On Mon 2005-02-14 (09:07), Johann Spies wrote: > On Fri, Feb 11, 2005 at 10:37:06AM -0500, Chris Santerre wrote: > > [...] > > > > Can we get the output from the "spamassassin -D --lint" please? > > I should have read what you meant! Ok. Here it is ( have since added > a few more rules ) : > > debug: SpamAssassin version 3.0.2 > [...] > debug: bayes: found bayes db version 3 > debug: using "/home/spamd/.spamassassin" for user state dir > debug: Score set 3 chosen.
I just upgraded to 3.0.2 from 2.64, and I'm also wondering why 3.0.2 has overall lower scores than 2.64. After looking closer, I realized that SA 3.0 zeros "score set 3" in 51 cases - just look at: score DRUGS_MANYKINDS 0.031 2.734 0 0.000 score EXCUSE_19 0.056 0.357 0.021 0.000 score FAKE_HELO_MAIL_COM 1.303 1.972 0.111 0.000 score FORGED_MUA_EUDORA 1.510 0.054 0.496 0.000 score FORGED_RCVD_HELO 0 0.050 0.266 0.000 score FREE_PORN 0.794 0.023 1.937 0.000 score FROM_ENDS_IN_NUMS 0.177 0.516 0.517 0.000 ... Version 2.64 in comparison always sets score set 3 to a non-zero value. Score set 3 - the fourth score - applies when Bayes is enabled and network tests are enabled. In addition, non-zero score set 3 values in 3.0.2 are much lower than those in 2.64, e.g. 2.64: score SORTED_RECIPS 4.299 4.300 2.696 2.699 3.0.2: score SORTED_RECIPS 0.879 1.155 1.759 0.887 2.64: score DEAR_FRIEND 1.888 1.065 2.397 1.846 3.0.2: score DEAR_FRIEND 0.542 0.766 1.288 0.070 This is probably intented - but I just don't understand why so many syntactical checks are completely disabled or strongly damped when using Bayes and network tests together. -Juergen Georgi
