On Mon, 2003-09-15 at 10:54, Alan DeKok wrote:
david nicol <davidnicol(_at_)pay2send(_dot_)com> wrote:
Okay, DRIP
http://www.ietf.org/internet-drafts/draft-brand-drip-01.txt
is very similar; however instead of
..
OMX would have example.com list
omx.m.example.com IN A 192.0.2.10
omx.m.example.com IN A 192.0.2.11
Which is overkill. The recipients of a message want to know if a
particular IP is marked as "sending MX". Forcing the recipient to
wade through tons of unwanted information is inefficient.
one UDP packet is not "tons of unwanted information." We're
looking at at most seven integers and seeing if the one we are
interested in is equal to one of them.
This was my preference:
http://www.ietf.org/internet-drafts/draft-fecyk-dsprotocol-02.txt
It appears to have expired, though.
"designated senders" are looked up by reverse IP, which is simple to
configure in DNS, and allows simple wildcarding for the "third-party
relaying" issue. Further, the response can be a TXT record, with
additional consent information.
The other benefit is that when the recipient MTA asks "is this IP
permitted to send messages as your domain?", then the response is
"yes", or "no", or "unknown". This response is much more helpful than
most similar proposals, which involve responding with huge amounts of
information.
Alan DeKok.
The "cram-them-all-into-A-records" method is meant to simplify
DNS administration. It is the simplest possible approach. The
second approach, suggested in the extended listing draft as
appropriate when there are many addresses to list, gives at least 32
bits of information in the response. We could mask off 32 bits of
AAAA record and have equivalence.
I understand your preference for more information in the response than
a binary listed/not-listed to imply a preference for defining only
a reverse-style listing and deprecating the arguably simpler
multiple A-records listing style at all, in proposed listing mechanisms.
--
david nicol / A thousand towers rise before me and I cannot climb them all.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg