Re: [Exim] Some more broken ACL's

Pàgina inicial
Delete this message
Reply to this message
Autor: Peter Bowyer
Data:  
A: Exim users list
Assumpte: Re: [Exim] Some more broken ACL's
Alan J. Flavell <a.flavell@???> wrote:
> On Mon, 19 Apr 2004, Peter Bowyer wrote:
>
>> I guess the secondary isn't under your control - if you feel you
>> really need a secondary, maybe it's time to rent a tiny corner of
>> the internet to run it on.
>
> You have a point, sure.
>
> Where the same techique can be really handy, however, is where users
> are getting mail forwarded from some other account elsewhere, e.g on
> alumni servers or from other research establishments where they work -
> those servers really aren't under our control, but just the same
> technique can be used for detecting spam "leakage" through them. It
> works well for us (admittedly, our rejection then probably provokes
> the would-be forwarding host into producing a "collateral bounce" for
> every faked item that we reject, but I'm afraid that's on their
> responsibility, not mine).


Agreed - I run such a service for a special interest group, and interact as
a client of some, too. I don't want to be obliged to relax my controls just
because the upstream forwarding service isn't as smart (IMHO) as I am, but
I'm a neutral carrier for some stuff and I don't want to be dropping peoples
mail either.

I guess I do something which is functionally similar - SpamAssassin rules
which crank up the spam score with RBL checks on the Received: headers. It
would take a lot of these plus something very spammy indeed to make a
REJECT, though.

> I really DO NOT want to start accepting mail that I don't want, and
> quietly dropping it into a black hole. That's no way to run a reliable
> mail service.


Completely agree.

Peter