I want to use Twitter to sign on in an iOS application. Once the user has authenticated with Twitter I want to pass these credentials to a server in order to match these with a user profile stored in a database.
I can comunicate with my server over https, but this still doesn't tell me that the credentials I am being passed from the app are:
Coming from the app - as someone else could start a secure session with my service
Have valid credentials - that the credentials are an actual signed in twitter account and not just some made up numbers (keys etc)
Do I need to implement reverse authentication to provide a method to do this securely? (If not what method should I use to know that the information provided is correct)
I think the methodology is once you got the token/secret from the iOS client is to call the GET account/verify_credentials, then you get the user account and you can compare twitter user id or name with the one stored on your user profiles.
Related
Boomerand and Streak and other similar browser extension apps that use oAuth2 to authienticate the user to google also have to authenticate the user to their backend servers. They dont use a separate login system to their web backend app.
I am assuming they piggyback on the initial oAuth2 traffic with google and set a 3th custom token to identify the user to their backend server.
It gets tricky when multiple gmail users sign in from the same browser. The extension must keep track of every gmail user and associate them with a new token.
Is this how they do it? What would your expert advise be on creating such a 3th token? I am thinking of taking the email address and whatever unique ID I can get from the client browser and encrypt this on the server to create a token for the user. Would this be secure enough? Do you have any better suggestions?
I am trying to support "Hybrid" Federated Login and oAuth2 (using logic from this document) for a webservice which will:
support Sign in using your Google account on my site. That is, from the documentation: You can also choose to use Google's authentication system as a way to outsource user authentication for your application. This can remove the need to create, maintain, and secure a username and password store.
Access the user's Google Analytics.
Here are the steps I have done.
I form my request to https://accounts.google.com/o/oauth2/auth with the scopes (Google Analytics) I want access to.
I Get redirected to google where it has my icon and which scopes I am requesting access to. I grant access.
I get redirected back to the callback page.
I get the tokens (access and refresh), as well as a huge id_token string (which I don't know) and store all of this information in my database.
I then make a call to https://www.googleapis.com/oauth2/v1/userinfo?access_token=xxxyyyzzz to get the user's email and name and store this information in my database too. I also notice it returns a id field which to my knowledge never changes and I presume is some sort of unique identifier. I store this too.
Question: If I go to Authorized Access to your Google Account section in my Google account, it shows that my site has access to "Google Analytics. BUT, it does not say Sign in using your Google account. This is what I am trying to accomplish. I would have thought using the logic would enable Sign in using your Google account. What am I doing wrong? And what would be the applicable call to google so that users can sign in to my site?
If your site has access to something like your Contacts or Analytics using OAuth, you'll never see "Sign in using your Google account". I'm pretty sure that's only if you use OpenID (not OAuth) only for sign-in.
Specifically, OAuth is used for giving you access to APIs to create/update/delete data, while OpenID is for signing in.
If you are asking how to identify user for future logins, you have two options:
Mix OAuth with OpenID, that is called Hybrid. I have described it on this answer.
Use userinfo scope and request userinfo (email, etc.) after successful OAuth authorization. It is described on Google OAuth 2 documentation.
If you mean automatically login to your web site in future visits you can use OpenID "immediate mode" (openid.mode parameter).
When the user is redirected back, you call the second request from your own (server-side?) code, and get their email address. When you successfully get it, that means that the user is logged on. You can add it to the session (e.g. as cookie), and as long as you have it, the user is logged on. You make the user log out by forgetting the email address, so by clearing the session/cookies.
Add this paramter to the https://accounts.google.com/o/oauth2/auth URL call: approval_prompt=force and then the Sign in using your Google account will always show regardless of whether the user was already signed into that or any other account.
So the call would be like this https://accounts.google.com/o/oauth2/auth?client_id=<client id>&redirect_uri=<uri>&scope=<scope>&access_type=<online or offline>&response_type=code&approval_prompt=force
So my intention is to have a login in my iOS app that allows for either our standard email/pwd registration, or login with Facebook. We are also creating rest services to get application info for a given user, e.g. https://url/getPosts/[userId]?userPwd=foo
I've implemented SSO with fb in a web application before but I have some concerns about the security of authentication in a iOS client scenario. The key difference from what I've done before is that in a web app, I was making a server to server call to Facebook to get the access token so I was reasonably assured that the user was authenticated and the web server made privileged calls to the database. In the iOS case, I have the mobile client app making the Facebook iOS authentication request itself and the server has to somehow trust that this user from the client app is indeed authenticated against the matching user record in our database.
My question is how do I generate a durable and secret unique key from the iOS SDK so that I can create and associate a matching user record in our database for users that authenticate only with Facebook. I want this to be seamless so the user would not have to manually fill out another form, and we would just automatically create this matching user record in our db.
I could insert a record into my own users table when they fbDidLogin with Facebook, using the Facebook Id as the unique identifier, and the fb access token as the pseudo password/key for my own user record. I would have to validate the access token with Facebook to make sure it's valid before saving it as a password for the user (the user would never see this password, it would just be passed by the client app during api calls). That way when the user makes a call to our own rest api via the iPhone app we can authenticate and authorize using this secret/pwd/key.
An alternative that would make this whole question moot is to just handle the authorization logic on the client app, and check that there is a valid fb session before making calls to our own apis which I secure with just a single application-wide secret, but that doesn't seem as secure since getting that one secret gives authorization to data on all users. I'd rather authorize at an individual user level. Is that the right choice? Am I being paranoid about iOS security?
