I am looking to allow a user to sign in with github for a website I am creating that will be API driven. However, I am running into a problem that there seems to be no way of authenticating a user and getting their email and name without redirecting them on the current page to github. I have tried to create a popup, but there is no way for me to get the necessary data from the popup, namely an api token to be used for the next calls.
The answer to the question as framed is "use a different grant flow" but your questions tell me you don't understand OAuth grant flows so you may be better served by a lecture on OAuth.
Popups and iframes are specifically designed to prevent code running outside them from fishing out data, because if you could do it so could invisible iframes in dodgy ads. Give up on that approach because in the unlikely event that you find a way, browser makers will immediately take steps to prevent your solution from working.
The behaviour you describe sounds to me like implicit grant flow.
Here's a walkthrough of the various ways OAuth2 can work.
Why so many ways?
Implicit grant is a flow that uses redirection to deliberately take control away from your code - which might be compromised or impersonated - and give it to a trusted third party, the OAuth provider: GitHub in this case. You pre-register the point to which control is returned after authentication, the point of which is the token is always given to your code.
Amnesia, popups, and what was I doing before we were so rudely interrupted?
There are flows that use popups. Usually this is the authorisation code grant flow. Popups are not suitable for mobile devices and this led to the creation of the implicit grant flow.
The usual reason for not liking redirection is the amnesia it causes. The solution is to store app state persistently prior to passing control to the OAuth provider, and to write a return-from-oauth handler that unpacks the JWT from the redirect parameter and then restores the app state you saved.
Note to self
If the user is anonymous prior to authentication serialise your state to a string and put it in a cookie or in localStorage.
If the user is not anonymous and is authenticating to elevate privilege you'll have to send any privileged data to the server and just persist a retrieval token on the browser. You must ensure that retrieval of this state requires both a valid OAuth token and a valid state token, and that it discards the state after honouring a request so that a replay attack will be conspicuous (you'll have to handle the fallout).
Having acquired a valid token you can use it to request profile information from GitHub. A very small amount may already be present in the JWT; you'll probably have the email address that served as a login, but this isn't necessarily the right one for correspondence.
GitHub OAuth documentation
It appears GitHub only supports code grant auth. That means you have to do the popup/iframe thing. Here's their doco:
https://developer.github.com/v3/oauth/
Code grant - why?
This flow doesn't give you an encoded token, it gives you a code you can redeem for an encoded token. The idea is that when the token expires you can use the code to get a refresh token without sending credentials over the wire.
Not only you have a redirection, but starting Dec. 2020, you also have:
OAuth 2.0 Device Authorization Flow now GA
GitHub Apps and OAuth Apps now feature GA support for the OAuth 2.0 Device Authorization Grant, in addition to the existing Web Application Flow.
This allows any CLI client or developer tool to authenticate using a secondary system with a browser.
GitHub CLI uses this authentication method on the login command.
Read the full documentation on Authorizing OAuth Apps and Authorizing Users for GitHub Apps for more information.
Confirmed March 2022:
Enable OAuth Device Authentication Flow for Apps
From today the OAuth Device Authorization flow feature must be manually enabled for all OAuth and GitHub Apps.
This change reduces the likelihood of Apps being used in phishing attacks against GitHub users by ensuring integrators are aware of the risks and make a conscious choice to support this form of authentication.
If you own or manage an OAuth App or GitHub App that makes use of the OAuth Device Authorization flow, you can enable it for your App via its settings page:
The OAuth Device Authorization flow API endpoints will respond with status code 400 to Apps that have not enabled this feature.
Documentation: "Device flow"
Related
I have a cli (which use a rest api) which needs authentication for use.
As of right now, it supports a token auth. This token is generated at the server on a request with username and password and given as the response.This is not ideal (due to man in the middle attacks) and I am looking for a better protocol to use to generate the tokens.
Users will use such a protocol from a cli, and may or may not have access to a browser on the same device (Though a protocol that requires opening a website is not a problem)
The OAuth device flow seems to be a very simple to use flow, but it is meant
for authorization and not authentication. I also do want to support OAuth as that will require a lot of work, and frankly not what I need.
What is the standard or recommended protocol to use in such a situation?
MITM should not be an issue if your server and app are properly securing the connection. There is nothing really wrong in using a username+password to connect to your backend services. After all, when you're logging into the site you're sending an HTTP request with your username and password to a backend the same way your cli app would do it.
But OAuth indeed can a better fit for cli app:
it's easier to revoke the stolen OAuth token than to force user to change a password,
an app doesn't have to deal with user credentials (although OAuth credentials flow is also exists),
it gives you flexibility when creating new tokens. For example, you may want to issue short lived tokens only to force the user to re-login each time the app is used or you may want to use long-lived refresh tokens.
As you already mentioned, OAuth doesn't handle authentication but you can use your current login flow to verify user credentials and issue an OAuth token (how exactly do that is a separate topic).
I don't think there is a special protocol targeting authentication in cli apps. In any case the app would need to send some secret to a backend. One of the possible solutions is to use OTP (e.g. SMS or email code). In this case you send the code the same way as you would send a password but it is better protected against MITM attacks because a code cannot be used more than once.
TL;DR When using google oauth on desktop app, what to save on disk to avoid repeated sign in? Save the google user id? or the token? or an session id?
I'm creating an little desktop app, whitch must authenticate to my REST API server. I'm using google oauth2 for that.
The idea is, that when the desktop app will be authentivated, it generates some data that will be send to my server. The server will store the data with the google user id received from https://www.googleapis.com/userinfo/v2/me.
On the first run of the desktop app, it will open the default browser, with and url for my server and start an local http server. then:
my server will redirect the browser to google (with the clientid, secret, etc.)
user logs in and it will be redirected back to the server with the oauth code
server uses the code to get the token, and then the user profile and stores the token and the profile in db, then redirects the browser to localhost with an paramerer
the desktop app catches the parameter and stores it in an file on the disk
next time the desktop app will start it only reads the file for the parameter to send the generated data with it to my server
my question is: what the parameter should be? the google user id? the oauth token? an generated session id for this desktop app? or something else?
when it will be the google user id, it can conveniently sent the data with the user id and the rest server will just store it in db as is. but I don't think it's safe
when it will be the token, the rest server has to with every request also get the user profile from google with the token. and imho sending the token with every request isn't safe either
generating an session id means to store it with the user and the token on the server and the desktop app will just store it and send it with every request. but I don't know if it's safe to do that
As it's normally the case in software development you have a couple of options depending on requirements.
The mandatory requirement is that your client (desktop) application needs to send something to your REST API so that the API can perform up to two decisions:
Decide who the user is.
Decide if the user is authorized to perform the currently requested action.
The second step may not be applicable if all authenticated users have access to exactly the same set of actions so I'll cover both scenarios.
Also note that, for the first step, sending the Google user ID is not a valid option as that information can be obtained by other parties and does not ensure that the user did authenticate to use your application.
Option 1 - Authentication without fine-grained authorization
Either always sending the id_token or exchanging that token with your custom session identifier both meet the previous requirement, because the id_token contains an audience that clearly indicates the user authenticated to use your application and the session identifier is generated by your application so it can also ensure that. The requests to your API need to use HTTPS, otherwise it will be too easy for the token or session ID to be captured by an attacker.
If you go with the id_token alternative you need to take in consideration that the token will expire; for this, a few options again:
repeat the authentication process another time; if the user still has a session it will indeed be quicker, but you still have to open a browser, local server and repeat the whole steps.
request offline_access when doing the first authentication.
With the last option you should get a refresh token that would allow for your application to have a way to identify the user even after the first id_token expires. I say should, because Google seems to do things a bit different than the specification, for example, the way to obtain the refresh token is by providing access_type=offline instead of the offline_access from OpenID Connect.
Personally, I would go with the session identifier as you'll have more control over lifetime and it may also be simpler.
