[opendmarc-users] Does my opendmarc 1.3.2 parse domains correctly?

[Dominic Raferd]

On 30 March 2017 at 15:41, Juri Haberland <juri at sapienti-sat.org> wrote:

> Dominic Raferd wrote:
>
> > Thanks Juri. It's not so much the absence of sp= (although that would be
> > nice to have appearing as well) it's that google shows 'header.from=
> emv5.com'
> > which is actually the domain that was/should_have_been tested (because
> > there is no DMARC TXT record for skimium.emv5.com) whereas openDMARC
> shows '
> >
> > header.from=skimium.emv5.com' so you don't know that openDMARC
> *actually*
> > looked at the DMARC TXT record for emv5.com (as it did, if I understand
> you
> > correctly).
>
> Ok, but still I think Google is wrong here and OpenDMARC is right - at
> least,
> if my interpretation of the RFC is right. There is nothing written, that
> the
> actually tested domain should be reported in the header.from field.
>
> What OpenDMARC actually tested should be recorded in the history file,
> which
> will be imported into the database.


I defer to your greater knowledge here. I don't use the history file. My
main concern was that openDMARC was testing the right thing, which you have
confirmed it is. In fact I haven't seen any cases where its result differs
from Google's.​

Thanks for your attention and if the sp= info can be added to the AR
header, so much the better. Maybe an *additional* field could be added to
the AR header only in cases where the appropriate valid DMARC record used
was found for a domain different to the 'from' header domain - in this
case, for example: tested_domain=emv5.com.

Dominic
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.trusteddomain.org/pipermail/opendmarc-users/attachments/20170330/e907eb01/attachment.htm>