Hello Roelof,
On Sat, 3 Jul 2004 03:00:58 +0200 UTC, Roelof Otten wrote:
RO> Hallo Charles,
RO> On Sat, 3 Jul 2004 01:02:13 +0200GMT (3-7-2004, 1:02 +0200, where I
RO> live), you wrote:
>>>> Danger, Will Robinson! Did you know that Outlook 2003 does not generate
>>>> them?
KA>>> Outlook doesn't have to. This is the job of the mail server.
CMG>> What rfc states this? It isn't. Why would a mta do that if it doesn't
CMG>> know where it's coming from?
RO> RFC2822 states this about message-id's:
RO> The "Message-ID:" field provides a unique message identifier that
RO> refers to a particular version of a particular message. The
RO> uniqueness of the message identifier is guaranteed by the host that
RO> generates it (see below).
"The host that generates it" is the host that runs the mua (your
machine), not the mta (the server).
RO> In all cases, properly-operating clients supplying correct
RO> information are preferred to corrections by the SMTP server. In all
RO> cases, documentation of actions performed by the servers (in trace
RO> fields and/or header comments) is strongly encouraged.
RO> Especially when you look at the last paragraph, it's clear that TB
RO> should destroy the message-id before sending the message out.
Quite the opposite. "properly-operating clients supplying correct
information are preferred to corrections by the SMTP server" means that
TB! (the client) should supply the correct information.
Or, lets substitute names here and change the plural of clients into
singular:
properly-operating clients => TheBat!
the SMTP server => qmail
The sentence then becomes:
In all cases, TheBat! supplying correct information is preferred to
corrections by qmail.
I fail to see why TB! should destroy the message-id.
--
Regards, Charles.
Using TB! 2.12 Beta/7 on Windows XP 5.1 Build 2600
________________________________________________
Current version is 2.11.02 | 'Using TBUDL' information:
http://www.silverstones.com/thebat/TBUDLInfo.html
