Re: Didn't receive the original post (to which this was a reply)

From	Greg Wooledge <greg@wooledge.org>
Newsgroups	linux.debian.user
Subject	Re: Didn't receive the original post (to which this was a reply)
Date	2026-04-30 21:20 +0200
Message-ID	<MPF5v-1ASj-3@gated-at.bofh.it> (permalink)
References	(5 earlier) <MOY0x-17PV-1@gated-at.bofh.it> <MP6AN-1dEL-1@gated-at.bofh.it> <MPD3H-1zF5-3@gated-at.bofh.it> <MPDQ5-1zVb-1@gated-at.bofh.it> <MPEsN-1AoB-1@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Apr 30, 2026 at 19:29:32 +0100, Joe wrote:
> Sorry, yes, not quite awake. It's the Reply To: where the bounce is
> sent, but if that isn't set in the headers, the envelope address is
> copied to a new Reply To. A spammer makes sure the Reply To: is set.

That's not correct.  The Reply-To: header is just as ornamental as
the From: header, as far as SMTP is concerned.  SMTP only looks at
the envelope addresses.

> The bounce mechanism is part of the message processing, after the SMTP
> transaction has already accepted the message. A rejection during the
> transaction simply terminates the SMTP handshake, sending an
> explanatory message back to the actual sending IP address and never
> actually receiving any of the message body or headers. There's no way to
> route the rejection to anywhere other than the sender IP address, since
> that's the other end of the transaction. The envelope sender address may
> not resolve to the sender IP address, though it generally must resolve
> in DNS.

That's almost correct.  It was correct until you mentioned IP addresses.
An MTA that needs to generate a bounce message will not remember the IP
address of the sender, nor attempt to make any connections based on the
original sending machine's IP address.  It looks *only* at the envelope
sender address, which is an email address, not an IP address.

There is no reason to assume that the machine which sent you a message
is the proper one to send a bounce to.  The machine that sent you the
message may not even be open to receiving messages at all.  It might
be a one-way sender.

> A rejection is far better than a bounce, using less bandwidth and never
> involving any other party.

Correct.

> One of the main problems of twenty years or
> so ago was than many businesses used an ISP to receive their email
> which was then downloaded to the company server by POP3, or very
> occasionally, IMAP4. Whatever the company server did, the ISP's server
> had already accepted the email, so if the recipient didn't exist, the
> company server had no option but to either bounce it or drop it.

That's a piece of the complications, yes.  Another piece is the
secondary MX.  A domain may declare more than one MX (mail exchange) in
DNS, with priorities, so that the primary MX is used unless it's down.
In that case, the secondary MX will receive incoming messages, and
will probably queue them up for delivery to the primary MX.

The issue with the secondary MX is that it might not know which local
user addresses are valid, and which are invalid.  Think about the
simplest case, where local users are defined in /etc/passwd on the
primary MX.  If the secondary MX doesn't have a copy of that passwd
file, it won't know which user accounts exist.  Then you add aliases,
~/.forward or equivalent files, virtual domains, etc.  It's a mess.

It's best not to have a secondary MX unless you have a mechanism that
ensures the secondary MX can reject invalid local users in the absence
of the primary MX.  For most organizations, that's just not going to be
feasible, so sticking with only the primary MX is the better choice.
Just don't let it remain down for very long.

