On Thu, 2004-12-09 at 02:01, D. D. Brierton wrote: > On Wed, 2004-12-08 at 17:07, Scot L. Harris wrote: > > > I don't think you can use those rbls with fetchmail. The message has > > already been delivered to your mailbox. > > Yes, you are right. > Well I am running Spamassassin anyway, although I haven't upgraded to > FC3 (and hence Spamassassin 3.0) yet (still waiting for Fedora Extras). I'm relatively sure that SA 2 still detects the sender from the rbl even though the mail went through fetchmail (I'm running a similar set-up and it seems SA catches spam that way). If that's really true then you could increase the score of your rbl-hits in SA assuming that you ultimately trust the rbls (but that's what you imply by trying to put them as a postfix-reject) > I am looking forward to upgrading to Spamassassin 3.0 as quite a few > spam messages are getting through now, even with Bayesian filtering. Yes, same here -- until I implemented greylisting. I found that I could also detect the "false cleans" by pulling in loads of RulesDuJour, but that was a bit too much load for my tiny & old mail server... -- HaJo Schatz <hajo@xxxxxxxx> http://www.HaJo.Net PGP-Key: http://www.hajo.net/hajonet/keys/pgpkey_hajo.txt
