Bayes db uninitialised error

[Jon Gerdes <GE...@whl.co.uk>]

Dear all

I have recently started training Bayes deliberately rather than relying on auto learning.  Now for some reason I get "debug: cannot use bayes on this message; db not initialised yet" when running spamassassin -D --lint (first dump).  

I know that everything was OK (see second dump).  All I have done is to run "sa-learn --spam (or ham) --showdots *" in my recently classified maildirs.

I am currently going through my command history to try and find out where it started failing.  Looking at some of my spam I can see BAYES_ rules being tripped.  I've tried sa-learn --rebuild to no avail.  

Just to reiterate, all I've done between dump 2 and 1 is to run sa-learn.  There was a similar post to mine on the list a little while ago but the poster didn't seen to get an answer.

Is this a bug or feature P-) 

SA 2.63, PERL 5.8.1

Cheers
Jon Gerdes


Dump 1
...
debug: bayes: 22175 tie-ing to DB file R/O /var/spamd/bayes_toks
debug: bayes: 22175 tie-ing to DB file R/O /var/spamd/bayes_seen
debug: bayes: found bayes db version 2
debug: Score set 3 chosen.
debug: Initialising learner
debug: Loading languages file...
debug: Language possibly: en,sco
debug: running header regexp tests; score so far=0
debug: running body-text per-line regexp tests; score so far=2.077
debug: bayes corpus size: nspam = 1671, nham = 420
debug: uri tests: Done uriRE
debug: tokenize: header tokens for *F = "U*ignore D*compiling.spamassassin.taint.org D*spamassassin.taint.org D*taint.org D*org"
debug: tokenize: header tokens for *m = " 1089803364 lint_rules "
debug: cannot use bayes on this message; db not initialised yet
debug: bayes: not scoring message, returning 0.5
debug: bayes: 22175 untie-ing
debug: bayes: 22175 untie-ing db_toks
debug: bayes: 22175 untie-ing db_seen
...

Dump 2
...
debug: bayes: 2617 tie-ing to DB file R/O /var/spamd/bayes_toks
debug: bayes: 2617 tie-ing to DB file R/O /var/spamd/bayes_seen
debug: bayes: found bayes db version 2
debug: Score set 3 chosen.
debug: Initialising learner
debug: Loading languages file...
debug: Language possibly: en,sco
debug: running header regexp tests; score so far=0
debug: running body-text per-line regexp tests; score so far=2.077
debug: bayes corpus size: nspam = 1582, nham = 256
debug: uri tests: Done uriRE
debug: tokenize: header tokens for *F = "U*ignore D*compiling.spamassassin.taint.org D*spamassassin.taint.org D*taint.org D*org"
debug: tokenize: header tokens for *m = " 1089795228 lint_rules "
debug: bayes token 'N:H*m:NNNNNNNNNN' => 0.997298245614035
debug: bayes: score = 0.997298245614035
debug: bayes: 2617 untie-ing
debug: bayes: 2617 untie-ing db_toks
debug: bayes: 2617 untie-ing db_seen
...

Re: Bayes db uninitialised error

[Theo Van Dinter <fe...@kluge.net>]

On Wed, Jul 14, 2004 at 12:48:45PM +0100, Jon Gerdes wrote:
> I have recently started training Bayes deliberately rather than relying on auto learning.  Now for some reason I get "debug: cannot use bayes on this message; db not initialised yet" when running spamassassin -D --lint (first dump).  
> 
> Is this a bug or feature P-) 
> debug: cannot use bayes on this message; db not initialised yet

Neither a bug nor a feature.  That message means the message didn't have
enough tokens that were also already in the DB.

In 3.0 the message was made a little clearer:

"cannot use bayes on this message; none of the tokens were found in the database"

:)

-- 
Randomly Generated Tagline:
"The one computer-language course I took was Cobol, and basically, I just
 slept the whole quarter. Then, the night before the final, I read the IBM
 Cobol manual, and I got the top score in the final." - Larry Wall