"Harry Katz" <hkatz(_at_)exchange(_dot_)microsoft(_dot_)com> wrote:
Suppose alice(_at_)mydomain(_dot_)example(_dot_)com sends mail to
bob(_at_)alumni(_dot_)almamater(_dot_)edu(_dot_) But bob has set up
forwarding of
bob(_at_)alumni(_dot_)almamater(_dot_)edu to bob(_at_)company(_dot_)com
While that's bob's perogative, it's entirely independent of Alice's
knowledge or intent.
On the second hop from alumni.almamater.edu to company.com, the message
still has MAIL FROM: alice(_at_)mydomain(_dot_)example(_dot_)com because
alumni.almamater.edu hasn't implemented SRS, say.
I would disagree from a philosophical perspective. It isn't a
"second hop" for the message. The "alumni.almamater.edu" site has
terminated an SMTP connection, and based on local site policy, decided
to send an SMTP message. The relationship between the two messages is
one of message body, not SMTP identity.
e.g. How is this case different from the situation where bob is
sitting at a desk at almamater.edu, and decides to forward the mail by
hand to bob(_at_)company(_dot_)com? In that case, the MAIL FROM may be
bob(_at_)alumni(_dot_)almamater(_dot_)edu(_dot_) Are there strong reasons why
that MAIL FROM
is different than the MAIL FROM for the automated forwarding case?
I understand that it's difficult for the automated forwarder to
manage bounces when it doesn't have a local mailbox to hold bounces.
But I would view that problem as one due to implementation, or
deployment. It's awkward to force the protocol to work around
difficulties with implementations.
To put it another way (yet again), look at it as a nested tunnel
issue.
alice(_at_)mydomain(_dot_)example(_dot_)com sends mail to
bob(_at_)alumni(_dot_)almamater(_dot_)edu
bob(_at_)alumni(_dot_)almamater(_dot_)edu sends mail to
bob(_at_)company(_dot_)com,
and tunnels alice's SMTP connection.
Why would the alumni.almamater.edu site, outside of the tunnel,
use an identity from inside of the tunnel?
Alan DeKok.