Re: [dmarc-ietf] Do is need a new ptype? Was Re: New authentication method, DNSWL

["Murray S. Kucherawy" <[e-mail]>]

On Mon, Oct 21, 2019 at 7:54 AM Alessandro Vesely <[e-mail]> wrote:

> On Wed 07/Aug/2019 17:16:29 +0200 Murray S. Kucherawy wrote:
> >
> >> If the definition of ptype smtp were "a parameter of the SMTP session
> used
> >> to relay the message" it would be perfect.  I'd propose that
> policy.iprev
> >> be deprecated and smtp.remote-ip used instead>>
> >
> > Given that RFC8601 was published just last month, it'll probably be a
> while
> > before this happens.
>
> Wouldn't an accepted erratum be enough to change the wording in the IANA
> page?
>

That's not what the RFC Editor erratum system is for.  The document
reflects what the WG intended to publish at the time, so this isn't an
erratum, it's a new change to the specification.

About the new ptype, a reviewer suggested to also use it to report whether
> the
> query supported DNSSEC.  No DNSWL that I know supports it.  However, I know
> some DKIM filters report that feature either as a comment or as a reason
> in the
> dkim= methodspec.  Using the new ptype might make that clearer.  Consider:
>
>     Authentication-Results: example.com;
>       dkim=pass dns.sec=yes header.i=@example.org header.b=j5aQ3SJv
>

Are there any MTAs that would take a different action based on this
information?

-MSK