this is my bayes config use_bayes 1 use_bayes_rules 1
# Bayesian classifier auto-learning (default: 1) # bayes_auto_learn 1 bayes_path /var/spamassassin/bayes bayes_file_mode 0777 shortcircuit BAYES_99 spam shortcircuit BAYES_00 ham score BAYES_00 -3.5 score BAYES_05 -2.0 score BAYES_20 -1.0 score BAYES_40 -0.5 score BAYES_50 1.8 score BAYES_60 3.5 score BAYES_80 5.0 score BAYES_95 6.5 score BAYES_99 7.5 score BAYES_999 0.4 On Tue, Jul 21, 2015 at 10:59 PM Roman Gelfand <rgelfa...@gmail.com> wrote: > I just ran spamassassin -D --lint. Below, is the bayes results. > > Jul 21 22:49:41.958 [8922] dbg: bayes: learner_new > self=Mail::SpamAssassin::Plugin::Bayes=HASH(0xb54d58c), > bayes_store_module=Mail::SpamAssassin::BayesStore::DBM > Jul 21 22:49:42.063 [8922] dbg: bayes: learner_new: got > store=Mail::SpamAssassin::BayesStore::DBM=HASH(0xb984c08) > Jul 21 22:49:42.067 [8922] dbg: bayes: tie-ing to DB file R/O > /var/spamassassin/bayes_toks > Jul 21 22:49:42.085 [8922] dbg: bayes: tie-ing to DB file R/O > /var/spamassassin/bayes_seen > Jul 21 22:49:42.121 [8922] dbg: bayes: found bayes db version 3 > Jul 21 22:49:42.130 [8922] dbg: bayes: DB journal sync: last sync: > 1437526591 > > Jul 21 22:49:42.310 [8922] dbg: bayes: DB journal sync: last sync: > 1437526591 > Jul 21 22:49:42.311 [8922] dbg: bayes: corpus size: nspam = 1213, nham = > 785 > Jul 21 22:49:42.336 [8922] dbg: bayes: cannot use bayes on this message; > not enough usable tokens found > Jul 21 22:49:42.337 [8922] dbg: bayes: not scoring message, returning undef > Jul 21 22:49:42.338 [8922] dbg: bayes: DB expiry: tokens in DB: 120539, > Expiry max size: 150000, Oldest atime: 1431846751, Newest atime: > 1437531525, Last expire: 1437456307, Current time: 1437533382 > Jul 21 22:49:42.339 [8922] dbg: bayes: DB journal sync: last sync: > 1437526591 > Jul 21 22:49:42.339 [8922] dbg: bayes: untie-ing > > > > On Tue, Jul 21, 2015 at 9:41 PM Roman Gelfand <rgelfa...@gmail.com> wrote: > >> Good question. I am not sure. How do I tell which dB it is pointing? >> It has been some time since I set it up. >> >> On Tue, Jul 21, 2015, 9:14 PM David B Funk <dbf...@engineering.uiowa.edu> >> wrote: >> >>> On Wed, 22 Jul 2015, Roman Gelfand wrote: >>> >>> > I did numerous sa-learn on this one type of email and spamassasiin >>> makes the following evaluation every time. I am not sure what I >>> > am doing wrong. >>> > >>> > X-Spam-Level: ** >>> > X-Spam-Status: No, score=2.3 required=5.0 tests=BAYES_50,HTML_MESSAGE, >>> > MIME_HTML_ONLY,RDNS_NONE,T_FILL_THIS_FORM_SHORT,T_REMOTE_IMAGE >>> autolearn=no >>> > version=3.3.2 >>> >>> Are you certain that the Bayes database that you did your manual >>> learning to is >>> the same one that your mail filtering system is looking at? >>> >>> "BAYES_50" implies that your Bayes database (as used by your mail >>> filtering >>> system) has no idea (good or bad) what that message is. This implies >>> that either >>> it has -no- tokens that match that message or has an almost equal number >>> of >>> spammy tokens and hammy tokens that match. >>> >>> -- >>> Dave Funk University of Iowa >>> <dbfunk (at) engineering.uiowa.edu> College of Engineering >>> 319/335-5751 FAX: 319/384-0549 1256 Seamans Center >>> Sys_admin/Postmaster/cell_admin Iowa City, IA 52242-1527 >>> #include <std_disclaimer.h> >>> Better is not better, 'standard' is better. B{ >> >>