[Top] [All Lists]

Re: MX to CNAME and (mis)interptretation of 2821

2007-12-14 10:50:56

Hector Santos wrote:
IMHO, 2821bis should state that "putting CNAMES as the target
of MX records is not allowed, as stated in RFC 2181"

Good idea, maybe it helps to reduce the debates about this issue.

this still won't change the BCP requirement for "Industry Ready" 
software to recognize it.

That BCP has no number yet, and Doug can be very creative to turn
anything in DNS into some kind of attack, with a slight prejudice
against SPF's mx-mechanism ;-)

-1 on changing RFC 29821bis.
  +1 for Paul's proposal, maybe adding "so far" to his note, that
     would help if your BCP gets a number (you could add "updates
     2181 and 2821bis" to this BCP as easy as only "updates 2181")

This SMTP Operational Experience is officially recognized in 
RFC 3974

With an IESG note almost half as long as that in RFC 4005..4008.
See also the comment about 3974 in 2821bis.  And of course 3974
talks about aliases for the MX, not aliases of the addresses.

See RFC 974 for historical perspective.

Oops, now I see why they picked 3974 ;-)  Skipping the WKS magic
RFC 974 states:

| Note that the algorithm to delete irrelevant RRs breaks if
| LOCAL has a alias and the alias is listed in the MX records
| for REMOTE.  (E.g. REMOTE has an MX of ALIAS, where ALIAS has
| a CNAME of LOCAL).  This can be avoided if aliases are never
| used in the data section of MX RRs.

Is DNAME relevant for "industry ready" MTAs, and while we're at
it, what about SRV ?  And should 2821bis mention these issues ?