[Rspamd-Users] Fwd: [pfx] Re: rspamd bugs ?

[G.W. Haywood]

Hi there,

On Fri, 24 May 2024, Benny Pedersen wrote:
> G.W. Haywood skrev den 2024-05-24 20:43:
>> On Fri, 24 May 2024, Benny Pedersen wrote:
>>>> ...
>>>> Authentication-Results	list.sys4.de; dkim=pass header.d=junc.eu;
>>>> arc=none (Message is not ARC signed); dmarc=pass (Used From Domain
>>>> Record) header.from=junc.eu policy.dmarc=reject
>>> 
>>> where or why is reject in policy.dmarc here ?
>> 
>> I guess it's the sender's policy:
>> 
>> 8<---------------------------------------------------------------
>> $ dig +short -t txt _dmarc.junc.eu
>> "v=DMARC1; p=none; sp=reject; pct=100; 
>> rua=mailto:abuse at junc.eu,mailto:dmarc_agg at vali.email; 
>> ruf=mailto:abuse at junc.eu; ri=86400; aspf=r; adkim=r; fo=0:1:d:s;"
>> 8<---------------------------------------------------------------
>
> p=none
>
> sp=reject means reject all other then parent domain imho, no ?

Yes, assuming there's no DMARC record for the subdomain in question,
and let's be clear (for those just tuning in) that it's the sender's
policy for what recipients should do on failing a signature check for
that (sub)domain's mail.  If you're saying that something looks a bit
strange in that Authentication-Results header, then I'd agree.

-- 

73,
Ged.