SA, bayes, BerkeleyDB

[Jeff A. Earickson]

Julian,
   Here's what I got:

sa-learn -D 1 --rebuild
please specify target type with --dir, --file, or --mbox: 1
debug: Score set 0 chosen.
debug: using "/opt/perl5/share/spamassassin" for default rules dir
debug: using "/etc/mail/spamassassin" for site rules dir
debug: using "/home/admin/jaearick/.spamassassin/user_prefs" for user
prefs file
debug: bayes: 9613 tie-ing to DB file R/O
/var/spool/spamassassin/bayes_toks
debug: bayes: 9613 tie-ing to DB file R/O
/var/spool/spamassassin/bayes_seen
debug: debug: Only 11 spam(s) in Bayes DB < 200
debug: bayes: 9613 untie-ing
debug: bayes: 9613 untie-ing db_toks
debug: bayes: 9613 untie-ing db_seen
debug: Score set 0 chosen.
debug: Initialising learner
debug: Initialising learner
debug: lock: 9613 created /var/spool/spamassassin/bayes.lock.emerald.9613
debug: lock: 9613 trying to get lock on /var/spool/spamassassin/bayes with
0 retries
debug: lock: 9613 link to /var/spool/spamassassin/bayes.lock: link ok
debug: bayes: 9613 tie-ing to DB file R/W
/var/spool/spamassassin/bayes_toks
debug: bayes: 9613 tie-ing to DB file R/W
/var/spool/spamassassin/bayes_seen
debug: bayes: 9613 untie-ing
debug: bayes: 9613 untie-ing db_toks
debug: bayes: 9613 untie-ing db_seen
debug: bayes: files locked, now unlocking lock
debug: unlock: 9613 unlink /var/spool/spamassassin/bayes.lock
debug: bayes: 9613 untie-ing

Note that On my first try, it complained about the "auto_report_threshold"
variable in spam.assassin.prefs.conf, so I commented that out.  I also
tried "spamassassin -D --lint" to see what directories it was using and
this same complaint.  My Sol 8 is up-to-date on patches.  My DB is 4.1.25,
built with Sun's Forte 7 compiler, not gcc.

--- Jeff


On Wed, 26 Mar 2003, Julian Field wrote:

> Date: Wed, 26 Mar 2003 17:47:44 +0000
> From: Julian Field <mailscanner at ECS.SOTON.AC.UK>
> Reply-To: MailScanner mailing list <MAILSCANNER at JISCMAIL.AC.UK>
> To: MAILSCANNER at JISCMAIL.AC.UK
> Subject: Re: SA, bayes, BerkeleyDB
>
> I was generating DB output by doing
>          sa-learn -D 1 --rebuild
> and it was saying the nspam and nham were 0.
>
> Maybe your Solaris version didn't have the same problem my Solaris did?
>
> At 16:24 26/03/2003, you wrote:
> >Julian,
> >    I'm on Solaris 8, db 4.1.  My /.spamassassin directory looks like thus:
> >
> >-rw-r--r--   1 root     daemon      1714 Mar 26 11:15 bayes_msgcount
> >-rw-r--r--   1 root     daemon     24576 Mar 26 11:15 bayes_seen
> >-rw-r--r--   1 root     daemon    671744 Mar 26 11:15 bayes_toks
> >
> >(BTW, defining auto_whitelist_path in the MailScanner spam.assassin.prefs.conf
> >file has no effect on where the .spamassassin dir goes, still into
> >the root directory).
> >
> >When I do "check_bayes_db -db /.spamassassin/bayes | more"
> >I get:
> >
> >0.000        0        0        0  non-token data: db format = on-the-fly
> >probs,
> >expiry, scan-counting
> >0.000        0        4        0  non-token data: nspam
> >0.000        0      150        0  non-token data: nham
> >0.000        0    14387        0  non-token data: ntokens
> >0.000        0        0        0  non-token data: oldest age
> >0.000        0     1698        0  non-token data: current scan-count
> >0.000        0        0        0  non-token data: last expiry scan-count
> >0.090        0        2     1542  Sentinel
> >0.020        0       11     1698  N:NNNNNNNN
> >0.149        0        1     1543  H*m:RCXN13905
> >(lots more)
> >
> > >From staring at the source code to check_bayes_db, this seems to be
> >correct behavior, I think.
> >
> >--- Jeff
> >
> >On Wed, 26 Mar 2003, Julian Field wrote:
> >
> > > Date: Wed, 26 Mar 2003 15:44:12 +0000
> > > From: Julian Field <mailscanner at ECS.SOTON.AC.UK>
> > > Reply-To: MailScanner mailing list <MAILSCANNER at JISCMAIL.AC.UK>
> > > To: MAILSCANNER at JISCMAIL.AC.UK
> > > Subject: Re: SA, bayes, BerkeleyDB
> > >
> > > At 15:40 26/03/2003, you wrote:
> > > >Julian,
> > > >    You commented on the list the other day "you need version 3
> > > >of BerkeleyDB installed, beware of version 4".  Why I wonder?
> > > >I have 4.1 installed in /usr/local/BerkeleyDB4.1 with a symlink
> > > >of /usr/local/BerkeleyDB -> BerkeleyDB4.1.  Things seem to be
> > > >working ok with SA, after I changed the "use AnyDBM_file" in
> > > >SA.pm to "use DB_file".   The bayes stuff in /.spamassassin is
> > > >updating for me...
> > >
> > > I was using DB4.1 on Solaris, and SpamAssassin wasn't working properly at
> > > all. It couldn't even get the nspam and nham counters out of the database
> > > files.
> > > It may well be okay on Linux.
> > > --
> > > Julian Field
> > > www.MailScanner.info
> > > MailScanner thanks transtec Computers for their support
> > >
>
> --
> Julian Field
> www.MailScanner.info
> MailScanner thanks transtec Computers for their support
>