begin Wednesday 21 January 2004 21:37, Za'mbori, Zolta'n quote:
Alain Knaff wrote:
begin Wednesday 21 January 2004 21:21, Za'mbori, Zolta'n quote:
I think it is detectable by the SMTP client that a long return path was
accepted by the receiver MTA or not. After the rejection, a new MAIL
command can be sended with a short return path.
But what happens if the receiving MTA needs to forward its message to
yet another MTA (maybe the receiving MTA was the secondary MX and
needs to forward it to the primary).
In that case, the next MTA could conceivably refuse the message, and
the forwarder would be none the wiser.
I dont understand Your problem exactly. As I understand SPF and SRS the
"receiving MTA" will rewrite the return path before forwarding to
"another MTA".
No. Only the forwarding MTA rewrites envelope From. Relaying MTAs
don't. Forwarding is where the Envelope To gets rewritten
user(_at_)alumniserver(_dot_)com becomes user(_at_)myisp(_dot_)com(_dot_)
Relaying is where
Envelope To gets not rewritten.
(Secondary MX transfer to the primary MX is an other case because
primary MX can trust in the secondary MX.)
In theory, this is the case. In practice however, there may be many
reasons why Secondary and "primary" may behave differently.
Just let's consider this hypothetical example:
1. Company uses some proprietary MTA "I" internally (Lotus Notes,
Exchange, whatever)
2. Company doesn't want to expose that proprietary MTA to the net, so
it puts a sendmail (or postfix, or whatever) "X" in front
3. Every incoming mail first hits the sendmail MTA. MTA is forgiving
about overlong addresses, so they are accepted. Forwarder "F" is happy
because mail was accepted.
4. Next step, the mail is relayed to the internal MTA. Internal MTA is
picky about the From. We have a bounce. Your mechanism of "After the
rejection, a new MAIL command can be sended with a short return path"
won't work because in this example, the rejection does not happen on
the transfer from "F" to "X", but only later on ("X" to "I"): we have
a bounce!
Note: X only does relaying, not forwarding, thus cannot reasonably be
expected to know about SRS.
Alain
-------
Sender Permitted From: http://spf.pobox.com/
Archives at http://archives.listbox.com/spf-discuss/current/
Latest draft at http://spf.pobox.com/draft-mengwong-spf-02.9.4.txt
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to
http://v2.listbox.com/member/?listname(_at_)©#«Mo\¯HÝÜîU;±€Ö€Íµø?¡