So you're AOL and you have just received 300 million pieces of
undeliverable mail...
do you drop them or try to generate 1 billion bounces that may come
right back at
you, not to mention the complaints as your mailers are used in reverse
attacks?
The best thing to do with undeliverable mail is not to accept it for
delivery. It's then the responsibility of the sending MTA
to generate any notification.
This is one reason why recipients policy would be best
enforced as near to the sender as possible.
So we need some method of propagating recipient policy back towards the
sender - i.e. per-user configuration of filtering at the MTA at the
organizational boundary (or at a filtering service provider).
Clearly there are issues involved with getting this to work with firewalls,
bastions, multiple MX etc.
I think that this is a interesting area for investigation. There are some
deployed systems which do this kind of thing already (sorry, can't point to
one - has been discussed on list).
This has been identified as a standards opportunity.
--
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg