AccuSpam wrote:
Seems to me that your implication that someone would need to
publish different mail servers depending on whether the
Return-Path or PRA header is being evaluated is not
mainstream.
It is. PRA does not work with legacy MUAs and MSAs, where the
sending site doesn't insert a "Sender" matching the MAIL FROM
(to overrule a 2822-From not matching the MAIL FROM).
In other words I daily send mails with "my" 2822-From, which
would FAIL in a PRA-Test, because "my" sender policy has scope
MAIL FROM and not PRA.
none of the scoped algorithms give a 100% final result
If "my" MAIL FROM results in a FAIL, then this is 100% final,
and if it results in a PASS then that's also 100% final. And
"my" 2822-From results in a 100% final NONE when it happens to
be the PRA.
Bye, Frank