We've been looking at implementing ARC in our MTA (Halon) and DKIM
library and wanted to share the following findings
A. If the DKIM selectors DNS record includes the flag t=s, the DKIM
standard says we should match the i= tag with the d= tag. This is of
course not possible for AMS since i= is reused for instance. This was an
exception we had make in our shared validator for AMS and DKIM. I think
it should be mentioned somewhere if it's not.
B. There is plausibly a typo in
draft-ietf-dmarc-arc-protocol-14#appendix-B.3.3 example where the
"Authentication-Results: i=3; gmail.com; ..." header should probably be
an AAR header since it begins with i=3; and is missing otherwise.
Regards
Erik
_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc
