Right now, the dns-01, http-01, and tls-sni-02 challenges require a
payload containing the keyAuthorization. However, there isn't any need
for this. For these challenges, the client only needs to convey "I am
ready to response to the validation requests." The server is capable of,
and should, calculate the keyAuthorization on its own.

Removing this would mean that triggering validation for one of these
three challenge types would consist of POSTing a signed, empty payload
to the challenge URL.

Thoughts on removing the keyAuthorization field from challenge POSTs? We
would of course still be using the keyAuthorization to produce the value
returned via HTTP, TLS, or DNS requests.

_______________________________________________
Acme mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/acme

Reply via email to