[Acme] Proactive vs On-Finalization Certificate Issuance

[Daniel McCarney <[e-mail]>]

Hi folks,

I wanted to revisit a discussion on the server proactively issuing
certificates. Section 4, "Protocol Overview"[0] currently has a paragraph
that
reads:

> Once the validation process is complete and the server is satisfied that
the
> client has met its requirements, the server can either proactively issue
the
> requested certificate or wait for the client to request that the
application be
> “finalized”, at which point the certificate will be issued and provided
to the
> client.

There is also an "Open Issue" remark related to proactive issuance in
Section
6.1.3[1].

I think the specification should decide whether proactive issuance or
on-finalization issuance should be used instead of allowing both with no
indication to the client which will happen.

It's the preference of Let's Encrypt that the specification only support
on-finalization issuance. This is cheaper to implement and seems to be the
path
of least surprise. It also seems to better support large volume issuance
when
the client may want explicit control over when the certificates for multiple
applications are issued.

In the earlier "Re: [Acme] Preconditions" thread[2] Andrew Ayer seems to
agree
that there should be only one issuance method to simplify client behaviours,
though he favours proactive issuance instead of on-finalization issuance.
While
it saves a round-trip message I'm not sure that this alone is convincing
enough
to choose proactive issuance as the one issuance method.

What are the thoughts of other list members?

- Daniel/cpu

[0]
https://github.com/ietf-wg-acme/acme/blob/3502ff0bfb6d434b4326e206ea7cae7b8434ac7d/draft-ietf-acme-acme.md#protocol-overview

[1]
https://github.com/ietf-wg-acme/acme/blob/3502ff0bfb6d434b4326e206ea7cae7b8434ac7d/draft-ietf-acme-acme.md#application-objects

[2] https://www.ietf.org/mail-archive/web/acme/current/msg01160.html