On 18 Sep 1997, Neal Becker wrote:
> Yes, they shouldn't, but they are. And users here are complaining
> about not getting their mail. I don't think we should insist that
> exim enforce facist policies. How does this sound:
I don't consider requiring mail to conform to the RFCs to be a facist
policy. Here is what RFC 821 has to say about the MAIL FROM command:
3.1. MAIL
[...]
The first step in the procedure is the MAIL command. The
<reverse-path> contains the source mailbox.
MAIL <SP> FROM:<reverse-path> <CRLF>
This command tells the SMTP-receiver that a new mail
transaction is starting and to reset all its state tables and
buffers, including any recipients or mail data. It gives the
reverse-path which can be used to report errors.
and here is what RFC 1123 says:
5.2.2 Canonicalization: RFC-821 Section 3.1
The domain names that a Sender-SMTP sends in MAIL and RCPT
commands MUST have been "canonicalized," i.e., they must be
fully-qualified principal names or domain literals, not
nicknames or domain abbreviations.
Notice that the <reverse-path> is supposed to provide useful
information, namely, where to send error reports. I fail to see the
usefulness of turning
MAIL FROM: unqualified
into
MAIL FROM: unqualified@unspecified-domain
*especially* when relaying the message on to somewhere else, because
this subtle change has then happened somewhere in the middle of a
delivery path which is not recorded anywhere. SMTP relays should not be
in the business of messing with the messages that pass through them.
This point has been made forcefully in other discussion groups.
> If sender matches sender_unqualified_local_hosts then append my
> default domain. Otherwise append sender_unqualified_default_domain if
> it is set. Otherwise append my default domain.
Well, I certainly can't support that. Even if you managed to persuade me
to provide options for you to break the rules with, I want the default
to follow the rules and reject unqualified addresses.
Heck! There's enough crud floating around the net as it is, and
sometimes I wonder if software should be as forgiving as much of it is.
It generally leads to trouble in the end. Somebody said recently that
"be liberal in what you accept and strict in what you generate" should
apply only when the specification is ambiguous, and I agree with that.
When the specification is clear, junk should be bounced. IMO.
--
Philip Hazel University Computing Service,
ph10@??? New Museums Site, Cambridge CB2 3QG,
P.Hazel@??? England. Phone: +44 1223 334714
--
* This is sent by the exim-users mailing list. To unsubscribe send a
mail with subject "unsubscribe" to exim-users-request@???
* Exim information can be found at http://www.exim.org/
