On Tue, Apr 18, 2017 at 1:34 AM, Hugo Landau <[email protected]> wrote:
> > In reviewing a PR today noting that a client can find the account URI > for > > a key pair using a new-account request with an empty payload [1], > Jacob > > and I thought it might be a little more robust to use an explicit > signal. > > I've posted a PR that adds a "recovery" field to indicate to the > server > > that it should not create an account if one does not exist already. > Which > > is a little ironic in a request to an endpoint designed to create an > > account, but saves us creating a whole new endpoint. > > [1]https://github.com/ietf-wg-acme/acme/pull/296 > I don't think this should be called recovery. It's too reminiscent of > the old removed account recovery mechanisms and will be confusing. > > How about "existing": true? > Good idea. To be extra clear, I think I'm going to with "only-return-existing".
_______________________________________________ Acme mailing list [email protected] https://www.ietf.org/mailman/listinfo/acme
