Using the auth0 passport strategy
https://github.com/auth0/passport-auth0
My callback always has null for a refresh token. Calling the auth0 lock function directly gives me a refreshtoken as expected, however using this oauth passport strategy doesn't return a refreshtoken. Do I need to pass an offline_access scope to auth0 - and if so, how do I pass it in using the Auth0Strategy?
passport.use(new Auth0Strategy({
domain: config.auth0.domain,
clientID: config.auth0.clientId,
clientSecret: config.auth0.secret,
callbackURL: '/login/return'
},
function(accessToken, refreshToken, extraParams, profile, done) {
console.log('refresh token is always null', refreshToken);
}
});
Yes, you need to request one (by adding scope=offline_access). You do this when you send the initial authorization request (e.g. via Lock, or auth0.js, or by simply following link with the right parameters). It is not dependent on passport really. Makes sense?
e.g. try with this:
app.get('/login',
passport.authenticate('auth0', {scope: 'offline_access'}), function (req, res) {
res.redirect("/");
});
Related
I am getting an access token like this:
private async _promptForToken(scopes: string[], prompt: "none" | "consent"): Promise<string> {
const that = this
return new Promise(resolve => {
const tokenClient = google.accounts.oauth2.initTokenClient({
client_id: this.clientId,
scope: scopes.join(' '),
callback: function (tokenResponse) {
that._storeTokenResponse(tokenResponse)
resolve(tokenResponse.access_token)
}
})
tokenClient.requestAccessToken({prompt})
})
}
I storing the token in local storage. If I leave the browser for an hour, so the access token expires, and I come back to my app and click a button that requires a new access token, I am requesting the new token using this code:
this._promptForToken(scopes, 'none')
In other words, I am asking for the same access permissions, but without consent. When I do that I get back a response like this:
{error_subtype: "access_denied", error: "interaction_required"}
Which I can't find documented anywhere, but that's another issue.
If instead, I ask for a new access token using consent i.e.
this._promptForToken(scopes, 'consent')
The Google dialog box for permissions pops up for a second, then disappears, which is horrible UX. And this will happen every time an access token expires. Horrible I say!
What is the recommended way to request a new access token?
Context: browser only, so implicit flow only, I do not want to have to maintain refresh tokens in the backend.
I have a .NET Web API and a small vanilla-JS app using ADAL.js, and I've managed to make them talk nicely to each-other and authenticate correctly.
If I console.log the token returned from adalAuthContext.acquireToken() and manually enter it as Authorization: Bearer {{token}} in Postman, I can also get a valid, authenticated, response from my backend.
However, I can't figure out how to configure Postman's built-in OAuth2.0 authentication UI to get me tokens automatically. I have managed to get tokens in several ways, but none of them are accepted by the backend.
How do I configure Postman to get a token the same way the ADAL.js library does?
For completeness, here's some code:
Backend configuration:
public void Configuration(IAppBuilder app)
{
app.UseCors(CorsOptions.AllowAll);
app.UseWindowsAzureActiveDirectoryBearerAuthentication(
new WindowsAzureActiveDirectoryBearerAuthenticationOptions
{
TokenValidationParameters = new TokenValidationParameters { ValidAudience = "<app-id>" },
Tenant = "<tenant>",
AuthenticationType = "WebAPI"
});
var config = new HttpConfiguration();
config.MapHttpAttributeRoutes();
app.UseWebApi(config);
}
ADAL.js configuration:
const backendUrl = 'http://localhost:55476';
const backendAppId = '<app-id>';
const authContext = new AuthenticationContext({
clientId: backendAppId,
tenant: '<tenant>',
endpoints: [{ [backendAppId]: backendAppId }],
cacheLocation: 'localStorage'
});
Actually making a request:
authContext.acquireToken(backendAppId, (error, token) => {
// error handling etc omitted
fetch(backendUrl, { headers: { Authorization: `Bearer ${token}` } })
.then(response => response.json())
.then(console.log)
})
So since the Azure AD v1 endpoint is not fully standards-compliant, we have to do things in a slightly weird way.
In Postman:
Select OAuth 2.0 under Authorization
Click Get new access token
Select Implicit for Grant Type
Enter your app's reply URL as the Callback URL
Enter an authorization URL similar to this: https://login.microsoftonline.com/yourtenant.onmicrosoft.com/oauth2/authorize?resource=https%3A%2F%2Fgraph.microsoft.com
Enter your app's application id/client id as the Client Id
Leave the Scope and State empty
Click Request token
If you configured it correctly, you'll get a token and Postman will configure the authorization header for you.
