You are here: Home > Participate > Join a Discussion > Mailman Archives
<<< Chronological >>> Author Index    Subject Index <<< Threads >>>

Re: Addition to spam definition

  • From: Anders Andersson < >
  • Date: Tue, 19 Jun 2001 20:19:47 +0200 (MET DST)

On Fri, 11 May 2001, Freek de Kruijf <F.deKruijf@localhost wrote:
>I would like to add to the definition also any (Commercial) e-mail with
>an improper envelope-from-address. Improper meaning that a non-delivery
>message to that address can not be delivered within a reasonable time to
>the mailbox represented by the envelope-from-address. I consider 1 hour
>a reasonable time, network outages excluded.

While this may take care of the immediate delivery notification
problems, what about return addresses that are technically valid
but lead back to an effective null sink, say nobody@localhost
or just any mailbox that is never checked for error returns?

I have seen a number of cases where a newsletter or similar is
sent to a recipient address with broken forwarding, resulting in
a bounce back to the sender, plus a note to the local postmaster
(i.e. me).  While I may take the opportunity to close the broken
account and begin rejecting any mail already during the SMTP
dialogue, this isn't always the desirable solution.  In some
cases, this has been going on for months.  If the sender never
bothers to check the bounces, what are the chances they will
remove the bad address once their server is forced to generate
the error messages itself?  Do mailing list maintainers consider
bounced mail a natural thing in life, just like rain?  Will they
continue to bother my mail server and syslog daemon forever, for
no good reason at all?

It's especially annoying to EXPN the sender's address at their
own server and find that it does indeed expand to "/dev/null".
I could tell them of course, but what's the point of us sending
them automatic notifications of invalid addresses if they are
just as automatically ignored?  I have tried talking to them,
and they simply ask: "Which address do you want removed?"  As
if MAILER-DAEMON hasn't told them a dozen times already...

In my opinion, failing to remove a user from a mailing list he
once subscribed to after several requests to be removed isn't
much different from outright spamming.  I once asked a company
to _replace_ my existing address in their mailing list with a
new one that I had set up explicitely for their mail.  They
acknowledged my request, and added the new address without
removing the old one, yielding two copies of their newsletter
in my mailbox.  Thank you. :-(

If the same error message is repeated for each message sent to
a bad recipient address over an extended period of time, the
error should be considered permanent and the error messages
equivalent to a removal request.  The sender cannot expect the
MAILER-DAEMON to follow their recommended unsubscribe method.
What is "an extended period of time" may depend on the volume
of mail sent via that mailing list.  If there is only a single
message per month, you can't expect a faster reaction than one
or possibly two months.

--
Anders Andersson, Dept. of Computer Systems, Uppsala University
Paper Mail: Box 325, S-751 05 UPPSALA, Sweden
Phone: +46 18 4713170   EMail: andersa@localhost





  • Post To The List:
<<< Chronological >>> Author    Subject <<< Threads >>>