On 7/7/20 7:00 AM, Hector Santos wrote:
Not sure, Benny. I still haven't gotten the "Ah Ha" with ARC.
My issue with ARC is that, as I understand it, receiving systems have to choose to trust each ARC signer's assertion signature. As such, there is a massive scalability problem to trust random / individual server operators (like myself).
Thus, I'm disinclined to do much with ARC if my ARC assertion signature is going to be ignored because I'm not on a trusted list.
Another thing that just came to mind is the number of messages in a digest.What if its set to 1? if possible. In our MLM, the admin sets this amount, not the subscriber. But I have seen another MLM, forget which one, allowing the user to set the digest amount. How would a digest of 1, if allowed, be different from a normal non-digest submission distribution. I see it as the only form of a MLM output that would have not any "argument" or "conflict" with DKIM/DMARC/ARC concerns. The list digest agent can do what it wants without any debate.
Mailman has the ability to ""wrap DMARC protected messages as an RFC 5322 attachment to thee outgoing message. This would be the 1-to-1 relation that I think you are talking about.
Seeing as how the RFC 5322 messages are actually /attachments/ -or- synthetic text, there is little, if any, concern about DMARC alignment of the outer wrapper message.
IMHO, The outer wrapper message would ideally be DKIM signed and DMARC protected. ;-)
-- Grant. . . . unix || die
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ dmarc mailing list dmarc@ietf.org https://www.ietf.org/mailman/listinfo/dmarc
