> -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of Scott > Kitterman > Sent: Thursday, January 05, 2012 2:41 PM > To: [email protected] > Subject: Re: [marf] I-D Action: draft-ietf-marf-authfailure-report-09.txt > > > spf: The evaluation of the author domain's SPF record produced > > something other than a "pass" result, which the report generator > > considers to be a reportable incident. This can include the usual set > > of failure results, or any result that is considered a failure under local > > policy constraints. > > I think it's a mistake to report non-SPF results as SPF result. This > includes policy overrides or the results of some hypothetical SPF > extension.
I'm specifically trying to enable your "none" case without calling it out, because I don't want to have to add another non-pass code later. "pass" is really the only case that isn't reportable, so the above language seems to cover all the possibilities without enumerating them. > That the message is intended to communicate some kind of failure is > implicit in the message type. The data element we are discussing is > for SPF results and should only include those (and it should include > all of them). The data element we're talking about is the Auth-Failure field, which is only used to indicate which module reported some kind of result that is reportable. The specific result reported will be carried in the Authentication-Results: field. _______________________________________________ marf mailing list [email protected] https://www.ietf.org/mailman/listinfo/marf
