On Wed 30/Dec/2020 16:17:21 +0100 Michael Thomas wrote:
On 12/30/20 7:06 AM, Laura Atkins wrote:
The auth-res result posted as an example of DMARC failing earlier in this
thread:
Authentication-Results:mx.google.com;
[email protected] header.s=ietf1 header.b=aayvF8Pg;
[email protected] header.s=ietf1 header.b="PwU4/yuQ";
dkim=neutral (body hash did not verify)[email protected]
header.s=201712 header.b=PRr8Q7Zv;
spf=pass (google.com: domain [email protected] designates
4.31.198.44 as permitted sender)[email protected];
dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=mrochek.com
The policy statement is right there: p=NONE.
No. That is not guaranteed whatsoever. It could say "(eat at Joe's)" and be
valid.
Correct. However, it could have been:
dmarc=fail policy.dmarc=none header.from=mrochek.com
where the property meaning is defined as "Evaluated DMARC policy applied/to be
applied after policy options including pct: and sp: have been processed. Must
be none, quarantine, or reject."
That value appears on the IANA page[*] referring to RFC7489 for its definition.
However, Section 11.1 of RFC 7489 only mentions the "from" property.
IMHO, dmarc=quarantine is more direct than dmarc=fail policy.dmarc=quarantine.
In addition, policy.dmarc as a ptype.property pair sounds redundant. Perhaps
it should be policy.result, to emphasize that it has been computed after
checking pct= and sp= (to be added: np=) and the alignment of header.from with
respect to the dns zone where the record was found.
Best
Ale
--
[*] https://www.iana.org/assignments/email-auth/email-auth.xhtml
_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc
