RE: [SAtalk] 2.60 Upgrade - SpamD not using trained bayes databas e

[Smart,Dan]

Title: Blank Stationery

It was suggested that this is an upgrade problem and I need to do a sa-learn --import.  No.  In fact, I run my existing mbox files with spam and ham through it.   For whatever reason, SA does not think I have any ham.  Yet when I run sa-learn --ham  I get:   ==== snip ==== [EMAIL PROTECTED] filter]$ sa-learn -D --ham --mbox djs-ham ==== snip ==== debug: using "/usr/share/spamassassin" for default rules dir debug: using "/etc/mail/spamassassin" for site rules dir debug: using "/home/filter/.spamassassin/user_prefs" for user prefs file debug: Failed to parse line in SpamAssassin configuration, skipping: timelog_path /tmp/satiming debug: Not allowing user rules! debug: bayes: 20606 tie-ing to DB file R/O /home/filter/.spamassassin_toks debug: bayes: 20606 tie-ing to DB file R/O /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: bayes: Not available for scanning, only 0 ham(s) in Bayes DB < 200 debug: bayes: 20606 untie-ing debug: bayes: 20606 untie-ing db_toks debug: bayes: 20606 untie-ing db_seen debug: Score set 0 chosen. debug: Initialising learner debug: Initialising learner debug: Syncing Bayes journal and expiring old tokens... debug: lock: 20606 created /home/filter/.spamassassin.lock.cavendish.vul.com.20606 debug: lock: 20606 trying to get lock on /home/filter/.spamassassin with 0 retries debug: lock: 20606 link to /home/filter/.spamassassin.lock: link ok debug: bayes: 20606 tie-ing to DB file R/W /home/filter/.spamassassin_toks debug: bayes: 20606 tie-ing to DB file R/W /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: Syncing complete. debug: Learning Ham debug: uri tests: Done uriRE debug: [EMAIL PROTECTED]: already learnt correctly, not learning twice debug: Learning Ham debug: uri tests: Done uriRE debug: [EMAIL PROTECTED]: already learnt correctly, not learning twice ==== snip ====   and on and on... but sa-learn --magic   ==== snip ==== [EMAIL PROTECTED] filter]$ sa-learn --dump magic 0.000          0          2          0  non-token data: bayes db version 0.000          0       1687          0  non-token data: nspam 0.000          0          0          0  non-token data: nham 0.000          0      36864          0  non-token data: ntokens ==== snip ====   What gives?   <<Dan>> From: Smart,Dan Sent: Tuesday, September 30, 2003 10:19 AM To: [EMAIL PROTECTED] Subject: RE: [SAtalk] 2.60 Upgrade - SpamD not using trained bayes databas e Got a direct email suggesting that the wrong config file may be being used (thanks F. Goudal).  Here's the full spamd -D transcript.  Note I've tried running this without the -u filter with no change.   ==== snip ==== -->spamd -D -x -u filter -H /home/filter debug: Preloading modules with HOME=/tmp/spamd-19659-init debug: ignore: test message to precompile patterns and load modules debug: using "/usr/share/spamassassin" for default rules dir debug: using "/etc/mail/spamassassin" for site rules dir debug: Failed to parse line in SpamAssassin configuration, skipping: timelog_path /tmp/satiming debug: Not allowing user rules! debug: bayes: 19659 tie-ing to DB file R/O /home/filter/.spamassassin_toks debug: bayes: 19659 tie-ing to DB file R/O /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: bayes: Not available for scanning, only 0 ham(s) in Bayes DB < 200 debug: bayes: 19659 untie-ing debug: bayes: 19659 untie-ing db_toks debug: bayes: 19659 untie-ing db_seen debug: Score set 1 chosen. debug: Initialising learner debug: bayes: 19659 tie-ing to DB file R/O /home/filter/.spamassassin_toks debug: bayes: 19659 tie-ing to DB file R/O /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: bayes: Not available for scanning, only 0 ham(s) in Bayes DB < 200 debug: bayes: 19659 untie-ing debug: bayes: 19659 untie-ing db_toks debug: bayes: 19659 untie-ing db_seen debug: is Net::DNS::Resolver available? yes debug: servers: vul.com debug: Looking up MX records for user specified servers: vul.com debug: trying (3) vul.com... debug: looking up MX for 'vul.com' debug: MX for 'vul.com' exists? 1 debug: MX lookup of vul.com succeeded => Dns available (set dns_available to hardcode) debug: is DNS available? 1 debug: all '*From' addrs: [EMAIL PROTECTED] debug: running header regexp tests; score so far=0 debug: running body-text per-line regexp tests; score so far=1.27 debug: running raw-body-text per-line regexp tests; score so far=1.27 debug: running uri tests; score so far=1.27 debug: uri tests: Done uriRE debug: running full-text regexp tests; score so far=1.27 debug: DCCifd is not available: no r/w dccifd socket found. debug: DCC is available: /usr/local/bin/dccproc debug: entering helper-app run mode debug: leaving helper-app run mode debug: DCC -> check timed out after 5 secs. debug: all '*To' addrs: debug: RBL: success for 1 of 1 queries debug: running meta tests; score so far=1.27 debug: is spam? score=1.27 required=6.5 tests=DATE_MISSING,NO_REAL_NAME server started on port 783/tcp (running version 2.60) server pid: 19659 logmsg: server started on port 783/tcp (running version 2.60) ==== snip ====   Its not a config file issue. From: Smart,Dan Sent: Tuesday, September 30, 2003 8:22 AM To: [EMAIL PROTECTED] Subject: RE: [SAtalk] 2.60 Upgrade - SpamD not using trained bayes databas e A little more info:  Just upgrade from SA 2.55 to 2.60.  I've got everything working except for SPAMD not using the Bayes database.  I need help debugging.   I run this on RH 7.3 with Postfix and Procmail   <<Dan>> From: Smart,Dan Sent: Monday, September 29, 2003 10:56 AM To: [EMAIL PROTECTED] Subject: [SAtalk] Not using bayes database I just taught. I just finished training my new bayes classifier with my spam and ham files.  Since I run SPAMD as a user "-u filter", I did this as user "filter".   'su filter', then sa-learn --ham and --spam.  The database has been created, and running sa-learn --dump I get:   ==== snip ====   0.000          0          2          0  non-token data: bayes db version 0.000          0       1687          0  non-token data: nspam 0.000          0          0          0  non-token data: nham 0.000          0      36864          0  non-token data: ntokens 0.000          0          0          0  non-token data: oldest atime 0.000          0          0          0  non-token data: newest atime 0.000          0          0          0  non-token data: last journal sync atime 0.000          0          0          0  non-token data: last expiry atime 0.000          0          0          0  non-token data: last expire atime delta 0.000          0          0          0  non-token data: last expire reduction count ==== snip ====   Note: not really sure what this means, but I checked my ham database with sa-learn -D --ham and it said it already learnt correctly the ham, so I know its trained.   When I run SPAMD -D it says:   ==== snip ==== debug: bayes: 17403 tie-ing to DB file R/O /home/filter/.spamassassin_toks debug: bayes: 17403 tie-ing to DB file R/O /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: bayes: Not available for scanning, only 0 ham(s) in Bayes DB < 200 debug: bayes: 17403 untie-ing debug: bayes: 17403 untie-ing db_toks debug: bayes: 17403 untie-ing db_seen debug: Score set 1 chosen. debug: Initialising learner debug: bayes: 17403 tie-ing to DB file R/O /home/filter/.spamassassin_toks debug: bayes: 17403 tie-ing to DB file R/O /home/filter/.spamassassin_seen debug: bayes: found bayes db version 2 debug: bayes: Not available for scanning, only 0 ham(s) in Bayes DB < 200 debug: bayes: 17403 untie-ing debug: bayes: 17403 untie-ing db_toks debug: bayes: 17403 untie-ing db_seen ==== snip ====   The database looks ok in /home/filter: -rw-rw-rw-    1 filter   filter    2621440 Sep 29 10:50 .spamassassin_seen -rw-rw-rw-    1 filter   filter   42082304 Sep 29 10:50 .spamassassin_toks What am I doing wrong?   <<Dan>>