=20226#p20226 > > > I just don`t get 2 things: > > > > 1. Why same mail with different recipient is treated different by SA > bayes? (one has bayes added other not) > 2. Why my bayes scoring is missing extreme values like BAYES_05 and > BAYES_99 ? > > > Thanks for listening and any input 😊 > > Regards,G > > ------------------------------ > *From:* Grega via users <[email protected]> > *Sent:* Wednesday, 25 September 2024 20:02 > *To:* [email protected] > *Subject:* Re: Bayes in V4 compared to V3 > > > Oh god I`m idiot... > > > I had: > > score BAYES_20 0.0 > > > So now every mail has bayes score in it (changed it to score BAYES_20 0.1) > > > Still puzzling why I have no extreme low or extreme high values. > > Also still puzzling why out of 3 identical mails one had bayes_60 and > other 2 bayes_20. > > > Autolearn is off. > > > > > ------------------------------ > *From:* Matija Nalis <[email protected]> > *Sent:* Wednesday, 25 September 2024 18:23 > *To:* [email protected] > *Subject:* Re: Bayes in V4 compared to V3 > > On Mon, Sep 23, 2024 at 01:14:25PM +0000, Grega via users wrote: > > Why one has "BAYES_60" and other 2 not? > > > > 4. Race condition (IDK I`m not coder) > > What backend are you using for storing bayer data? > > I'm not yet on 4.x (Debian Stable FTW), but in SA 3.x default was a > local file storage (BDB?) which used file locking, and that locking was > prone to timing out when several mails came in quick succession. > > For me, switching to MySQL backend for Bayes (and AWL) fixed such issues... > > -- > Opinions above are GNU-copylefted. >
