Arnt Gulbrandsen wrote:
Dilyan Palauzov writes:
Does anybody see any disadvantages of ending the session
...
with
550 Mail rejected by mailing-list(_at_)example(_dot_)org
which implies, that the mail was accepted by all but
mailing-list(_at_)example(_dot_)org?
The problem I see is that 550 does not imply "accepted" at all. 550 is
not a partial success code.
If its not already clearly stated in 2821 and in 2821bis, then maybe
there should be a statement added to 2821bis stating to the effect that
a 550 not only means a permanent rejection for the transaction, but it
also *implies* to the SENDER that the recipients ARE NOT going to
receive the mail - atleast not "officially."
I really hope we don't open the DOOR to the idea that a SMTP server can
begin to tell the SENDER mail was rejected but actually LIE about this
delivery status and delivered the mail anyway.
I can understand and see how GREYLISTING implementations can help "mold"
the thinking in this directon.
--
Sincerely
Hector Santos, CTO
http://www.santronics.com
http://santronics.blogspot.com