Re: Benjamin Kaduk's Yes on draft-ietf-httpbis-http2-tls13-02: (with COMMENT)

[David Benjamin <[e-mail]>]

On Wed, Oct 16, 2019 at 5:27 PM Benjamin Kaduk via Datatracker <
[e-mail]>; wrote:

> Benjamin Kaduk has entered the following ballot position for
> draft-ietf-httpbis-http2-tls13-02: Yes
>
> When responding, please keep the subject line intact and reply to all
> email addresses included in the To and CC lines. (Feel free to cut this
> introductory paragraph, however.)
>
>
> Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
> for more information about IESG DISCUSS and COMMENT positions.
>
>
> The document, along with other ballot positions, can be found here:
> https://datatracker.ietf.org/doc/draft-ietf-httpbis-http2-tls13/
>
>
>
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
>
> Thanks for this; I just have some minor nit-level comments; no response
> necessary.
>
> Abstract
>
>    This document updates HTTP/2 to prohibit TLS 1.3 post-handshake
>    authentication, as an analog to existing TLS 1.2 renegotiation
>    restriction.
>
> nit: either "restrictions" or "the existing".
>

Done in https://github.com/httpwg/http-extensions/pull/955. (I'll merge to
the repository and publish a -03 shortly.)


> Section 1
>
>    TLS 1.3 [RFC8446] updates TLS 1.2 to remove renegotiation in favor of
>    separate post-handshake authentication and key update mechanisms.
>    The former shares the same problems with multiplexed protocols, but
>    the prohibition in HTTP/2 only applies to TLS 1.2 renegotiation.
>
> nit: I'd suggest referring to a specific RFC rather than "HTTP/2" --
> this document will in some sense become part of "HTTP/2" upon
> publication :)
>

Done.


> Section 3
>
>    HTTP/2 servers MUST NOT send post-handshake TLS 1.3
>    CertificateRequest messages.  HTTP/2 clients MUST treat TLS 1.3 post-
>    handshake authentication as a connection error (see Section 5.4.1 of
>    [RFC7540]) of type PROTOCOL_ERROR.
>
> nit: is it the authentication or the request thereof that is the
> connection error?
>

Reworded to say the message is the authentication error.


> Section 4
>
>    Unless the use of a new type of TLS message depends on an interaction
>    with the application layer protocol, that TLS message can be sent
>    after the handshake completes.
>
> I don't see anything better to say than this, but ... will
> draft-ietf-tls-exported-authenticator be considered to "depend on an
> interaction with the application layer protocol"?
> (Also, nit: hyphenate "application-layer".)
>

(Hyphenation fixed.)

draft-ietf-tls-exported-authenticator doesn't define a post-handshake TLS
message, so it hopefully shouldn't apply?