Hi, On 19.05.2026 00:30, Tero Kivinen wrote:
Yes, at least I think there is work we can do to make ARC better, and I still find it useful. The main problem in the ARC is the C part of it. It was never and will never be a chain. Solution that requires every single node who processes email to be updated simply will not work in real world. This is the reason I do not belive DKIM2 will help at all.
I can concur that ARC has its practical use-cases.We are using it to forward mail between different clusters without any allowlisting. It does help in some odd cases coming from Microsoft or Google like you've described - they make a promise about the state of authentication (like SPF) at reception. Rescues more than a hundred thousand emails from DMARC rejection a week (mostly because people aren't using DKIM everywhere), probably more from quarantine.
I think the record so far has been four hops, ending with Microsoft breaking the chain. I suspect DKIM2 will encounter something similar with its chains. So it seems rather vital to really think through how it works when the chain is broken or re-established.
If DKIM2 does not provide a way for making such promises or their origin can't be determined, then DKIM2 won't be a useful replacement of ARC for us.
Labeling ARC "historic" before DKIM2 proves itself as a replacement would serve no good purpose.
Best, Taavi
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ dmarc mailing list -- [email protected] To unsubscribe send an email to [email protected]
