It appears that MCN-IETF <[email protected]> said: >A (i=1) -> B (i=2) -> C > | > | discard existing chain > v > C (i=1) -> D/Receiver > >Have I understood correctly that C is able to do this, and that D has no >protocol-level way to determine that the message previously had a valid DKIM2 >chain? > >In other words, would the result at D be indistinguishable from a message >whose DKIM2 history genuinely started at C?
Pretty much. If the From: header address is still A's or B's, the d= on C's signature won't match that domain and we expect i=1 messages where they don't match to have a poor reputation. On the other hand, if C changes the address to one in C's domain so the signature does match, that is for all purposes a new message from C. R's, John _______________________________________________ Ietf-dkim mailing list -- [email protected] To unsubscribe send an email to [email protected]