Back to linux.debian.user | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Moribund mailing list CGS <etphonehomefrance@gmail.com> - 2026-04-28 17:10 +0200
  Re: Moribund mailing list Norwid Behrnd <nbehrnd@yahoo.com> - 2026-04-28 17:30 +0200
    Re: Moribund mailing list "Jeremy Nicoll" <jn.ml.dbi.73@letterboxes.org> - 2026-04-28 19:20 +0200
      Re: Moribund mailing list David Wright <deblis@lionunicorn.co.uk> - 2026-04-28 20:20 +0200
        Re: Moribund mailing list "Jeremy Nicoll" <jn.ml.dbi.73@letterboxes.org> - 2026-04-28 20:40 +0200
          Re: Moribund mailing list Andy Smith <andy@strugglers.net> - 2026-04-28 21:50 +0200
          Re: Moribund mailing list David Wright <deblis@lionunicorn.co.uk> - 2026-04-29 01:00 +0200
            Re: Moribund mailing list "Jeremy Nicoll" <jn.ml.dbi.73@letterboxes.org> - 2026-04-29 01:40 +0200
  Re: Moribund mailing list Felix Miata <mrmazda@earthlink.net> - 2026-04-28 17:50 +0200
    Didn't receive the original post (to which this was a reply) (was: Re: Moribund mailing list) rhkramer@gmail.com - 2026-04-28 18:30 +0200
      Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Nicolas George <george@nsup.org> - 2026-04-28 19:30 +0200
        Re: Didn't receive the original post (to which this was a reply) (was: Re: Moribund mailing list) rhkramer@gmail.com - 2026-04-28 21:50 +0200
      Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Andy Smith <andy@strugglers.net> - 2026-04-28 20:10 +0200
        Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Jan Claeys <lists@janc.be> - 2026-04-28 21:20 +0200
          Re: Didn't receive the original post (to which this was a reply) nwe <nwe@gitcoding.net> - 2026-04-28 21:50 +0200
          Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Andy Smith <andy@strugglers.net> - 2026-04-28 21:50 +0200
            Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Joe <joe@jretrading.com> - 2026-04-28 23:00 +0200
              Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Nicolas George <george@nsup.org> - 2026-04-28 23:20 +0200
                Re: Didn't receive the original post (to which this was a reply) Nicolas George <george@nsup.org> - 2026-04-28 23:30 +0200
                Re: Didn't receive the original post (to which this was a reply) nwe <nwe@gitcoding.net> - 2026-04-29 00:00 +0200
                Re: Didn't receive the original post (to which this was a reply) Nicolas George <george@nsup.org> - 2026-04-29 11:10 +0200
                Re: Didn't receive the original post (to which this was a reply) nwe <nwe@gitcoding.net> - 2026-04-28 23:30 +0200
                Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Joe <joe@jretrading.com> - 2026-04-29 09:10 +0200
                Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Nicolas George <george@nsup.org> - 2026-04-29 11:10 +0200
              Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) <tomas@tuxteam.de> - 2026-04-29 12:00 +0200
              Re: Didn't receive the original post (to which this was a reply) (was: Re: Moribund mailing list) "Roy J. Tellason, Sr." <roy@rtellason.com> - 2026-04-29 18:50 +0200
                Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Nicolas George <george@nsup.org> - 2026-04-29 19:00 +0200
            Re: Didn't receive the original post (to which this was a reply) Stefan Monnier <monnier@iro.umontreal.ca> - 2026-04-28 23:20 +0200
              Re: Didn't receive the original post (to which this was a reply) Chris Green <cl@isbd.net> - 2026-04-29 08:30 +0200
                Re: Didn't receive the original post (to which this was a reply) Joe <joe@jretrading.com> - 2026-04-29 09:00 +0200
                Re: Didn't receive the original post (to which this was a reply) Stefan Monnier <monnier@iro.umontreal.ca> - 2026-04-30 17:50 +0200
                Re: Didn't receive the original post (to which this was a reply) Joe <joe@jretrading.com> - 2026-04-30 19:10 +0200
                Re: Didn't receive the original post (to which this was a reply) Greg Wooledge <greg@wooledge.org> - 2026-04-30 20:00 +0200
                Re: Didn't receive the original post (to which this was a reply) Joe <joe@jretrading.com> - 2026-04-30 20:40 +0200
                Re: Didn't receive the original post (to which this was a reply) Greg Wooledge <greg@wooledge.org> - 2026-04-30 21:20 +0200
              Re: Didn't receive the original post (to which this was a reply) CGS <etphonehomefrance@gmail.com> - 2026-04-30 16:50 +0200
            Google employee expertise (was: Re: Didn't receive the original post (to which this was a reply) (was: Re: Moribund mailing list)) rhkramer@gmail.com - 2026-04-30 16:00 +0200
              Re: Google employee expertise (was: Re: Didn't receive the original post (to which this was a reply) (was: Re: Moribund mailing list)) rhkramer@gmail.com - 2026-05-01 02:30 +0200
                Re: Google employee expertise (was: Re: Didn't receive the original  post (to which this was a reply) Thomas Dineen <tdineen2021@gmail.com> - 2026-05-01 22:20 +0200
                On tolerance [was: Google employee expertise ...] <tomas@tuxteam.de> - 2026-05-02 07:40 +0200
      Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Jeffrey Walton <noloader@gmail.com> - 2026-04-28 21:10 +0200
        Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Joe <joe@jretrading.com> - 2026-04-28 21:30 +0200
      Re: Didn't receive the original post (to which this was a reply)  (was: Re: Moribund mailing list) Anders Andersson <pipatron@gmail.com> - 2026-04-28 21:30 +0200
      Re: Didn't receive the original post (to which this was a reply) Richmond <dnomhcir@gmx.com> - 2026-04-28 22:20 +0200
        Re: Didn't receive the original post (to which this was a reply) rhkramer@gmail.com - 2026-04-28 22:30 +0200
  Re: Moribund mailing list Radhitya <alif@radhitya.org> - 2026-05-01 15:50 +0200

csiph-web