Greetings folks,

Wondering if anyone here is still using the MAPS RBL? They recently
stopped allowing DNS zone xfers and are forcing clients to use direct
lookups to their servers. Typically my postfix config (with zone
xfers) would look like this:

smtpd_recipient_restrictions =
<snip>
reject_rbl_client rbl-plus.mail-abuse.org,
<snip>

Mail was being rejected properly with the default RBL reply:

mx03 postfix/smtpd[12565]: [ID 197553 mail.info] NOQUEUE: reject: RCPT
from c55-239.icpnet.pl[62.21.55.239]: 554 Service unavailable; Client
host [62.21.55.239] blocked using rbl-plus.mail-abuse.org;
from=<dfhpjsqaomyq@es.oetiker.com> to=<rgehl@example.com> proto=SMTP
helo=<c55-239.icpnet.pl>

---

The new cfg looks like this:

# smtpd_recipient_restrictions=
...
reject_rbl_client activationcode.r.mail-abuse.com,
....

(where "activationcode" is the special string that Trend sends you)

# Add the command to check for a rbl_reply map.

* rbl_reply_maps = hash:/$config_directory/rbl_reply

Then create the rbl_reply map:

activationcode.r.mail-abuse.com 550 Service unavailable; $rbl_class
[$rbl_what] blocked using Trend Micro RBL+. Please see
http://www.mail-abuse.com/cgi-bin/lookup?ip_address=$rbl_what${rbl_reason?;
$rbl_reason}

....then postmap hash:rbl_reply, then reload postfix and it's supposed
to fire right up.

Thus far, however, I've put these different cfgs in place in the
main.cf, created and postmapped the reply file, and there have been
zero hits on this (compared to roughly one every 5-10 mins when using
the old config).

Any ideas where I may have gone wrong?