...... Original Message .......
On Wed, 23 Feb 2005 13:58:33 -0700 David MacQuigg <dmq(_at_)gain(_dot_)com>
wrote:
At 12:01 PM 2/23/2005 -0500, you wrote:
...... Original Message .......
On Tue, 22 Feb 2005 15:28:11 -0700 David MacQuigg wrote:
At 09:55 PM 2/22/2005 +0000, you wrote:
Are we in agreement that bounces may come *after* a forwarder's SMTP
session is closed? To repeat my earlier statement: A bounce might come as
late as several hours, when the recipient hits a "Reject as Spam" button.
That reject should be treated as a "bounce" and follow authenticated
addresses all the way back to its source.
No. Not generally. The only time you can bounce without causing
backscatter is if SPF passed.
So how should a forwarder handle a bounce that comes from his receiver
*after* the session with his sender is closed?
I agree an SPF check *during* the session is necessary, but that doesn't
mean the bounce can't come later.
-- Dave
I was referring to the creation of a bounce, not forwarding it.
A forwarder should deal with a bounce much like any other message. Reject
if it Fails SPF. Forward if it doesn't.
The big difference is that if mail-from/return-path is null, the forwarder
must check HELO/EHLO (not a bad idea generally). There has been previous
discussion that perhaps for HELO/EHLO, Neutral or Softfail should be
treated like Fail.
Scott Kitterman