Now about that authorization URL.
Make sure you specify either your AAD tenant id or a verified domain name instead of yourtenant.onmicrosoft.com.
Or you can use common if your app is multi-tenant.
The resource is the most important parameter (and non-standards-compliant).
It tells AAD what API you want an access token for.
In this case I requested a token for MS Graph API, which has a resource URI of https://graph.microsoft.com.
For your own APIs, you can use either their client id or App ID URI.
Here is a screenshot of my settings:
I am using the Passport AAD project with the bearer strategy to protect my endpoints. After I receive tokens with the OIDC strategy when logging in, I can't seem to get the bearer strategy to validate the signature of the access token. I get:
authentication failed due to: invalid signature
I have no problems validating the id_token, but I would prefer not to use this for our client app if the id_token can't be refreshed with AAD. Also, when using jwt.io to test the validation with the published public keys, I see the same issue (can validate the id_token, but not the access_token).
Am I missing a step when grabbing the access token, or is there a gap in my understanding of how access tokens are validated?
Update with more details
I am requesting an access token from my tenant:
identityMetadata: https://login.microsoftonline.com/your_tenant_name.onmicrosoft.com/.well-known/openid-configuration,
responseType: 'id_token code'
Using the OIDCStrategy in the AAD Passport project.
const callbackOIDC = (iss, sub, profile, accessToken, refreshToken, params, done) => {
return done(null,{
profile,
accessToken,
refreshToken
});
};
passport.use(new OIDCStrategy(config.creds, callbackOIDC));
Then I run authenticate, shown below:
auth.adCallback = function (req, res, next) {
passport.authenticate('azuread-openidconnect', {
response: res,
resourceURL: 'https://graph.microsoft.com',
session: false
}, function (err, user, info) {
console.log(user.access_token);
})(req, res, next);
};
I think I may have been asking for a graph access token above by specifying the resource URL. If I remove that resource URL, I still get an access token, but the bearer strategy throws an invalid token error (instead of an invalid signature error). Is there a different resource URL I should be setting to match with my tenant and get the access token I'm looking for?
What access tokens are you requesting? If the access token is meant to be used against the Microsoft Graph, for example, it is the Graph's task to validate them- not your app's.
Can you expand on the exact scenario you are trying to implement, and at what point you need to refresh id_tokens?
Here is my work flow for getting access token and refresh token for youtube api. Im generating authorization url with parameters
access_type=offline, response_type=code, redirect_uri=uri, scope=scopes, state=state, client_id=id
from authorization url I´m receiving authentication code, then I´m generating another url to get access_token and refresh_token using code from authorization url with these parameters
code: code, client_id: CLIENT_ID, client_secret: CLIENT_SECRET, redirect_uri: serviceCallback, state: state.callback, grant_type: "authorization_code"
As far as I know user should complete this process only once and then it should be automatic. My problem is that I´m always have to complete authorization and I´m getting always new access_token and refresh_token without forcing it on request.
here is code part where I´m getting authentication url
getAuthUrl: function(scopes, applicationCallback, serviceCallback, siteId,
selectChannel, websiteUrl) {
var requestedClientId = CLIENT_ID;
var scopess =
"https://www.googleapis.com/auth/yt-analytics.readonly https://www.googleapis.com/auth/youtube.readonly https://www.googleapis.com/auth/userinfo.email " +
scopes.replace(",", " ");
return "https://accounts.google.com/o/oauth2/auth?" +
"access_type=offline" +
"&response_type=code" +
/*"&approval_prompt=auto" +*/
"&redirect_uri=" + serviceCallback +
"&scope=" + scopes +
"&state=" + JSON.stringify({
service: NAME,
callback: applicationCallback,
scopes: scopes,
siteId: siteId,
selectChannel: selectChannel,
websiteUrl: websiteUrl
}) +
"&client_id=" + requestedClientId;
},
From there Im getting back code and using that code, clientID and clientSecret to get access token and refresh token
getAuthTokens: function(code, state, res, serviceCallback) {
// Google oAuth endpoint
var endpoint = "https://www.googleapis.com/oauth2/v4/token";
const scopes = state.scopes.split(" ");
// Setup request data
var data = {
code: code,
client_id: CLIENT_ID,
client_secret: CLIENT_SECRET,
redirect_uri: serviceCallback,
state: state.callback,
grant_type: "authorization_code"
};
request.post(endpoint).send(data).type('form').set('Accept',
'application/json').end(function(err, oAuthResponse) {});
},
I was using wrong endpoint url I changed it to different one to one provided by youtube api documentation and removed state parameter from data variable but still doesnt fix the problem
new endpoint url
var endpoint = "https://accounts.google.com/o/oauth2/token";
I´m really confused right now because I´m not forcing authorization and on google apps section there is my app already authorized and it does not update authorization that means it gives permission only first time and after that when I´m pressing allow it doesn´t do anything. OAuth should check if I have refresh token or not, so my conclusion is that I don´t fully understand how it should work or I´m somehow testing everything on debug or test mode where authorization prompt is automatically forced.
