Re: [OAUTH-WG] Second AD Review: draft-ietf-oauth-mtls

[Brian Campbell <[e-mail]>]

Thanks for the additional review, Roman. I feel lucky, it's not often one
gets *two* AD reviews :)  Please see below for replies inline with a few
followup questions.



On Sat, Jun 22, 2019 at 12:29 PM Roman Danyliw <[e-mail]> wrote:

> Hi!
>
> I conducted as second AD review of draft-ietf-oauth-mtls per the AD
> hand-off.  I have the following additional feedback:
>
> ** Per ekr's earlier review at
> https://mozphab-ietf.devsvcdev.mozaws.net/D3657, paraphrasing:
> -- Section 2.1.2, How is these metadata parameters being obtained?
>

The authorization server can obtain client metadata via the Dynamic Client
Registration Protocol [RFC7591], which is referenced in the top of that
section. Also the metadata defined by RFC7591, and registered extensions to
it, implies a general data model for clients that is used by most
authorization server implementations even when the Dynamic Client
Registration Protocol isn't in play. Such implementations typically have
some sort of  user interface available for managing client configuration.

Dose that answer your question? Do you believe more should be said in the
document to better explain or clarify that?


-- Section 3.2, Figure 3.  In this example, what new information is the
> auth server providing to the relying party here?
>

The new info here (and in Section 3.1 too) is the hash of the client
certificate to which the access token is bound, which is in the "cnf"
confirmation method at the bottom as the "x5t#S256" member.



>
> ** Section 2.0.  What is the expected behavior if the presented
> certificate doesn't match expected client_id?  How is this signaled?
>

With a normal OAuth 2.0 error response using the "invalid_client" error
code per https://tools.ietf.org/html/rfc6749#section-5.2

Do you think that needs to be stated more explicitly in this document?



>
> ** Section 2.2.  Per the sentence "As pre-requisite, the client registers
> its X.509 certificate ... or a trusted source for its X.509 certificates
> ... with the authorization server.
> -- Editorial: s/As pre-requisite/As a prerequisite/
>

done


> -- What's a "trusted source" in this case?  Is that just a jwks_uri?  If
> so, maybe s/a trusted source/a reference to a trust source/.  If not, can
> you please elaborate.
>

Yes, it's just a jwks_uri. I'll change that.


>
> A few editorial nits:
> ** Section 2.2.2.  Typo.  s/sec 4.7/Section 4.7/
>

fixed


>
> ** Section 3.1  Cite DER encoding as:
>     [X690]     ITU-T, "Information Technology -- ASN.1 encoding rules:
>               Specification of Basic Encoding Rules (BER), Canonical
>               Encoding Rules (CER) and Distinguished Encoding Rules
>               (DER)", ITU-T Recommendation X.690, 2015.
>

will do


> ** Section 5.  Typo. s/metatdata/metadata/
>

yup


> ** Section 6.  Typo.  s/The the/The/
>

got it


>
> ** Section 7.2. Typo.  s/the the/the/
>

done


>
> ** Appendix. Cite the figures numbers (#5 - 7) in the text describing the
> contents of the section.
>

will do


>
> The shepherd write-up is in good shape.  Thank you.
>
> Regards,
> Roman
>
> _______________________________________________
> OAuth mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/oauth
>

-- 
_CONFIDENTIALITY NOTICE: This email may contain confidential and privileged 
material for the sole use of the intended recipient(s). Any review, use, 
distribution or disclosure by others is strictly prohibited.  If you have 
received this communication in error, please notify the sender immediately 
by e-mail and delete the message and any file attachments from your 
computer. Thank you._