Option 2 - Authentication + fine-grained authorization
If you need a fine-grained authorization system for your REST API then the best approach would be to authenticate your users with Google, but then have an OAuth 2.0 compliant authorization server that would issue access tokens specific for your API.
For the authorization server implementation, you could either:
Implement it yourself or leverage open source components
⤷ may be time consuming, complex and mitigation of security risks would all fall on you
Use a third-party OAuth 2.0 as a servive authorization provider like Auth0
⤷ easy to get started, depending on amount of usage (the free plan on Auth0 goes up to 7000 users) it will cost you money instead of time
Disclosure: I work at Auth0.
There should be no problem sending the access_token with every request since they are created for that purpose and are thus short lived. You can use the Google Authorization Server endpoint to verify a token instead of using it to do a request for a users profile.
If you're only relying on Google for authentication, here's how your workflow can look:
the client (desktop application, in your case) retrieves the
Google id_token following the user's log in, and then sends it to
the server
the server validates the integrity of said token and extracts the user's profile data; this could mean a simple GET on Google's endpoint to verify this token: https://www.googleapis.com/oauth2/v3/tokeninfo?id_token={0}
On subsequent requests, nothing should change really, except that the user's login process will be automated (since he's given permissions & all), and thus much faster. #danielx is right, there's no problem with sending the token each and every time.
Use-case: Emails to be sent from a web-application upon an event, as someuser#somedomain.com via MS Exchange or Outlook.com, using the RESTful APIs exposed by Outlook.com. Only HTTP access allowed (=> no SMTP/IMAP).
All documentation seems to mention that the app has to forward users to MSOnline, sign-in and then use the authorization code sent back by MS online.
But, this won't work for a background task (=> no sign-in possible!) where a pre-built token(with some predefined scope) is necessary so that Outlook.com can be accessed via APIs to send mail as someuser#somedomain.com.
Any hints/pointers to how it could be done? Basically, automated authentication without explicitly signing in as 'someuser#somedomain.com' on the MS Online login page.
I did not find M$ documentation regarding Outlook REST APIs to be of any great help and found it to be pretty difficult to navigate/understand. :(
Thanks!
At some point you will have to have the user sign in to grant access to your app. So you would need to have some sort of user-facing web front end where they can do this. Once they have signed in and you've obtained an access token/refresh token, your background app should be able to use those tokens silently, without user interaction, at least until the user either revokes access or the refresh token expires.
Currently Azure (which provides the login/token functionality) does expire the refresh tokens after some time (90 days), at which point the user must sign in again to grant your app continued access.
Preface
I'm developing several web services and a handful of clients (web app, mobile, etc.) which will interface with said services over HTTP(s). My current work item is to design an authentication and authorization solution for the product. I have decided to leverage external identity providers, such as Facebook, Google, Microsoft, Twitter, and the like for authentication.
I'm trying to solve the problem of, "when a request comes to my server, how do I know who the user is and how can I be sure?". More questions below as well...
Requirements
Rely on external identities to indicate who I'm dealing with ('userId' essentially is all I care about).
The system should use token-based authentication (as opposed to cookies for example or basic auth).
I believe this is the right choice for scaling across multiple clients and servers while providing loose coupling.
Workflow
Based on my reading and understanding of token-based authentication, the following is how I imagine the workflow to be. Let's focus for now on Facebook in a web browser. My assumption is that other external identity providers should have similar capabilities, though I have not confirmed just yet.
Note, as of writing, I'm basing the following off of Facebook login version 2.2
Client: Initiates login to Facebook using the JavaScript SDK
Facebook: User authenticates and approves app permissions (to access user's public profile for example)
Facebook: Sends response to client which contains user’s access token, ID, and signed request
Client: Stores user access token in browser session (handled by SDK conveniently)
Client: Makes a request to my web service for a secure resource by sending along the user’s access token in the authorization header + the user’s ID (in custom header potentially)
Server: Reads user access token from request header and initiates verification by sending a request to the debug_token graph API provided by Facebook
Facebook: Responds back to the server with the user access token info (contains appId and userId)
Server: Completes verification of the token by comparing the appId to what is expected (known to itself) and the userId to what was sent on the client request
Server: Responds to the client with the requested resource (assuming the happy authorization path)
I’m imagining steps 5-9 would be repeated for subsequent requests to the server (while the user’s access token is valid – not expired, revoked from FB side, app permissions changed, etc.)
