---- Original Message -----
From: "Daniel Feenberg" <feenberg(_at_)nber(_dot_)org>
To: "Frank Ellermann" <nobody(_at_)xyzzy(_dot_)claranet(_dot_)de>
Cc: <asrg(_at_)ietf(_dot_)org>
Sent: Thursday, December 14, 2006 11:50 PM
Subject: Re: [Asrg] Re: bounceone
On Thu, 14 Dec 2006, Frank Ellermann wrote:
Tom Petch wrote:
Less good is
postmaster@<domain-name>
since, while technically correct, it is unlikely to convey the
meaning clearly to a lay user, especially as the <domain-name>
may be unknown to the user.
...
I have always found the overuse of the pronoun "localhost" in DSN messages
a stumbling block to quick understanding. Substituting the actual host
name wouldn't cost anything, would it? I understand there is no actual
ambiguity, just a delay while the user figures it out (or gives up
trying).
I assume you are referring to something like
"The original message was received at Fri, 15 Dec 2006 15:53:51 +0900
from localhost [127.0.0.1]
----- The following addresses had permanent fatal errors -----
<inform(_at_)adscorp(_dot_)jp>
(reason: 553 5.3.0 <inform(_at_)adscorp(_dot_)jp>... No such user here)"
which ties in with a similar use of 'localhost' in the "Received:" of the DSN's
headers. I imagine that this is the way they have configured their MTAs and
yes, I find it unhelpful.
If the DSN included the message-id, it would be possible to automate the
dropping of DSN messages arising from forged return addresses on messages
to misconfigured MTAs. Most of the DSNs I get have the subject line, but
few have the message-id. If no reasonably reliable way to determine the
legitimacy of a DSN is established, then more MTAs will start discarding
all of them, which would be significant reduction in service and not at
all desirable.
I do see the message-id in the text/rfc822-headers, which a typical UA will
display in all its glory, and also in message/rfc822, which it will not. I
haven't yet counted how many of my 400 bounces do use either format but most of
that data is in bouncetwo.
Tom Petch
Daniel Feenberg
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg