> -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of Scott > Kitterman > Sent: Tuesday, January 24, 2012 2:02 PM > To: [email protected] > Subject: Re: [marf] I-D Action: draft-ietf-marf-spf-reporting-03.txt > > On Tuesday, January 24, 2012 09:28:49 PM John Levine wrote: > > Nits: in Sec 3, the description of r= still refers to rf=. > > Thanks. Fixed locally. > > > In 6.3, how come it says that a bounce address MUST pass SPF but a > > HELO only SHOULD pass? (I'm not necessarily arguing, I'm asking.) > > > > Other than that, looks fine. > > Sending the report from an address the passes SPF, modulo bugs and DNS > errors, solves the potential looping problem. If you're using NULL Mail > From and the only identity you have for an SPF check is EHLO/HELO then > there's still no loop problem, although you might generate an extra > report if there's a rejection due to EHLO/HELO not passing SPF.
The text in -03 says: "Per Section 2 of [DSN], the NULL envelope sender address of the report MAY be chosen to ensure that no delivery status reports will be issued in response to the report itself." That's not consistent with Section 2 of [DSN], which says SHOULD and MUST. We should either drop the reference, or be consistent with it. If we drop it, we need to be prepared to explain why we're contradicting [DSN], which is a draft standard. Also: 1) 6.3 has a missing "<" before an xref, so the XML is revealed in the produced document. 2) I've changed "r=" to "ra=" so that all of the reporting tags are distinguishable as "r[a-z]". Seem reasonable for yours as well? -MSK _______________________________________________ marf mailing list [email protected] https://www.ietf.org/mailman/listinfo/marf
