Hi Andrii,

> Am 07.02.2021 um 21:30 schrieb Andrii Deinega <andrii.dein...@gmail.com>:
> 
> 
> Hi Torsten,
> 
> thank you for your response.
> 
> My use case is pretty straight forward
> 
> An OAuth client queries the AS to determine the active state of an access 
> token and gets the introspection response which indicates that this access 
> token is active (using RFC7662).
> 
> An OAuth client queries the AS to determine the active state of a refresh 
> token and gets the introspection response which indicates that this refresh 
> token is active (using RFC7662).
> 
> An OAuth client queries the AS to determine the active state of an access 
> token and gets the introspection response (JWT) which indicates that this 
> access token is active (using this draft).
> 
> Now, an OAuth client queries the AS to determine the active state of a 
> refresh token (using this draft)... How will the introspection response look 
> like assuming that the client provides the valid refresh token and 
> technically, nothing stops it from doing so.

why should the state be provided as JWT?I think the plain JSON response is 
sufficient in that case.  I also think using token introspection for checking 
the state of a token from the client side has limited utility. The definitive 
decision is always made when the client tries to access a resource. 

I‘m therefore leaning towards explicitly stating in our draft that it is not 
intended to be used with refresh tokens.

best regards,
Torsten.

> 
> Regards,
> Andrii
> 
> 
>> On Sun, Feb 7, 2021 at 4:14 AM Torsten Lodderstedt <tors...@lodderstedt.net> 
>> wrote:
>> Hi Andrii,
>> 
>> thanks for your post. 
>> 
>> The draft is intended to provide AS and RS with a solution to exchange 
>> signed (and optionally encrypted) token introspection responses in order to 
>> provide stronger assurance among those parties. This is important in use 
>> cases where the RS acts upon the introspection response data and wants the 
>> AS to take liability re the data quality. 
>> 
>> I’m not sure whether there are similar use cases if a client introspects a 
>> refresh token. What is your use case?
>> 
>> best regards,
>> Torsten.  
>> 
>> > Am 07.02.2021 um 08:41 schrieb Andrii Deinega <andrii.dein...@gmail.com>:
>> > 
>> > Hi WG,
>> > 
>> > draft-ietf-oauth-jwt-introspection-response-10 states that "OAuth 2.0 
>> > Token Introspection [RFC7662] specifies a method for a protected resource 
>> > to query an OAuth 2.0 authorization server to determine the state of an 
>> > access token and obtain data associated with the access token." which is 
>> > true. Although, according to RFC7662, the introspection endpoint allows to 
>> > introspect a refresh token as well. Hence, the question I have is how will 
>> > a token introspection response look like when the caller provides a 
>> > refresh token and sets the "Accept" HTTP header to 
>> > "application/token-introspection+jwt"?
>> > 
>> > I expect there will be no differences, right?
>> > 
>> > If so, I suggest to
>> >       • replace "a resource server" by "the caller" in section 4 
>> > (Requesting a JWT Response)
>> >       • change "If the access token is invalid, expired, revoked" by "If a 
>> > given token is invalid, expired, revoked" in section 5 (JWT Response)
>> > If not, my suggestion would be to clarify what the AS should do when it 
>> > asked to introspect the refresh token in general and additionally, what 
>> > should happen in the same case based on the type of the caller from the 
>> > AS's point of view.
>> > 
>> > Regards,
>> > Andrii
>> > 
>> 

Attachment: smime.p7s
Description: S/MIME cryptographic signature

_______________________________________________
OAuth mailing list
OAuth@ietf.org
https://www.ietf.org/mailman/listinfo/oauth

Reply via email to