On Thu, Feb 21, 2013 at 6:53 AM, Cyborg <cyborg2@???> wrote: > This seems to be a design issue :
It's not an exim design issue. It is a design issue of the SMTP
protocol. Once you get to the data phase of an smtp email delivery,
you are only allowed to issue one answer for all intended recipients:
accept, reject, or defer.
> Too inform the sender, that his attempt did not work, we have to reject it
> on SMTP level, means inside acl_check_data or a similar capable acl.
> Otherwise, we have to create a new delivery message and inform the sender
> about the "later" failure inside a router. Which will not work with spams as
> of the nature of most faked sender addresses.
Correct, it's called backscatter.
> a new acl "acl_per_recipient" could solve such problems on an elegant way.
An experimental proposed SMTP protocol called PRDR (Per Recipient Data
Responses) from several years back is exactly what you describe. It
just so happens that Jeremy Harris and I coded up some experimental
PRDR support and have merged it into the master development branch
(which leads to the next release of Exim, whenever that turns out to
be).
It is sparsely tested, but I am running the code on a production
server and it has not blown up yet, though I'm not really using the
PRDR acl for anything that the DATA acl doesn't already do.
If you want to test it, you can checkout master, build, install, and
run it. If you happen to run exim on a CentOS 5.x machine that is not
plesk or cpanel, I can put up for download on my personal website the
rpms that I built and use in production. My build also has
experimental DMARC enabled, which means that libopendmarc must be
installed, which I also have available for download from my personal
website.
...Todd
--
The total budget at all receivers for solving senders' problems is $0.
If you want them to accept your mail and manage it the way you want,
send it the way the spec says to. --John Levine