The fb access token expires so that might not seem durable, however if I enable offline access that token won't expire but creates a scarier looking oauth dialog window. An alternative to the access token is to hash the fb Id with an application secret key on the iOS client, and use that as the Facebook user's password in our db. However, that again is a single secret key that could perhaps be reverse compiled from the iOS client app?
Design for Facebook authentication in an iOS app that also accesses a secured web service
This post helped me undesrtand it more. If I am not mistaken, the flow goes like this:
User authenticates in iOS app
iOS app takes auth token, sends it to the rails app
Rails app takes auth token and sends it to graph.facebook.com/?auth_token=XXX to get back the user if authentication was successful.
Rails app takes the user info and matches/creates user in own database table. Sends some kind of authentication key back to iOS app.
iOS app saves the authentication key so it can use it to communicate with the rails app.
Let me know if I am missing anything.
Have you looked at the iOS docs for Single Sign On (SSO)? https://developers.facebook.com/docs/guides/mobile/#ios
You can share an app ID across mobile, canvas and web site and the same user auth works for each environment.
Check out: https://developers.facebook.com/docs/authentication/
Facebook Platform provides a number of ways to use the above OAuth flows in different app types, including Websites, Apps on Facebook.com, Mobile and Desktop Apps.
You just need to insert users Facebook key to your database to know if its authenticated with Facebook. Use OAuth at ios side authenticate user take users secret key send it to your rest web-service and save it with users other info.
On iOS (or any mobile), after a successful login, how should I handle login with my own server?
Should I send the token and user id to my server and re-verify it, or is there a better practice?
I obviously don't want to send only the FB user id.
The FBConnect SDK should automatically save login information on the device, so every time the user goes to use the Facebook feature of your app, they are automatically logged in.
The current version of FBConnect doesn't provides obvious proxy authentication, but the old version did. I think you could create this proxy manually, ie. pass all login data, except api secret key to proxy, proxy adds secret key and sends request to FB. Then it returns response of FB server as it is.
Goal:
Allow a user to authentication with Facebook into an iOS application which requires access to a protected web service that I'm running.
Assumptions:
There is a native authentication (and registration) system in place for those users that opt not to use Facebook for sign in.
Details:
Assume we want to offer the option for a user to sign in with Facebook without creating a separate account/credential for our system.
Because we support our own native auth mechanism (username and password) we have our own user IDs and issue an authentication token that is used for subsequent interactions after the initial credential validation.
I'm surprised that Facebook doesn't have best practices for this in their developer documentation. All the existing documentation is either assuming you are building FB auth into a website, or a standalone mobile app with no service that requires authentication.
Here's my initial thoughts on how this would be designed but want validation on whether it's correct.
Client pops the Facebook iOS Login
UI User signs in with Facebook credentials and gets access token
iOS App passes access token to our server
Our server talks to FB graph API using access token to (a) validate the token and (b) get the FB user ID for that access token.
e.g. Our server would call https://graph.facebook.com/me/?access_token=XYZ which would return profile info in a JSON object
Assuming it's valid, our server extracts the User ID from the JSON object and checks whether the user already has an account. If so, we issue our own auth ticket to client to use for that session. If user doesn't have an account, we create a new one with the Facebook User ID, assign our own unique UserID and issue our auth ticket.
Client then passes auth ticket back on subsequent interactions that need authentication.
This seems like the right approach to me but not sure if I'm missing something insanely basic and going down the wrong (complicated) path.
I just dealt with this myself, and here's the part that bit me:
In your step 5... It's possible for a user to register for an account with you entirely separate from their Facebook ID, right? Then some other time they log in with Facebook.... And you just created them a second account and lost their first one.
There needs to be a way to be logged in to your web service, then log in to facebook, and capture the association between the facebook ID and the local account.
Apart from that, your plan sounds solid.
Update: Facebook has added a doc outlining such a scenario HERE
Use https to transmit the auth token to your server, as stated by Facebook
Sharing of Access Tokens
Our Data Policies explicitly prohibit any sharing of an Access Token
for your app with any other app. However, we do allow developers to
share Tokens between a native implementation and a server
implementation of the same App (ie. using the same App ID) as long as
the transfer takes place using HTTPS.
One problem I can see with this strategy, is that somebody can give you an access token obtained for a different facebook app. As far as I know, there's no way to verify that the access token is for your application, so you'll just go on and use it.
It doesn't sound very harmful, though. Generally people/apps try to protect the access tokens, rather than sharing them.
One possible exploit of this would be, for somebody to create their own site or mobile app, obtain access tokens for their users and try to authenticate them, using your API. If this succeeds (the user is has a facebook account in your site), the malicious site will be able to use your API impersonating the user.
It's a bit of a long shot, but I think it could work.
Edit: It looks like there is a way to validate the access token after all. See the answer by #Daaniel on question Get application id from user access token (or verify the source application for a token).
your solution totally works.
Maybe an alternative: why not just get the email on the client from the initial social service request and send to your web service? The web service could just store the email, and maybe a social_provider as well. I understand that your web service will not be able to validate where the email came from, but isn't there a high-trust relationship between your web service and your client? If there is, seems like you can depend on the email coming from the right place. Someone please let me know what obvious thing I'm missing that makes the email-based approach silly...