Re: [OAUTH-WG] Review draft-ietf-oauth-spop-00

[John Bradley <[e-mail]>]

We did add a diagram but have not published that version yet due to the freeze on document updates prior to the IETF meeting this week.

in sec 4.1 we recommend using a 32 octet (256bit) random number that is base64url encoded to create a 42 octet string with the correct URLsafe character set.

With the S256 transform more bits don't add to the security.   

We limit the string size to 128 octets to limit the storage burden on the AS and to prevent the URI from becoming to long.  longer than 128 characters adds no real value for those transforms and is enough for a SHA512 hash if someone wants to define that transform at some point in the future.

If we allowed public key transforms 128 characters might be too small, but that would be an extension profile.

John B.

On Sep 2, 2014, at 9:57 PM, Eduardo Gueiros <[e-mail]> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
> 
> Review of draft-ietf-oauth-spop-00
> 
> Gentleman, here are some notes on the OAuth SPOP Draft.
> 
> 
> Review Summary
> 
> A few grammar errors, no major flaws, some suggestions that could
> possibly make some passages clearer. Some questions as well.
> Also, providing a flow diagram would help clarify some of the
> interactions.
> 
> Abstract
> 
>> The OAuth 2.0 public client utilizing authorization code grant is 
>> susceptible to the code interception attack.
> 
> Suggestion for clearer reading:
> OAuth 2.0 public clients utilizing Authorization Code Grant (RFC 6749
> - - 4.1) are susceptible to an interception attack.
> 
> Introduction
> 
> Suggestion for clearer reading:
>> Public clients in OAuth 2.0 [RFC6749] is susceptible to the "code" 
>> interception attack.  The "code" interception attack is an attack 
>> that a malicious client intercepts the "code" returned from the 
>> authorization endpoint and uses it to obtain the access token.
> 
> Public clients utilizing OAuth 2.0 are susceptible to authorization
> code interception attack. A malicious client intercepts the
> authorization code returned from the authorization endpoint and uses
> it to obtain the access token.
> 
> Suggestion for clearer reading:
>> This is especially true on some smartphone platform in which the
> "code" is
>> returned to a redirect URI with a custom scheme as there can be 
>> multiple apps that can register the same scheme.
> 
> This is especially true on smartphones applications where the
> authorization code can be returned
> through custom URL Schemes where the same scheme can be registered by
> multiple applications.
> 
> Insert article before ‘dynamic’
>> To mitigate this attack, this extension utilizes dynamically
>> created cryptographically random key called 'code verifier'.
> 
> To mitigate this attack, this extension utilizes a dynamically created
> cryptographically random key called 'code verifier'.
> 
> Missing commas for context
>> The code verifier is created for every authorization request and
>> its transformed value called code challenge is sent to the
>> authorization server to obtain the authorization code.
> 
> The code verifier is created for every authorization request and its
> transformed value, called code challenge, is sent to the authorization
> server to obtain the authorization code.
> 
> 2 Terminology
> 2.1
> Question
> Random String: What constitutes ‘big enough entropy’? Shouldn’t
> minimum length be specified (e.g. 32 characters)?
> 
> 3 Protocol
> 3.1
> Question
> This paragraph states that the client must, through some mechanism,
> verify if the server supports this specification. Shouldn’t this
> mechanism be part of OAuth and therefore have an specification
> document on how to implement and perform the aforementioned verification?
> 
> Suggestion: Change MUST to SHOULD
>> The client that wishes to use this specification MUST stop
>> proceeding if the server does not support this extension.
> 
> I think a client can use this mechanism to implement a more secure
> transaction, but by verifying it, the client can be configured to
> continue performing the regular authorization grant if it chooses so.
> Hence, SHOULD instead of MUST.
> 
> 
> 3.3
> Question
> Goes with question on 2.1, why less than 128 bytes?
> 
> Suggestion
>> string of length less than 128 bytes
> string of length no less than 128 bytes
> 
> 
> Eduardo Gueiros
> Software Developer | Jive.com
> Jive Communications, Inc | egueiros at jive.com
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG/MacGPG2 v2.0.22 (Darwin)
> Comment: GPGTools - https://gpgtools.org
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
> 
> iQEcBAEBCgAGBQJUBsnrAAoJEInLDRowktwYn6sH+wUGGCimGSRaSfy4LeN9ug9e
> VN5R/N4eWhEKl5iydMZskeMovYsH4PhEP19m56mWGhRn53CxMB5dlHkTw56JX4mS
> qnPu96Ot6HoCzzCVj8PtGyAZUwWFyd57Ff7uUuSQaVghhIfLXzggFfciiErJpV5H
> wwQo+eFp98fx6uGEeCr4Olr6PsJ8Ajn5SnaCA/dPr7YWAUrnpSb55NCB4twtp4y6
> 36EqjcuvafudTEuYJOoGqYJppfpcZ+Da6uKZNRTahkN1ivv1C+PdNRWkfHbB47mu
> IF4u3b6tDhiymPNFjCABZ6gB5ZbXmUbVrkhVKwJpf/87/fiaScGmZG+6rRZLP5s=
> =XQSw
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> OAuth mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/oauth