Here's a diagram to help go along with the steps. Please understand this system is not a single page application (SPA). The web services mentioned are API endpoints serving JSON data back to clients essentially; they are not serving HTML/JS/CSS (with the exception of the web client servers).
Questions
First and foremost, are there any glaring gaps / pit falls with the described approach based on my preface and requirements?
Is performing an outbound request to Facebook for verifying the access token (steps 6-8 above) per client request required / recommended?
I know at the very least, I must verify the access token coming from the client request. However, the recommended approach for subsequent verifications after the first is unknown to me. If there are typical patterns, I’m interested in hearing about them. I understand they may be application dependent based on my requirements; however, I just don’t know what to look for yet. I’ll put in the due diligence once I have a basic idea.
For instance, possible thoughts:
Hash the access token + userId pair after first verification is complete and store it in a distributed cache (accessible by all web servers) with expiry equal to access tokens. Upon subsequent requests from the clients, hash the access token + userId pair and check its existence in the cache. If present, then request is authorized. Otherwise, reach out to Facebook graph API to confirm the access token. I’m assuming this strategy might be feasible if I’m using HTTPS (which I will be). However, how does performance compare?
The accepted answer in this StackOverflow question recommends creating a custom access token after the first verification of the Facebook user token is complete. The custom token would then be sent to the client for subsequent requests. I’m wondering if this is more complex than the above solution, however. This would require implementing my own Identity Provider (something I want to avoid because I want to use external identity providers in the first place…). Is there any merit to this suggestion?
Is the signedRequest field present on the response in step #3 above (mentioned here), equivalent to the signed request parameter here in the ‘games canvas login’ flow?
They seem to be hinted as equivalent since the former links to the latter in the documentation. However, I’m surprised the verification strategy mentioned on the games page isn’t mentioned in the ‘manually building a login flow’ page of the web documentation.
If the answer to #3 is ‘Yes’, can the same identity confirmation strategy of decoding the signature and comparing to what is expected to be used on the server-side?
I’m wondering if this can be leveraged instead of making an outbound call to the debug_token graph API (step #6 above) to confirm the access token as recommended here:
Of course, in order to make the comparison on the server-side, the signed request portion would need to be sent along with the request to the server (step #5 above). In addition to feasibility without sacrificing security, I’m wondering how the performance would compare to making the outbound call.
While I’m at it, in what scenario / for what purpose, would you persist a user's access token to a database for example?
I don’t see a scenario where I would need to do this, however, I may be overlooking something. I’m curious was some common scenarios might be to spark some thoughts.
Thanks!
From what you describe I'd suggest to use a server-side login flow as described in
https://developers.facebook.com/docs/facebook-login/manually-build-a-login-flow/v2.2
so that the token is already on your server, and doesn't need to be passed from the client. If you're using non-encrypted connections, this could be a security risk (e.g. for man-in-the-middle attacks).
The steps would be:
(1) Logging people in
You need to specify the permission you want to gather from the users in the scope parameter. The request can be triggered just via a normal link:
GET https://www.facebook.com/dialog/oauth?
client_id={app-id}
&redirect_uri={redirect-uri}
&response_type=code
&scope={permission_list}
See
https://developers.facebook.com/docs/facebook-login/manually-build-a-login-flow/v2.2#login
(2) Confirm the identitity
GET https://graph.facebook.com/oauth/access_token?
client_id={app-id}
&redirect_uri={redirect-uri}
&client_secret={app-secret}
&code={code-parameter}
https://developers.facebook.com/docs/facebook-login/manually-build-a-login-flow/v2.2#confirm
(3) Inspect the access token
You can inspect the token as you already said in your question via
GET /debug_token?input_token={token-to-inspect}
&access_token={app-token-or-admin-token}
This should only be done server-side, because otherwise you'd make you app access token visible to end users (not a good idea!).
