Re: [exim] Bad addresses in forwards

Page principale
Supprimer ce message
Répondre à ce message
Auteur: Patrick von der Hagen
Date:  
À: exim-users
Sujet: Re: [exim] Bad addresses in forwards
On 08.01.2016 16:07, Jeremy Harris wrote:
> On 08/01/16 14:37, Robert Blayzor wrote:
>> Is there a way/best practice to verify the forwarding address before accepting the message being delivered? This would allow us to reject it at SMTP time rather than letting thousands of messages queue up for delivery and then ultimately bounce.
>
> recipient verify callout

I certainly agree that recipient verification via callout is a must.
However, since there are still issues like "you accept the content, the
address you are forwarding to rejects it, so you have to send bounces
and get a bad reputation", I'd try to do some more.

1. activate callouts
2. establish a policy how old forwards can be disabled or deleted, like
causing many bounces due to content based filters at the receiving
server, inactivity, age, recipient frequently over quota...
3. consider running stronger content filtering for forwards than for
locally delivered mail, e.g. a mail that would be locally delivered to a
junk folder might be rejected if the local address forwards instead
4. monitor logs for bounces caused by forwards and react. E.g. having
dozens of old accounts forwarding all the spam they receive to gmail is
likely to cause issues (see policy above).
5. Optional: consider using e.g. Mail::CheckUser to periodically check
all forwards. Mail::CheckUser checks syntax, DNS records and can perform
callouts. Since exim callouts will only trigger for local addresses that
actually receive emails, Mail::CheckUser can help you identify unused
but broken forwards.



--
Karlsruher Institut für Technologie (KIT)
Steinbuch Centre for Computing (SCC)

Patrick von der Hagen

Zirkel 2, Gebäude 20.21, Raum 004.2
76131 Karlsruhe
Telefon: +49 721 608-46433
E-Mail: hagen@???
Web: http://www.scc.kit.edu

KIT - Universität des Landes Baden-Württemberg und
nationales Forschungszentrum in der Helmholtz-Gemeinschaft