It would be nice to be able to do spam control from one location.
I would like to see a sender_host_reject_bydns option. It should be
designed to allow the checking done by the sender_host_reject via
an DNS query instead of a local lookup.
My specific need is to be able to have a central database on one
machine that I can maintain easily. Our mail network is growing and I
don't want to have to try and maintain a flat file on multiple machines.
I am close to being able to do all of our internal per user mail routing
via exim with the exact same set of configuration files on each mail
server. I don't want to introduce a file that must be maintained on
multiple machines if I don't have to.
This might also be a suitable way for multiple ISP's to combat
spam by having one server that can be maintained and queried by
any mail servers that might want to.
For example we specify that sender_host_reject_bydns is a string
containg a domain suffix to add to any host name that you would
then attempt a lookup a TXT record for. E.g. if we had:
send_host_reject_bydns = blacklist.poste.com
Then DNS queries for TXT could be made to see if a host was blacklisted.
E.g.:
spam-sender@???
would result in a DNS queries for:
spam-sender.spam.cyberpromo.com.blacklist.poste.com.
If there is a TXT record returned for any of them the message is
refused. The TXT data could be put in the rejectlog giving some
information on when and why the address was added to the list.
For the above example I used the following in my zone file:
cyberpromo.com.blacklist IN TXT "major spam site"
*.cyberpromo.com.blacklist IN TXT "major spam site"
All mail from anything under cyberpromo.com will be rejected.
Another example:
spam.allvip.com.blacklist IN TXT "another major spam site"
Only mail from user spam@??? or anything@??? would
be rejected.
--
Stuart Lynne <sl@???> 604-933-1000 <http://www.poste.com>
PGP Fingerprint: 28 E2 A0 15 99 62 9A 00 88 EC A3 EE 2D 1C 15 68