See
https://developers.facebook.com/docs/facebook-login/manually-build-a-login-flow/v2.2#checktoken
(4) Extending the access token
Once you got the (short-lived) token, you can do a call to extend the token as described in
https://developers.facebook.com/docs/facebook-login/access-tokens#extending
like the following:
GET /oauth/access_token?grant_type=fb_exchange_token
&client_id={app-id}
&client_secret={app-secret}
&fb_exchange_token={short-lived-token}
(5) Storing of access tokens
Concerning the storing of the tokens on the server, FB suggests to do so:
https://developers.facebook.com/docs/facebook-login/manually-build-a-login-flow/v2.2#token
(6) Handling expired access tokens
As FB doesn't notify you if a token has expired (and if you don't save the expiry date and compare this to the current timestamp before making a call), it's possible that you receive error messages from FB if the token got invalid (after max. 60 days). The error code will be 190:
{
"error": {
"message": "Error validating access token: Session has expired at unix
time SOME_TIME. The current unix time is SOME_TIME.",
"type": "OAuthException",
"code": 190
}
}
See
https://developers.facebook.com/docs/facebook-login/access-tokens#expiredtokens
If the access token becomes invalid, the solution is to have the person log in again, at which point you will be able to make API calls on their behalf once more. The login flow your app uses for new people should determine which method you need to adopt.
I dont' see any glaring gaps / pit falls, but I'm not a security expert.
Once your server has verified the given token (step 8), as you said:
The accepted answer in this StackOverflow question recommends creating a custom access token after the first verification of the Facebook user token is complete. The custom token would then be sent to the client for subsequent requests. I’m wondering if this is more complex than the above solution, however. This would require implementing my own Identity Provider (something I want to avoid because I want to use external identity providers in the first place…). Is there any merit to this suggestion?
IMHO is the way to go. I would use https://jwt.io/ which allows you to encode values (the userId for example) using a secret key.
Then your client attach this token to every request. So you can verify the request without need to a third party (you don't need database queries neither). The nice thing here is there is no need to store the token on your DB.
You can define an expiration date on the token, to force the client authenticate with the third party again when you want.
Let's say you want your server be able to do some action without the client interaction. For example: Open graph stories. In this scenario because you need to publish something in the name of the user you would need the access token stored on your DB.
(I can not help with the 3 and 4 questions, sorry).
Problem with Facebook is that they do not use OpenId connect on top of Oauth (https://blog.runscope.com/posts/understanding-oauth-2-and-openid-connect).
Thus resulting in their custom ways of providing Oauth authentification.
Oauth2 with OpenId connect identity services usually provide issuer endpoint where you can find URL (by appending ".well-known/openid-configuration") for jwk's which can be used to verify that JWT token and its contents were signed by the same identity service. (i.e access token originated from the same service that provided you jwk's)
For example some known openid connect identity providers:
https://accounts.google.com/.well-known/openid-configuration
https://login.microsoftonline.com/common/v2.0/.well-known/openid-configuration
(btw it is not a coincidence that Attlasian provides only these two services to perform external login)
Now as you mentioned, you need to support multiple oauth providers and since like Facebook not all providers use same configuration of oauth (they use different JWT attribute names, toke verification methods, etc. (Openid connect tries to unify this process)) i would suggest you to use some middleware identity provider like Oauth0 (service not protocol) or Keycloak. These can be used with external identity providers (Social pages as you mentioned) and also provides you with custom user store.
Advantage is that they unify authentication process under one type (e.g both support openid connect). Whereas when using multiple oauth providers with not unified authentication workflow you will end up with redudant implementations and need for merging different information's under one type (this is basically what mentioned middle-ware identity providers solve for you).
So if you will use only Facebook as identity provider in your app then go for it and make implementation directly for Facebook Oauth workflow. But with multiple identity providers (which is almost always case when creating public services) you should stick with mentioned workaround or find another one (or maybe wait till all social services will support Openid connect, which they probably wont).
