[Top] [All Lists]

Re: Last Call: draft-klensin-rfc2821bis

2008-03-26 14:36:35
Keith Moore wrote:
because conditions are different now than it was when RFC 974 was 
written, and the case that compelled fallback to A in RFC 974 does not 
significantly exist for AAAA.

I agree with Keith here.

The published RFC 2821 states in section 5:
  "... The lookup first attempts to locate an MX
   record associated with the name.  ...   If
   no MX records are found, but an A RR is found, the A RR is treated as
   if it was associated with an implicit MX RR, with a preference of 0,
   pointing to that host.  If one or more MX RRs are found for a given
   name, SMTP systems MUST NOT utilize any A RRs associated with that
   name unless they are located using the MX RRs; the "implicit MX" rule
   above applies only if there are no MX records present."

There's too much running code to change that now.  However, that doesn't
mean we have to subject AAAA only hosts to the same treatment.  See
Keith's previous post referencing "every power meter, parking meter,
traffic signal" to really understand why this wouldn't be a good idea.
For this reason, I don't agree with section 5 of

For those purists who don't want *any* lookups beyond the MX: sorry, but
we have to be backwards compatible here.  When trying to deliver mail,
we should do an MX lookup.  If there are no MX RRs, we still need to
lookup the A record using "implicit MX" rule to stay backwards
compatible.  That doesn't mean you need to lookup AAAA records, though.

The only real argument I've heard for using AAAA records as implicit MX
RRs is to try and keep IPv4 like IPv6 as far as a few lines of code go
for the application side of things.

Just my thoughts,
IETF mailing list