My system works as follows;
I have an ASP.Net RESTful API server, which contains a user database.
I also have a Xamarin.Forms application, with a registration and login page.
What I want is to have the ability to authenticate a login using a social IdP, and then, if that user has not been logged in before, register that user in my local database.
I have been reading up on how to implement OAuth 2.0 with OpenID Connect to authenticate my users with a social IdP, however, I cannot seem to wrap my head arround it. From what I've read, I shouldn't use an Access token for authentication, since that i what the ID token is for, however, I have also read that the only intended purpose for an ID token, is the client.
My problem then is, how can I make sure that calls made to my ASP.Net server, has been made by "real person", and how do I determine who makes the call?
Access token will be used determine whether the client application was authorized by a user to access a resource. The concept of ID token comes from OpenID Connect. Main purpose of the ID token is to authenticate the user to the client application (i.e. letting the client application know that the person who authorized the access is a valid person).
To do this, you have to validate the ID token. This can be done using third party libraries such as nimbusds or auth0. You can validate the signature of the token verify the integrity of the token and check the claims included in the token (by comparing them with expected values) to verify the user details. Also, you can add custom claims (any claim that is specific for your application/implementation) to the tokens through your identity provider so that you'll be able to validate those particular claims in order to verify the user.
Related
My current setup is the following:
Single Page Application with Vue
Rest Backend
The SPA has social login functionality. It uses the authorization code flow with PKCE to retrieve:
Access Token / Refresh Token
ID-Token
As it is an SPA, the information is stored in local storage and therefore not 100% secure.
But how do I use this information to actually authenticate against my rest backend? The rest backend actually contains the user data I need after all.
Originally, I thought I could just send my access token or id token to my backend and the backend uses this as proof that I'm the correct user.
Main problems I see:
Make sure the access token / id token is actually from my application (the SPA)
Be protected if the tokens are stolen / Minimize the impact of this
For the first problem, the client_id might help, which is embedded in the id-token. It is kind of public information (because it is an SPA) and there is no client_secret. But the redirect_uri is specific to my SPA. Is this enough protection?
If it is my backend could have a list of allowed client ids and providers and check if the client id of the token is one of them.
The second problem is the lifetime of the token. Access tokens are only valid for a short time period and refresh token should be rotated. So it's kind of okay to store these in local storage. But the id_token is valid for a longer time. What do to about this?
Is this in general the right track? Or is my approach completely wrong?
I like to implement the functionality
where if two users are trying to login with the same credentials then the first user should log out as soon as the second user login.
consider user one is logged in with his credentials from one machine
and he/ another user is trying to log in from another machine
then the user one session should be removed as soon as user one logged in.
Ps:
I tried to implement that by saving the current session id in the user table and overriding the OnCreated method from the IAuthSession interface and then checking in that if the request sessionId is the same as the saved session Id if same then process the request else call the lout endpoint.
But It will be not good for performance and I am not sure if it is a good way to do that?
PS: I am using a JWT token.
Update :
I am able to clear the session by using ICacheClient to get the session and then remove a session from the server using IRequest.RemoveSession(sessionId), but is it not log out the specific user.
You can't invalidate a user authenticating with stateless authentication like JWT which has the signed authentication embedded in the Token which is valid until the JWT expiry.
i.e. you can't revoke a JWT Token after it's already been issued.
There is a JwtAuthProvider.ValidateToken filter you can use to execute custom logic to prevent a user from authenticating which you may be able to use however that would require that you manage a collection of Token info you want to prevent from authenticating before its Token expiry.
In Firebase, it is possible to log in a user as anonymous with a token, and when the user decides to register, just update the credentials. I have a hard time understanding whether the same is possible with Auth0.
We are now using Auth0 as identity provider, the flow is the following:
The user is using the app anonymously with limited access.
User registers in the app with Auth0.
Auth0 issues a token
Firebase receives the token and lets the user use the restricted parts. All the data generated by the anonymous user is lost
What I want to achieve:
When the user starts using the app, Auth0 immediately creates a user token
The token is valid to access certain parts of Firebase database
If/when the user decides to register, their token remains valid but receives additional credentials
Firebase lets the user use the restricted parts
It's Konrad from Auth0 Community Team. Yep as Baskaro said unfortunately it's not supported from our side of stack. It will be great if you can submit it as a feature request to our product team using our feedback form (you will be contacted by one of our product managers within 10 business days):
https://auth0.com/feedback
There is an app that wants to authenticate with my users using oAuth2.
