Random thought:
selective rejections after DATA. Julian even found a draft trying
to support this - after it was implemented worldwide, so this
probably won't pass any "FUSSP" giggle test
Actually not funny, it's a potential hole in the SMTP HEAD draft:
IIRC the HEAD idea was to split DATA into HEAD and BODY, roughly
derived from the TOP in POP3. If the header doesn't pass muster,
there's no point in sending the complete data (i.e. the body).
But that's still after the RCPT TO. For a selective reject we'd
need the order EHLO - MAIL FROM - HEAD - RCPT TO - BODY - QUIT:
S: ready to receive fresh spam
C: EHLO some.thing.example
S: 2xx HELP HEAD 8BITMIME
C: MAIL FROM:<some(_at_)body(_dot_)example>
S: 2xx
C: HEAD
S: 3xx - send dot to finish header
C: foo: bar
...
C: .
S: 2xx - noted header, now what?
C: RCPT TO:<userA(_at_)example>
S: 2xx - userA loves spam
C: RCPT TO:<userB(_at_)example>
S: 5xx - forget it, not for userB
...
C: BODY
S: 3xx - send dot to finish spam
C: spam: spam spam
...
C: .
S: 2xx - received spam for userA etc.
...
C: QUIT
S: 2xx - a pleasure to talk to spammers
Frank
-------
Sender Policy Framework: http://www.openspf.org/
Archives at http://archives.listbox.com/spf-discuss/current/
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to http://v2.listbox.com/member/?list_id=735