On Fri 14/Feb/2020 16:32:52 +0100 Viktor Dukhovni wrote:
Postfix supports a "relocated_maps" feature, which aims to refer the
bounce recipient to a more appropriate address:
http://www.postfix.org/postconf.5.html#relocated_maps
When a RCPT TO hits a listed address it looks like the response will be:
550 5.1.6 User has moved to <lookup result>
That would look like a perfect 551 candidate, wouldn't it?
The 5.1.6 is determined in the mailer triple resolution code, which is
is SMTP independent, and 3-digit SMTP reply code is determined in the
SMTP engine, which maps replies with 550 for all addresses that
resolve to the error channel.
https://tools.ietf.org/html/rfc3463#section-3.2
X.1.6 Destination mailbox has moved, No forwarding address
I'm not familiar with extended status codes, but "No forwarding address" seems
to mean that there is no <lookup result>. Curiously, looking in the IANA page:
https://www.iana.org/assignments/smtp-enhanced-status-codes/smtp-enhanced-status-codes.xhtml
One can find other uses of 251 and 551, not related to forwarding for address
correction. OTOH, no extended status code provides for address correction. Is
that the reason why Postfix relocated maps reply 550 rather than 551, perhaps?
But the result is correct, Postfix is not returning a machine-readable
551 referral, it is returning a human readable[1] 550 reject.
[1] By humans who can read and understand a short English phrase.
I'm sure most people on this list can read NDN. However, generic users who
have no idea of the SMTP dialogue find them difficult to grasp, especially
backscatter. Maybe we're not so human, after all...
Best
Ale
--
_______________________________________________
ietf-smtp mailing list
ietf-smtp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf-smtp