use_bayes not working
Marc Lucke
marc at marcsnet.com
Wed Jun 25 11:47:13 IST 2008
spamassassin broke completely. sa-learn didn't work. Perhaps there's a
Perl version / rpm version conflict?
Hugo van der Kooij wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Marc Lucke wrote:
> | Yes, that's checked. This was a working system. Anywhere I can get it
> | to give me debugging info?
>
> You allready gave it:
>
> [10721] dbg: bayes: tie-ing to DB file R/O /root/.spamassassin/bayes_toks
> [10721] dbg: bayes: tie-ing to DB file R/O /root/.spamassassin/bayes_seen
> [10721] dbg: bayes: found bayes db version 3
> [10721] dbg: bayes: DB journal sync: last sync: 1213214587
>
> [10721] dbg: bayes: DB journal sync: last sync: 1213214587
> [10721] dbg: bayes: corpus size: nspam = 63340, nham = 154630
> [10721] dbg: bayes: score = 0.215791458401563
> [10721] dbg: bayes: DB expiry: tokens in DB: 120875, Expiry max size:
> 150000, Oldest atime: 1212853477, Newest atime: 1214166417, Last expire:
> 1213193735, Current time: 1214192307
> [10721] dbg: bayes: DB journal sync: last sync: 1213214587
>
> This seems to indicate your bayesian database is there and contains both
> HAM and SPAM entries. (Check the atime info on:
> http://esqsoft.com/javascript_examples/date-to-epoch.htm)
>
> But what did you change to make it a non-working system? That is where
> you should be looking for trouble.
>
> For now I would backup the bayesian database and start a fresh set just
> to see if that resolves the issue.
>
> Hugo.
>
> - --
> hvdkooij at vanderkooij.org http://hugo.vanderkooij.org/
> PGP/GPG? Use: http://hugo.vanderkooij.org/0x58F19981.asc
>
> A: Yes.
> >Q: Are you sure?
> >>A: Because it reverses the logical flow of conversation.
> >>>Q: Why is top posting frowned upon?
>
> Bored? Click on http://spamornot.org/ and rate those images.
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.7 (GNU/Linux)
>
> iD8DBQFIYhCBBvzDRVjxmYERAjpUAJ9xpFgkxtcSLYIsl9Uj0vdcgkwJJwCgqr/m
> 3zlXqLFQQ2pJ2440w7QYB74=
> =PT/R
> -----END PGP SIGNATURE-----
More information about the MailScanner
mailing list