So they open a window, with the authorize URL, and parameters (such as redirect uri)
Like: https://my-website.com/api/authLauncherauthorize?redirect=SOME_URI
Now I have my own firebase-login, and when the user logs in, I get their access token from firebase. Which is what I want to respond with.
However, in oAuth2 guides/explanations like https://aaronparecki.com/oauth-2-simplified/ I see I am supposed to return an authorization code, and I don't understand where can I get that from?
What I can do, is generate a bullshit code, pair it in the DB to the access token, and then in the "token" request, send the correct access token. Is that what I am supposed to do?
Just to be clear, this is my first time writing an oAuth2 service myself.
OAuth is a system that provides authenticated access to resources. This resource can be for example a user page or editing rights to that user page. So your goal is to provide access to permissions to the right people.
When someone logs in, they get a token. Your part is to generate that token however you want, may it be some form of userdata into base64 or completely random. Take this token and link it against permissions, like viewing a page, editing it or even simpler things like viewing the email of a user.
OAuth2 tokens and/or permissions should be revokable without deleting a user. You should not use OAuth2 to identify someone.
If I am understanding your question correctly:
User visits some website
User wants to register or login using your websites OAuth2
You redirect back to the original page and send your generated token
The page can access content on your site with this token
Assuming you are the Host Site, given a User who wants to connect a 3rd party application, then the flow would be like this:
User lands on site - Clicks Login with Github
User is redirected to Github site where they login and click "Authorize"
Github redirects user back to your site /authorize with an auth token.
Your site then passes that token back to the 3rd party API (github in this case) in exchange for an access token and refresh token.
You can then pass that Authorization token to an API endpoint to get details about it. If the token expires, you can use the refresh token to get a new Auth token. Both Tokens should be stored in your database for your user.
However writing that all out I realize you are asking how do you generate the Authorization token, so I'm guessing you're actually the 3rd party API in this example. So you would want to generate an Authorization token using a random generator. Since you are using firebase, you'll probably wanna try out their token generator: https://github.com/firebase/firebase-token-generator-node
There's also some more up-to-date info here I believe: https://firebase.google.com/docs/auth/admin/#create_a_custom_token
And like you said, you would store that in a database associated with the user, and then when the Host Site sends that user's auth token to your server, you exchange it for the Authorization token (and refresh token if requested).
It's also worth reading through how google does it, because you'd be doing something similar: https://developers.google.com/identity/protocols/OAuth2UserAgent#validatetoken
JWT is another option of generating tokens: https://jwt.io/
I want to enable Facebook authentication and the FB-Graph in my website, which already has forms authentication. Using http://multitiered.wordpress.com/2010/08/05/getting-started-with-the-facebook-c-sharp-sdk/, I was able to figure out how to login server-side.
However, the problem with this approach is that a secure cookie will not be created, since the call returns the authentication code in the querystring via a callback. This means that the user will have to login every time.
I can see two ways around this:
Store the access token in a secure cookie manually
Instead of the above approach, use the FB JS API to login - this stores a secure cookie with the access token automatically
I would prefer not to use the second approach, as I would like the login code to be server-side.
Which would be the better approach? Am I missing something?
I use the JavaScript method to first authenticate the user, the JS SDK then writes an encrypted cookie (called "fbs_[YourAppID]") when a connected user hits your page; using one of the many Facebook c# SDKs, this cookie can be decoded using your application secret giving you the user ID, oAuth token, expiry date etc.
Then I hook into the AuthenticateRequest event of my .NET application, check the presence of the cookie, decode if it found, and then find a user who has been assigned this facebook ID (your user table must have a extra field for storing the ID of their facebook account).
If a match is found, I write a normal forms authentication cookie for this user, then .NET will recognise them for all future requests. If no user is found, then this is a brand new user who has just connected. Use the SDK again to query the graph API using their oAuth token, get things like their name/email etc and create a new account, then issue a authentication token as normal.
By writing a normal authetication cookie, the user will stay logged into to your site for all requests, just as if they were a normal user.
One side point, when using email address, check for duplicates, and check for the facebook cookie in all requests. For example, an existing registered logged in user may have just connected.