I feed all of my spam into Bayes. Stuff that slips through gets fed into
bayes as well.
but I never see any Bayes hits.
spamassassin -D --lint gives me this:
[28551] dbg: uri: running uri tests; score so far=0.96
[28551] dbg: bayes: DB journal sync: last sync: 1150988415
[28551] dbg: bayes: corpus size: nspam = 24485, nham = 3481
[28551] dbg: bayes: score = 0.0757736831405458
[28551] dbg: bayes: DB journal sync: last sync: 1150988415
[28551] dbg: bayes: untie-ing
[28551] dbg: bayes: untie-ing db_toks
[28551] dbg: bayes: untie-ing db_seen
[28551] dbg: plugin: registering glue method for check_uridnsbl
(Mail::SpamAssassin::Plugin::URIDNSBL=HASH(0x82f01e4))
[28551] dbg: plugin: registering glue method for
_mimeheader_eval___ZMIde_EBAYBILL2
(Mail::SpamAssassin::Plugin::MIMEHeader=HASH(0x903ae64))
[28551] dbg: plugin: registering glue method for
_mimeheader_eval___ZMIde_EBAYBILL1
(Mail::SpamAssassin::Plugin::MIMEHeader=HASH(0x903ae64))
[28551] dbg: rules: ran eval rule BAYES_20 ======> got hit
[28551] dbg: rules: running raw-body-text per-line regexp tests; score
so far=0.22
[28551] dbg: rules: running full-text regexp tests; score so far=0.22
[28551] dbg: plugin: registering glue method for check_razor2_range (M
*snipped some seemingly unrelated stuff*
[28551] dbg: plugin: registering glue method for
check_from_in_auto_whitelist
(Mail::SpamAssassin::Plugin::AWL=HASH(0x9028054))
[28551] dbg: config: using "/home/jnichols/.spamassassin" for user state dir
[28551] dbg: locker: safe_lock: created
/home/jnichols/.spamassassin/auto-whitelist.lock.mailgate.pbp.net.28551
[28551] dbg: locker: safe_lock: trying to get lock on
/home/jnichols/.spamassassin/auto-whitelist with 0 retries
[28551] dbg: locker: safe_lock: link to
/home/jnichols/.spamassassin/auto-whitelist.lock: link ok
[28551] dbg: auto-whitelist: tie-ing to DB file of type DB_File R/W in
/home/jnichols/.spamassassin/auto-whitelist
[28551] dbg: auto-whitelist: db-based
[EMAIL PROTECTED]|ip=none scores 0/0
[28551] dbg: auto-whitelist: AWL active, pre-score: 2.166, autolearn
score: 2.166, mean: undef, IP: undef
[28551] dbg: auto-whitelist: DB addr list: untie-ing and unlocking
[28551] dbg: auto-whitelist: DB addr list: file locked, breaking lock
[28551] dbg: locker: safe_unlock: unlink
/home/jnichols/.spamassassin/auto-whitelist.lock
[28551] dbg: auto-whitelist: post auto-whitelist score: 2.166
[28551] dbg: rules: running body-text per-line regexp tests; score so
far=2.166
[28551] dbg: uri: running uri tests; score so far=2.166
[28551] dbg: rules: running raw-body-text per-line regexp tests; score
so far=2.166
[28551] dbg: rules: running full-text regexp tests; score so far=2.166
[28551] dbg: check: is spam? score=2.166 required=5
[28551] dbg: check:
tests=BAYES_20,MISSING_SUBJECT,NO_REAL_NAME,NO_RECEIVED,NO_RELAYS,TO_CC_NONE
[28551] dbg: check:
subtests=__HAS_MSGID,__MSGID_OK_DIGITS,__MSGID_OK_HOST,__NONEMPTY_BODY,__SANE_MSGID,__SARE_WHITELIST_FLAG,__UNUSABLE_MSGID
mailgate log #
---
So, the test hits Bayes, but no spam is. Razor2 is helping out quite a
bit, but I never see anything from DCC or BAYES anymore. :-(
I have use_bayes 1 in local.cf, and I'm using amavisd-new. I don't
remember the last time Bayes was working, because things have been ok.
Just recently a LOT of spam has been slipping through. :-|
If anyone wants the full output from spamassassin -D --lint, let me know.
Otherwise..any ideas?
--
Jonathan