There may be hope.. This year, Facebook have announced a "limited login" feature, which, if they were to add to their javascript sdks would certainly make my life easier:
https://developers.facebook.com/blog/post/2021/04/12/announcing-expanded-functionality-limited-login/
At the time of writing, I can only find reference to iOS and Unity SDKs, but it does seem to return a normal JWT, which is all I want!
We have a regular web application with cookie based auth and now we want to split frontend and backend (api) in order to have third-party public API. So our backend will be on one domain and frontend on another one.
For authorization we would like to switch for OAuth 2 with JWT. In this case our frontend app will have to use access_token instead of cookie session and it brings a big old question:
How To Remain Logged In - The Infamous "Remember Me" Checkbox (part II from Form based authentication for websites)
From OAuth2 point of view our frontend application going to use something between Resource Owner Password Credentials Grant and Implicit Grant. It is closer to Password Credentials Grant since we still going to use usual login form and won't redirect user to another domain in order to sign in. At the same time it is closer to Implicit Grant since it's all going to be browser-only & JavaScript based when access_token will be saved in browser.
The RFC says the authorization server MUST NOT issue a refresh token if you use Implicit Grant and my question is if it's still valid in this use case when you don't really use a 3-d party OAuth but your own api? Instinctively I feel that having refresh_token in browser is a security hole and would like to confirm it with you guys, but that refresh_token seems to be the only way to have persistent login working the same way as we had with cookies.
**UPD** after #FlorentMorselli comment:
The OpenID specs still do not answer my question if I can use refresh_token with browser only application
Google says they provide refresh_token only for access_type=offline
OpenID Connect Core says you cannot use Refresh Token with Implicit Flow
OpenID Connect Core says nothing about using refresh_token with Hybrid Flow
There's only one place where it says something promising about refresh_token with Hybrid Flow, but nothing precise
UPD2 thanks to #reallifelolcat
It looks like OpenID Connect does not explicitly support Resource Owner Password Credentials Grant, meaning you have to redirect user to OpenID Connect server to perform login. Do you know if there is another way to authenticate with user credentials over OAuth 2.0?
I believe splitting api and frontend is getting more common these days and I'd appreciate it if you share how you solve this Persistent Login issue and if you drop it completely and force user to re-login every X weeks.
Thanks!
Access tokens and refresh tokens have nothing to do with login with OpenID Connect. These are only for authorizing access to user profile information and for perhaps authenticated service calls to your public API after the fact of login. Refer to the spec for the difference between the ID Token and the Access Token.
If you are going to use OpenID Connect for login, then from what you've wrote so far, it sounds like you need to host your own OpenID Provider (OP) since you want to avoid going to another domain to sign in:
we still going to use usual login form and won't redirect user to another domain in order to sign in.
If you want to be your own Identity Provider, then more power to you. This means that you going to have to deploy your own working instance of an OpenID Connect server, complete with authorization and token endpoints.
Now this is the part where your persistent login comes in. Your browser webapp will be a relying party to the OP server you now have. When a user tries to login to your browser app using OpenID Connect, they will need to authenticate themselves to your OP server. Going through the OIDC flow, your browser app will get an ID token containing an issuer/subject pair identifying the user.
It's up to you to determine how the user stays logged into your OP server, but as long as the user at least authorizes the browser app once:
http://openid.net/specs/openid-connect-core-1_0.html#Consent
then you can save that consent for all future requests by this browser app to login, and therefore maintain a persistent login.
You're going to have to consider how you're going to handle sessions management, but it sounds like you have some cookie thing going already so you might be able to use that (see this answer: OpenID sign in mechanism - Stay signed in ). Otherwise, you're going to end up with a situation where your browser webapp has to get a new id token all the time.
Also as Florent mentioned, there are security considerations you should consider when doing a public client thing that your browser based webapp would be. Example:
https://www.rfc-editor.org/rfc/rfc6749#section-10.16