At 09:03 PM 3/6/2004 -0500, you wrote:
From: Alex van den Bogaerdt <alex(_at_)ergens(_dot_)op(_dot_)het(_dot_)net>
Subject: Re: [spf-discuss] Strictly my opinion
Date: Sat, 6 Mar 2004 23:11:56 +0100
On Sat, Mar 06, 2004 at 10:42:44AM -0700,
administrator(_at_)yellowhead(_dot_)com wrote:
After examining Microsoft's CallerID proposal in more detail, it is my
opinion that any system which attempts to reject/return mail after the DATA
phase is inherantly flawed.
Maybe it's just me, but I am not sure what you mean here.
There are two entirely different situations:
both: The sending server issues DATA, receives "250 go ahead", sends
its message and ends with CRLF.CRLF
#1: The receiving server then rejects the message
#2: The receiving server accepts the message, then sends a bounce
(probably as the result of scanning the body)
The first situation I call "after DATA", the second situation I
call "after accepting the message".
I base that opinion on the thousands of attempted bounces that we have
[snip]
To me this looks like you are referring to #2 yet you are saying
you refer to #1. Please be explicit.
cheers,
Alex
****************** REPLY SEPARATER *******************
Correct me if I am wrong, but once the DATA phase has begun, there is no
more negotiation until the "." on a new line, followed by a QUIT command
and disconnection. A sender is supposed to wait for responses to HELO, MAIL
FROM:, and RCPT TO:, but some spam engines don't even do that. Examining
anything after the DATA phase has begun seems rather pointless, because the
spammer has already started spewing his garbage and will not pay any
attention to an error code sent at the end. Since spammers don't play by
the rules (and they account for 70% of our message attempts), we must
adjust our thinking to match the majority of the traffic.