Alex van den Bogaerdt wrote:
Tony is under the impression that you (and I, and the rest
of the world) know when an email we send is going to be
forwarded.
Not my impression, he just wants to do nothing. And if the
next hop rejects an SPF FAIL he has to bounce it back to
the SPF-FAIL user. Almost the same effect as if he would
reject it directly with a "551 user not local" from my POV,
because I use the smart host of my provider - or in other
words I'd get a bounce in both cases.
I also don't see how "551" can help here, perhaps you mean
"251" ?
261 = forwarder tries it, next hop might reject => bounce
551 = forwarder rejects it, my smart host bounces it to me
Almost the same effect from my POV. IIRC Scott said that
he had some of these 551-effects in the last year. I had
exactly one case so far in about fourteen months. It was
simple, the bounce contained the new address, I just sent
it again directly (bypassing the forwarder), no problem,
It would be less simple when the bounce doesn't show the
new address. In that case I'd be forced to use another
MAIL FROM (not protected by a FAIL), but fortunately so
far I didn't need this plan B.
I'm going to let this topic rest in peace for a while.
ACK, no new insights. I liked the case where the final
hop is a C/R system, that could have really odd effects
for SRS ;-)
Bye, Frank