I use
reject_rbl_client zen.spamhaus.org,
reject_rbl_client b.barracudacentral.org,
reject_rbl_client cbl.abuseat.org,
which I find catches about 98% of SPAM.
I also receive mail at an address that is a forwarding mailbox and
sends mail to my Postfix server. The provider of that mailbox uses a
SPAM filtering service that is significantly less effective than my RBL
recipe above. Since, from my server's viewpoint, the client is the
forwarding service provider (which is trusted), all that SPAM makes it
into my mailbox.
What I'd like to do is apply the same RBL client filtering to hosts
further back in the delivery chain than the immediate client. I.e.
given a chain of Received headers like this:
> Received: from acmsmtp01.acm.org (ACMSMTP01.acm.org [64.238.147.78])
> Received: from in-002.ord.mailroute.net
> Received: from localhost (localhost.localdomain [127.0.0.1])
> Received: from in-002.ord.mailroute.net ([199.89.2.5])
> Received: from theshoemart.wc09.net (theshoemart.wc09.net
[74.203.48.129])
> Received: from arbt04.whatcounts.com (172.16.3.34) by theshoemart.wc09.net
run all the hosts through the RBL lookup and reject if any of the
hosts get a positive result. Is this possible?
-- Jim Garrison
