On 3/16/2014 11:20 AM, Ahmad riza h nst wrote: > Dear all, > > I have some experiences on postfix. > > Currently i work on isp and we have one outgoing server using > postfix+amavis+clamav+spamassassin on freebsd, running well. > > We have plan to migrate our MXs from qpsmtpd+qmail to postfix+amavis > and my problem is i can't find way to implement badmailfromto > filtering on postfix. > > Please point me how to setup this type filtering on postfix. > > About badmailfromto (from qpsmtpd plugins website): > Much like the similar check_badmailfrom, this plugin references both > the FROM: and TO: lines, and if they both are present in the > badmailfromto config file (a tab delimited list of FROM/TO pairs), > then the message is blocked as if the recipient (TO) didn't exist. > This is specifically designed to not give the impression that the > sender is blocked (good for cases of harassment). > > Based heavily on check_badmailfrom. > > Thank you. >
For two-factor decisions, postfix has the built-in smtpd_restriction_classes http://www.postfix.org/RESTRICTION_CLASS_README.html However, this quickly becomes unmaintainable after more than just a few conditions. For a more scalable approach, you can use an external policy service such as postfwd, or write your own. http://www.postfix.org/SMTPD_POLICY_README.html http://www.postfix.org/addon.html#policy Or use a milter. There are probably existing milters with this feature. http://www.postfix.org/MILTER_README.html -- Noel Jones
