Benny Pedersen via Postfix-users wrote in <[email protected]>: |Noel Butler via Postfix-users skrev den 2023-11-05 04:53: |> On 03/11/2023 10:22, Steffen Nurpmeso via Postfix-users wrote: ... |> DKIM, was a problem, but for several years now mailman can do the |> right thing by stripping out the original DKIM headers and rewrites |> sender (although you need to know where to add the former) - but only |> if the admins set those settings (its more a mailman problem for not |> enabling them by default) | |fud as you self write above on spf, if mailman did not break dkim, |problem would be gone, but mailman breaks last 15+ years dkim, so all |kind of fixup is needed to allow dkim to pass later, i dont get it, it |worked on cloud9, no ?
Only to add that the ARC draft of that Google employee i mentioned, at least the version i read months ago, almost starts with the sentence (and one could read humour in between the lines) that "mailing-lists break DKIM", or similar. (And that is not the only IETF document which mentions this.) I personally think it has to be designed around the human, and not vice versa. Mailing-lists use footers and Subject modifications (and inject headers), so there (regarding my design optimum that is) can be only one mentally healthy solution. Also i never aligned to "dmarc is the problem" that also (the famous) Dave Crocker once mentioned on another list. I for one would love DKIM to be cryptographically verifieable down-top, from receiver to sender. At least in the end (that hyptothetically one that will not happen as the IETF will go the ARC way). If that approach is taken modifications do not matter in practice. --steffen | |Der Kragenbaer, The moon bear, |der holt sich munter he cheerfully and one by one |einen nach dem anderen runter wa.ks himself off |(By Robert Gernhardt) _______________________________________________ Postfix-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
