That sounds closer.  I think Todd knows what he needs to change, so I'd prefer 
to see what goes into the document than to continue to tweak it.

Scott K

On Thursday, January 20, 2022 12:18:03 PM EST Douglas Foster wrote:
> So if the first tree walk stops at a a PSD=y policy, then the match string
> used for alignment is the organizational domain, one segment down from the
> PSD policy.  Any SPF or DKIM domain must match or be a child of the
> organizational domain, so there is no secondary tree walk,
> 
> Does that correct the problem?
> 
> 
> 
> On Thu, Jan 20, 2022 at 11:58 AM Scott Kitterman <[email protected]>
> 
> wrote:
> > On Thursday, January 20, 2022 9:57:48 AM EST Douglas Foster wrote:
> > ...
> > 
> > > -- If a policy is found with PSD=y, the domain does not participate in
> > > DMARC but may need to be tested for non-existence.  If the policy also
> > > specifies NP=reject, query the next-lower domain name for a resource
> > > record.   If the DNS query result is NXDOMAIN, processing stops and the
> > > DMARC policy is also "NXDOMAIN".  (I recommend using NXDOMAIN as a
> > separate
> > > result code from REJECT, as it seems to be a stronger repudiation.)
> > 
> > ...
> > 
> > No.  In this case policy discovery is complete and that's the policy that
> > should be applied.  Additionally, that's not how the np= tag works.
> > 
> > All psd=y means is don't use this domain for determining alignment.  For
> > policy discovery it's like any other.
> > 
> > Scott K
> > 
> > 
> > _______________________________________________
> > dmarc mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/dmarc




_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc

Reply via email to