I agree, it is a matter of tradeoffs, but I think you have the idea. As for your question in 3, think of rotation as following the same practice as periodic re-authentication of browser based users. If you never do that, you probably don't care to re-authenticate client apps either. But if you do it hourly, you probably want your access tokens to expire hourly too. In that case refresh is useful since it keeps clients from having to go through re-authorization (which usually involves the end-users).
One other thought. If you are using bearer tokens, then a compromise of your token db would lead to more exposure than a compromise of refresh tokens. Why? Because refresh tokens require authentication from the client before they can be used. It is just another layer of protection that again may or may not have value for you. Hope that helps! Phil @independentid www.independentid.com phil.h...@oracle.com On 2011-09-07, at 5:22 PM, Dave Rochwerger wrote: > 1. "The user does not have to be present." > Maybe I should be more clear. What benefit does that have over just a > long-lived (forever) access token? The cost is the extra complication for 3rd > party developers to have to worry about refresh tokens. I can not see a > benefit in our model (everything over SSL, etc) to use refresh tokens. > I want to use refresh tokens - but only if there is a reason for them, which > I can not see at the moment. > > 2. "As Eran points out, you'd have to have do a DB lookup to have true > revocation." > The act of revoking tokens is not a common occurrence, DB lookups to revoke > tokens is not a concern as there is more time spent by the user navigating > the UI (or network latency, etc) than the cost of the DB call. > > 3. "In this sense you get the best of a long-lived credential, combined with > good key rotation and authorization re-verification without having to > re-involve the end-user." > That all sounds good, but in our situation (all SSL, etc) - what do we want > key rotation and re-verification for? I fail to see a reasonable vector for > access token leakage to warrant any of this in our case. > > > On Wed, Sep 7, 2011 at 5:08 PM, Phil Hunt <phil.h...@oracle.com> wrote: > See below... > Phil > > @independentid > www.independentid.com > phil.h...@oracle.com > > > > > > On 2011-09-07, at 4:57 PM, Dave Rochwerger wrote: > >> Hi Phil, >> >> >> The client is then forced to periodically reauthenticate (without the >> >> user) before getting a new access token. >> What benefit does that have? > > The user does not have to be present. > >> >> >> Refresh also gives the authzn server a chance to revoke access. Hence it >> >> is better to use shorter lived access tokens with long lived refresh >> >> tokens. >> That doesn't follow - we can just as easily revoke the single long-lived >> access token. > > As Eran points out, you'd have to have do a DB lookup to have true > revocation. But, by having a short expiration time on the access token (say 1 > hour or less), you get quasi-revocation which has to be re-validated after > the access token expires and the client has to re-authenticate and provide a > valid refresh token. In this sense you get the best of a long-lived > credential, combined with good key rotation and authorization re-verification > without having to re-involve the end-user. > >> >> Dave. >> >> On Wed, Sep 7, 2011 at 4:24 PM, Phillip Hunt <phil.h...@oracle.com> wrote: >> You can also use a long lived refresh token in combination with a short >> access token. The client is then forced to periodically reauthenticate >> (without the user) before getting a new access token. >> >> Refresh also gives the authzn server a chance to revoke access. Hence it is >> better to use shorter lived access tokens with long lived refresh tokens. >> >> Phil >> >> On 2011-09-07, at 15:27, William Mills <wmi...@yahoo-inc.com> wrote: >> >>> I'll talk to the refresh token question: they give you a hook for >>> extensibility and key rotation. If you want to rotate your encryption keys >>> or extend the data carried in the token in any way then you want to be able >>> to cleanly refresh your tokens. Note that the refresh flow allows you to >>> issue a new refresh token at the same time. It also allows a clean path to >>> convert tokens in a new client if you decide you want SAML tokens instead >>> of MAC for example. >>> >>> If you want those things you want to use refresh tokens. You can have long >>> lived access tokens too, and just use the refresh tokens when you want to >>> do something new with the access tokens. >>> >>> -bill >>> >>> From: Dave Rochwerger <da...@quizlet.com> >>> To: oauth@ietf.org >>> Cc: Quizlet Dev Team <devt...@quizlet.com> >>> Sent: Wednesday, September 7, 2011 2:15 PM >>> Subject: [OAUTH-WG] OAuth2 Implementation questions (client secret and >>> refresh tokens) >>> >>> Hi all, >>> >>> I have been implementing OAuth2 based on the various drafts for our new >>> API. Initially, I implemented everything as per the spec, but due to our >>> particular scenario and restrictions we have in place, there are some >>> fundamental questions that I am unable to defend. >>> >>> I am hoping this group could help answer them for me. >>> >>> Our scenario: >>> ========== >>> * We are implementing an API to allow 3rd party developers to access users' >>> protected resources via their applications. The applications will mostly be >>> native phone apps, but some will have web server backends (javascript-only >>> applications are not a concern at the moment). >>> * We want to provide very long-lived (forever) tokens. >>> * We are implementing the "authorization code" flow as that seems best >>> suited to us (we don't want the implicit flow because end-users would have >>> to re-authorize every hour). >>> >>> Our architecture: >>> ============ >>> * We control both the API server and the authorization server. >>> * All requests to protected resources (ie: to the API server) are always >>> done over SSL. >>> * All requests to the authz server (token and authorize endpoints) are >>> always done over SSL. >>> * We enforce that every client must supply the state parameter (and our >>> guidelines say they must verify the state for CSRF mitigation). >>> * We enforce that every client must register a redirect URI. >>> * We validate the redirect_uri used to request an access token is the same >>> that was used to obtain the auth code. >>> * The only time a request is not made over SSL is the redirect with the >>> auth_code which is very short-lived (30 seconds) and is tied to a verified >>> redirect URI. >>> * We enforce that access tokens must be provided using the Authorization >>> header only (and of course, over SSL). >>> * We have guidelines saying that all mobile apps must use the native >>> browser (and not an embedded web UI). >>> >>> Questions: >>> ======== >>> 1. Given the above scenario, what use are refresh tokens? >>> - Access tokens can not leak because every request (to resource and authz >>> server) containing an access token is done over SSL. We control both the >>> authz and resource servers, so tokens in logs (and other suggested reasons >>> in the archives) are not an issue. >>> - Long-lived refresh tokens and short-lived access tokens are supposed to >>> provide security due to possible access token leakage... but in our 100% >>> SSL scenario, if access tokens are able to leak, then so would the client >>> id, secret and refresh token. >>> - Having a long-lived refresh token that can be exchanged for another >>> access token adds a level of complexity (a second HTTPS request every so >>> often) and seems to provide no benefit for our case. >>> >>> >>> 2. What is the point of the client secret (in our scenario)? >>> - We originally were treating the clients as confidential, but after >>> re-reading the native-application section, it seems we really should treat >>> them as public (phone apps can be decompiled and the secret discovered). >>> - The spec says that the authz server should authenticate confidential >>> clients, but public clients are allowed to just send their public client id >>> (and no secret). >>> - The only verification then, is to enforce redirect URI registration and >>> to validate the redirect URI between authorization and token steps. >>> >>> So, the question is, assuming that we, one: "enforce redirect-URI >>> registration" and two: "validate that URI" - why can't we treat all clients >>> as public and not worry about a secret? >>> What is the benefit of having confidential clients (and a secret) at all? >>> >>> >>> Our API source is not available, but the oauth2 server implementation can >>> be seen here: https://github.com/quizlet/oauth2-php >>> >>> Regards, >>> Dave >>> >>> >>> _______________________________________________ >>> OAuth mailing list >>> OAuth@ietf.org >>> https://www.ietf.org/mailman/listinfo/oauth >>> >>> >>> _______________________________________________ >>> OAuth mailing list >>> OAuth@ietf.org >>> https://www.ietf.org/mailman/listinfo/oauth >> > >
_______________________________________________ OAuth mailing list OAuth@ietf.org https://www.ietf.org/mailman/listinfo/oauth
