On Sat, Aug 06, 2016 at 11:55:26AM -0700, Jacob Hoffman-Andrews wrote:
> At IETF 96 it was proposed to drop this issue:
> https://www.ietf.org/proceedings/96/minutes/minutes-96-acme.
> 
> The rationale from the notes is that nonces are not a scarce resource.
> However, cachability and idempotence of GETs were not addressed. I think
> it's worth not requiring nonces on GETs purely for those reasons. In
> practical terms, this difference has caused real bugs for Let's Encrypt.
> 
> Would someone like to present a specific defense of providing a unique
> nonce with every GET?

The example that I and others originally voiced on this list was that you
shouldn't have to do a POST that you know will fail (because you don't
yet have a valid nonce) to get a valid nonce.

But as long as there is some trivial way to GET a fresh nonce (similar
to what we are already doing now with a HEAD request to the directory)
I don't see a problem with not providing a nonce with _every_ GETable
resource.

IIRC you objected to using the directory for this, but a separate "get
me a new nonce" resource would probably be fine for that too.  modulo
needing a flag day for the client code to transition to that if Boulder
doesn't implement both for some period of time.

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

Reply via email to