Bump your BAYES_99 score. That's the beauty of spamassassin, it is highly customizable. The determinant of whether it is spam or not is the total score versus your threshold. You can change either. I have my BAYES_99 set to 5.0 points. My threshold is 4.0 points. If there is enough about that e-mail to be a 99% chance it's spam based on what I've fed bayes, that's good enough for me. If you have your threshold set to 10, you can't expect bayes to know this and adjust accordingly.
>>> On 7/26/2007 at 7:17 AM, martin f krafft <[EMAIL PROTECTED]> wrote: Hi list, I just had a flood of spam coming through, which SA classified as ham. On closer inspection, it turns out that the only tests triggered for all those mails were HTML_MESSAGE and BAYES_99. HTML messages are commonplace today (unfortunately), so they don't add anything to the score. BAYES_99 yields 3.5 points. What's curious is that in this scenario, even though SA thinks that the message is 99%-100% likely to be spam, it will always classify it as ham, and further learning does not have any noticeable effect. I know how SA scores are computed. I do wonder how that algorithm applies to the BAYES_* tests though. Don't you think BAYES_99 should yield > 5 points to trigger the threshold on default installs? Shouldn't thus BAYES_* be renormalised? -- martin; (greetings from the heart of the sun.) \____ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] spamtraps: [EMAIL PROTECTED] "... alle sätze der logik sagen aber dasselbe. nämlich nichts." -- wittgenstein
