| On 06/03/2013 17:06, Chris Siebenmann wrote:
| > PS: we don't do this because we do accept-then-bounce; we do this
| > because some of our users forward *all* of their mail, spam
| > included, off to outside places which then reject the spam at
| > SMTP time when we try to deliver it to them. We want to eat the
| > bounces for such rejected forwarded spam.
|
| Ahh, you want cutthrough delivery. Feature due in next release;
| present in git now if you're into compiling your own.
I took a look at the docs in the Exim git and unfortunately I
suspect that we won't really be able to use cutthrough delivery.
Partly this is because of the single recipient limitation and
partly this is because of our existing mail architecture with a
separate external MX and central mail server (we'd have to use
cutthrough on both, and there's bonus oddness in the mix).
I also wish that cutthrough delivery used a separate router flag
besides verify/no_verify. In our current router architecture we have a
very strong split between routers that simply validate local addresses
(which are basically verify-only) and routers that go to the work of
actually expanding and delivering them. I am not so keen to expose the
latter to anyone who wants to do VRFYs.
(I am not even sure that all of our real routers *work* if you do VRFY
without the message headers and body available.)
- cks