I would be really thankful for any kind of help because I feel like I tried everything.
The issue is that the access token that you are using has expired before the next time you use as you have not updated the access token manually using the refresh token.
You need to use the refresh token to update the access token if [ (time you last updated the access token) + (the expiry time) ] has already surpassed.
The concept of refresh tokens is that if an access token is compromised, as it is short-lived, the attacker has a limited time period in which it can be used. Refresh tokens, if compromised, are useless because the attacker requires the client id and client secret in addition to the refresh token in order to gain an access token.
The YouTube API documentation demonstrates the procedure here
By default, the expiry time is around 3 seconds.
This will surely, work in your case.
Adding the following parameter to your authentication object may help...depending on your requirements:
prompt: 'none'
This would mean no consent is gained or needed, after an initial authorization to use the app.
Go to the my accounts settings of google for this account---> go to connected apps and sites ----> manage apps:
Over there can you see the permissions for youtube like this:
I have setup my facebook auth per passportjs docs:
var passport = require('passport')
, FacebookStrategy = require('passport-facebook').Strategy;
passport.use(new FacebookStrategy({
clientID: FACEBOOK_APP_ID,
clientSecret: FACEBOOK_APP_SECRET,
callbackURL: "http://www.example.com/facebook/callback"
},
function(accessToken, refreshToken, profile, done) { ... });
}
));
app.get('/login/facebook', passport.authenticate('facebook'))
.get('/facebook/callback', passport.authenticate('facebook', {successRedirect: '/', failureRedirect: '/login'}));
All this works fine. However, there are cases (such as token expiration) when I want to automatically redirect the user to the page that the user was on before initiating the login request. So I tried to plumb a query string param through the login request (from client to server to facebook and back). But I cant see a way to specify that in the callbackURL.
Furthermore, when I tried hard-coding some context param to the config callbackURL (eg: "http://www.example.com/facebook/callback?redir=lastUserPage") I get an OAuth parse error. Interestingly enough, Facebook does respond correctly with the access code as well as the redir param, but it fails with OAUTH exception:
FacebookTokenError: Error validating verification code. Please make sure your redirect_uri is identical to the one you used in the OAuth dialog request
at Strategy.parseErrorResponse (C:\Sources\node_modules\passport-facebook\lib\strategy.js:198:12)
at Strategy.OAuth2Strategy._createOAuthError (C:\Sources\node_modules\passport-facebook\node_modules\passport-oauth2\lib\strategy.js:345:16)
at C:\Sources\node_modules\passport-facebook\node_modules\passport-oauth2\lib\strategy.js:171:43
at C:\Sources\node_modules\passport-facebook\node_modules\passport-oauth2\node_modules\oauth\lib\oauth2.js:177:18
at passBackControl (C:\Sources\node_modules\passport-facebook\node_modules\passport-oauth2\node_modules\oauth\lib\oauth2.js:124:9)
at IncomingMessage.<anonymous> (C:\Sources\node_modules\passport-facebook\node_modules\passport-oauth2\node_modules\oauth\lib\oauth2.js:143:7)
at IncomingMessage.emit (events.js:117:20)
at _stream_readable.js:943:16
at process._tickCallback (node.js:419:13)
Note that I had this working using WIF before. I don't see any security concerns with passing additional query string parameters through the OAuth process..
Any idea how I can get past this?
I'm not sure how to do what you're asking, but for your desired end goal you could:
Save a cookie before authenticating
Authenticate the user
on the resulting callback page, check for the cookie and redirect if present.
Wouldn't this work just as easily?