AMaViS and DSPAM Training
uxbod at splatnix.net
Wed Feb 27 14:14:09 CET 2013
we have amavis and SA already in place, which is working very well indeed, though we would like to enable DSPAM as-well. What we have found is that the auto-learn functionality does not seem to train DSPAM when SA triggers either spam/ham. At present we have this in our configuration file:
@spam_scanners = (
['SpamAssassin', 'Amavis::SpamControl::SpamAssassin' ],
['DSPAM', 'Amavis::SpamControl::ExtProg', $dspam,
[ qw(--stdout --classify --deliver=innocent,spam --mode=tum
--tokenizer=chained,noise --user), $daemon_user ],
learn_ham => [ qw(--class=innocent --source=error--user), $daemon_user ],
learn_spam => [ qw(--class=spam --source=error --user), $daemon_user ],
mail_body_size_limit => 64000, score_factor => 0.1,
Is this the correct way of defining DSPAM ? I read an archive posting, from last year, http://firstname.lastname@example.org/msg01943.html that it would not work due to the headers and that the OP ended up writing a custom hook.
Would welcome advice on how we can get it to work please ?
More information about the amavis-users