[Mailman-Users] List posts sent to AOL list subscribers bounce as undeliverable (v. 2.1.20)

Stephen J. Turnbull stephen at xemacs.org
Sat Dec 5 09:34:57 EST 2015

Mark Sapiro writes:

 > >    This is a permanent bounce due to:
 > > 
 > >        * RFC2822 From domain does not match the rDNS of sending server.

 > The first reason above is really bad. It's looks like AOL's own
 > extension to DMARC that ignores any DMARC policy that might actually be
 > published by the From: domain.

AOL fired most (all?) of their competent people a couple of years ago.
The ones who are left probably don't have time to write super-precise
documentation.  I just read that one as a reject of a message without
a valid DKIM signature that fails the SPF test too, according to the
DMARC policy of the From domain.  True, it could be as bad as you say
-- this is AOL we're talking about -- but I think it's probably just
hard-to-understand docs.

 > On the other hand, I don't expect AOL to properly document what
 > they are doing anyway. You could try setting from_is_list on your
 > list's General Options page to Munge From or Wrap Message.

Note that Wrap Message is the theoretically correct answer, but it
causes pain to at least iPhone users (as of last year).  I can't
imagine it works better than Munge From as a DMARC mitigation
(except for Emacs/Gnus and mutt users ;-).  I wouldn't bother trying
it if I were Woody, and much as it pains me to say it (it was my
proposal, after all) I don't think we should recommend it to anybody
who can't figure out what it does for themselves. :-(

Also, if this is happening to digests (as Woody says in a later post),
that seems weird to me.  Digests should be sent From:
mailman at sending-host.tld, and so should be DMARC-safe, right?  Of
course the sending host may not be DMARC-ly correct, but if so I would
think they would suffer greatly in many ways.

So I tend to think it's the IP reputation that's the main problem, but
what AOL considers in computing reputation I don't know.


More information about the Mailman-Users mailing list