Ian Eiloart <iane(_at_)sussex(_dot_)ac(_dot_)uk> wrote:
I understand the spec. I've not read the discussions leading to it, but if
there's a GOOD reason to assume every AAAA record (without a corresponding
MX record) points to a mail host, then I'd like to hear it.
There are a number of historical assumptions behind this rule, for
example:
- any email address is an honest attempt to reach a person or process;
- any host connected to the Internet should be able to receive email;
- publishing MX records is difficult for some of these hosts...
The rule actually made sense when it was first written. I believe the
three assumptions listed above are all false today, but I agreed to
accept the consensus call on this issue for RFC 5321 because with Doug
Otis on one side and Dave Crocker on the other, I didn't want to wait
for one of them to die.
;^)
You can see that the middle assumption is "close enough" to your
question: in actuality, the assumption is that any host with an address
record in DNS, whether IPv4 or IPv6, is worth trying in the interests
of making email delivery reliable.
We could escape from this rule by issuing a separate RFC listing
justification for exceptions to the rule, but I had to agree with John
Klensin that "fixing" this as _part_of_ the transition from Proposed
Standard to Draft Standard was arguably an over-reach.
--
John Leslie <john(_at_)jlc(_dot_)net>
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg