Re: [OAUTH-WG] Using OAuth error_code to glean information from the server

[William Mills <[e-mail]>]

I don't think the leak of client IDs is a big issue, in any cleartext protocol you can just sniff them.  We don't mandate SSL on the protected resources.


Anyone relying on ClientID for a stronger assurance will want to be using SSL and an unguessable ID anyway.



________________________________
 From: Mike Jones <[e-mail]>
To: "[e-mail]" <[e-mail]> 
Sent: Monday, December 12, 2011 4:51 PM
Subject: [OAUTH-WG] Using OAuth error_code to glean information from the server
 

 
I recently received an inquiry regarding invalid_client vs. invalid_grant.  It seems that there is a potential information disclosure in the specification with respect to how these error codes are used:
 
invalid_client
               Client authentication failed (e.g. unknown client, no
               client authentication included, or unsupported
               authentication method).  The authorization server MAY
               return an HTTP 401 (Unauthorized) status code to indicate
               which HTTP authentication schemes are supported.  If the
               client attempted to authenticate via the "Authorization"
               request header field, the authorization server MUST
               respond with an HTTP 401 (Unauthorized) status code, and
               include the "WWW-Authenticate" response header field
               matching the authentication scheme used by the client.
invalid_grant
               The provided authorization grant (e.g. authorization
               code, resource owner credentials, client credentials) is
               invalid, expired, revoked, does not match the redirection
               URI used in the authorization request, or was issued to
               another client.
 
If one uses invalid_client when the client is unknown and invalid_grant when the client credentials are invalid, then an attacker could deduce whether or not a particular client exists.
 
First, do people agree that this is a potential information leak and that the leak is meaningful?  If so, what mitigation might be suggested?  For instance, might a server choose to use a single error code for both (and potentially other) cases?
 
                                                            Thanks,
                                                            -- Mike
 
_______________________________________________
OAuth mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/oauth