From: "Derek J. Balling" <dredd(_at_)megacity(_dot_)org>
Message-Id: <6DCBBFCC-4FED-11D7-8D9B-000A27AF5202(_at_)megacity(_dot_)org>
Date: Thu, 6 Mar 2003 11:05:19 -0500
So in the DS model, you would configure your internal relay to do the
DS test, which had a sender of <asrg-admin(_at_)ietf(_dot_)org>, and ask IETF
"is
132.151.1.19 authorized to send mail for ietf.org?", get an assurance
of trust, and move on.
You could then set up your final-MX to say "trust all mail from my
internal relay, because it's already done those checks itself".
The received headers never come into play. It's strictly concerned
sender-domain (from the envelope) and connecting IP.
BUT there is no envelope sender re-writing for (in sendmail parlance):
- aliases without list owners
- .forward forwarding
Should there be? It would be a different world.
While that doesn't rule out adoption of RMX, it does make even
its adoption have pretty big impact on everyday normal practices.
P.S. I have now twice succumbed to the temptation to add
to the high-volume small-topic flood. Now it's out of my
system, and I see the wisdom in Paul Judge's trying to move
the discussion toward longer and systematic drafts rather
than point-by-point reaction. So I'll be quiet for a while.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg