Sunny
did use the -p /spam.assassin.pref.conf flag on th sa-lean and
spamassassin calls so the things get the correct bayes_path etc?
Also bayes won't kick in till you've given it at least 200 spam AND ham
examples..
--
Martin Hepworth
Snr Systems Administrator
Solid State Logic
Tel: +44 (0)186
I've decided to wipe out my old bayes db from SA 2.6. I'm trying to
rebuild the bayes db by sa-learning some directories full of spam saved
by MailScanner. Sa-learn shows the dots, says it learned from messages,
but when I run a spamassassin -D -t it says there are 0 spams. When I do
a --dump magic
