-----Original Message-----
From: Scott Nelson [mailto:scott(_at_)spamwolf(_dot_)com]
At 06:24 PM 6/6/03 -1000, Peter Kay wrote:
I agree, a challenge can be seen as a DSN or a new message.
My point is
that in either case, the MAIL FROM or REPLYTO should be
consistent with
the challenger's email address, otherwise 2 C/R systems will
challenge
each other's challenge.
If you're sending a DSN, then the From: header could be your
address, but the MAIL FROM should probably be <>. That
prevents a lot of bad things from happening.
I think DSN in the CR space is a matter of semantics and definition. I'm
not 100% whether a challenge email is a DSN or not, and like I said
earlier I can see a case for it ether way.
If you're saying that if it's a DSN, then the MAIL FROM s/b <>, then
I'll say that challenge emails are not DSNs.
Setting the MAIL FROM to your address means that you could
end up challenging a non-existent address, then the
mailer_daemon that bounces the mail back to you. If the C/R
system was really broken, it could challenge bounces
from a non-existent address endlessly, but in the normal case
this is only twice as much work as you need to do.
The MAIL FROM needs to be the address of the challenge sender in order
to allow CRI systems to interoperate. While some CR systems may inspect
the header, other systems that bounce the email outright (like ours)
need to look at the MAIL FROM, for reasons I've mentioned earlier in
this thread.
You are correct about that fact that you "could end up challenging a
non-existent address". Most spammers use bad addresses. Our system
handles mail that is bounced because of non-existent email and this also
results in reduced load on the MTS due to more intelligent mailer_daemon
handling.
Peter
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg