I am implementing a REST API that has both mobile application and browser based clients and users. Based on questions I've asked and previous questions here and at security.stackexchange, I have come to the conclusion that to stay as "RESTful" as I can for as long as I can, HTTP Basic Auth over SSL is sufficient for Authentication. The problem is I'd also like to implement Two Factor Authentication along with it. Is it acceptable to add headers in the 401 Authorization header response, like username:password:token, or in a totally separate request header, but in the same payload as the basic auth response by the client? Since I'm using node.js + express/connect, I have access to the entire HTTP protocol stack, but want to remain as restful as possible for scalability reasons. On the browser side, I guess I could do the basic auth, and if it passes, ask for the TFA token, and only if it passes consider the user authenticated.
You can technically make up new authentication schemes to extend from HTTP Basic Auth, but they generally won't be supported by browsers. In your example, I don't believe any browser would be able to natively ask for and send username:password:token in the same way they can easily ask for username and password.
Generally two-factor authentication schemes work by putting the user into an intermediary state using some form of sessions as you mentioned in your second example. A user who has passed the first factor, say username/password via Basic Auth, has a session opened but not marked as really logged in until they also pass the second factor. Inputting a dongle code or something like that. Once both factors are passed their session is marked as fully logged in and they can access their account/data/whatever.
Related
I have a Cloudflare Worker that presents a registration form, accepts input from the user that is posted back to the Worker which then sends that on to an Node HTTP API elsewhere (DigitalOcean if that matters) that inserts the data into a MongoDB (though it could be any database). I control the code in both the CF-Worker and the API.
I am looking for the best way to secure this. I am currently figuring to include a pre-shared secret key in the API call request headers and I have locked down what this particular API can do with database access control. Is there an additional way for me to confirm that only the CF Webworker can call the API?
If this is obvious to some I apologize. I have always been of the mind that unless you are REALY good at security it is best to consult those who are.
You can research OAuth2.0 standard. That is authorization standard for third party clients. Here is link: https://oauth.net/2/
This solution is the most professional.There are other less secure ways to do it, but easier to implement. Password and username, x-api-key, etc..
It sounds to me that you can also block all IPs and allow only requests from that specific domain name (CF Worker)
I'm developing my web-site project based on NodeJs/Express, and for some UI parts I'm using Jquery ajax request to fetch secondary data.
How can we handle some basic control on our Rest API end-points that are used for ajax calls by the browser?
I was thinking about some kind of token authorization , but it can be also used by other clients (scripts etc.) once it has been intercepted , so how can we protect our server from unwanted requests? What other controls should be used in this cases (recognize too many request from same client, clients black list,etc)?
There are three main topics Authentication, Authorization, Security. I will give links and only shortly answers. Subject is enough big to write few books.
Authentication - who is the one who is making request. There are many 'strategies' to authentication user. Please check most pupular module for this : http://passportjs.org/docs.
Of course you can inplement one or more of this strategies alone.
For stateless authentication jwt tokens are very convenient. If you want to code it yourself (Passport has this strategy) check this link (one of many in web) https://scotch.io/tutorials/authenticate-a-node-js-api-with-json-web-tokens.
How to prevent from token interception? Use always https and set token expiration time short.
Where to store your token client side? for detail look at this https://stormpath.com/blog/where-to-store-your-jwts-cookies-vs-html5-web-storage/ In short don't store in web storage because of XSS attacks. Use cookies, when they are correctly configured they are safe (more in attached link), if not configured they are very exposed to threats.
Authorization : we know user, but he has access only to some resources. Please check https://github.com/OptimalBits/node_acl
There is gist with node_acl and passport : https://gist.github.com/danwit/e0a7c5ad57c9ce5659d2
In short passport authenticate user. We now who want what. We setup roles and resources and define roles and resources relation. Then we set for each user roles. Module will check for us user permission.
Security: please look for this subject in documentation of sails framework http://sailsjs.org/documentation/concepts/security they describes attacks and how framework prevent form them. I write about express:
DDOS: (part of your question "too many request from same client") "At the API layer, there isn't much that can be done in the way of prevention". This is subject most for servers admins. In short use load balancer. If it is one IP (not hundreds) then blacklist or deley response (for start look at this https://www.npmjs.com/package/delayed-request but I thing that solution must be more sophisticated).
CSRF: "type of attack which forces an end user to execute unwanted actions on a web application backend". Look at this module https://www.npmjs.com/package/csrf
XSS: "type of attack in which a malicious agent manages to inject client-side JavaScript into your website" don't trust any data from user. Always validate, filter, santize. Look at this https://www.npmjs.com/package/xss
In documentation of sails, there is more attack types but above are most popular.
Use express session + passport (http://passportjs.org/)
Basically you should have a login to the website and than only authenticated users can call the REST apis.
now... if you don't want a login, than you can't really protect the APIs as the website by design is open.
You didn't specify much info so it is hard to say more than that.
Also DoS attacks can't be protected by your code, and usually its not the responsibility of the app server (in your case node.js express) to provide such protection. If someone wants your website down by doing DoS attacks, than without other layers (see https://en.wikipedia.org/wiki/Denial-of-service_attack#Defense_techniques) which mostly mean it is up to a router/switch/so on... to implement.
For a project I’m working on currently I am developing an API using Node/Express/Mongo and separately developing a website using the same tools. Ideally I want to host these on separate servers so they can be scaled as required.
For authentication I am using jsonwebtoken which I’ve set up and I’m generally pleased with how it’s working.
BUT…
On the website I want to be able to restrict (using Express) certain routes to authenticated users and I’m struggling a little with the best way to implement this. The token is currently being saved in LocalStorage.
I think I could pass the token through a get parameter to any routes I want to protect and then check this token on the website server (obviously this means including the jwt secret here too but I don’t see a huge problem with that).
So my questions are
Would this work?
Would it mean (no pun intended) I end up with ugly URLs
Would I just be better hosting both on the same server as I could then save the generated token on the server side?
Is there a better solution?
I should say I don’t want to use Angular - I’m aware this would solve some of my problems but it would create more for me!
First off, I'll answer your questions directly:
Will this work? Yes, it will work. But there are many downsides (see below for more discussion).
Not necessarily. I don't really consider ugly urls to include the querystring. But regardless, all authentication information (tokens, etc.) should be included in the HTTP Authorization HEADER itself -- and never in the URL (or querystring).
This doesn't matter so much in your case, because as long as your JWT-generating code has the same secret key that your web server does, you can verify the token's authenticity.
Yes! Read below.
So, now that we got those questions out of the way, let me explain why the approach you're taking isn't the best idea currently (you're not too far off from a good solution though!):
Firstly, storing any authentication tokens in Local Storage is a bad idea currently, because of XSS (Cross Site Scripting attacks). Local Storage doesn't offer any form of domain limitation, so your users can be tricked into giving their tokens up quite easily.
Here's a good article which explains more about why this is a bad idea in easy-to-understand form: http://michael-coates.blogspot.com/2010/07/html5-local-storage-and-xss.html
What you should be doing instead: storing your JWT in a client-side cookie that is signed and encrypted. In the Node world, there's an excellent mozilla session library which handles this for you automatically: https://github.com/mozilla/node-client-sessions
Next up, you never want to pass authentication tokens (JWTs) via querystrings. There are several reasons why:
Most web servers will log all URL requests (including querystrings), meaning that if anyone gets a hold of these logs they can authenticate as your users.
Users see this information in the querystring, and it looks ugly.
Instead, you should be using the HTTP Authorization header (it's a standard), to supply your credentials to the server. This has numerous benefits:
This information is not typically logged by web servers (no messy audit trail).
This information can be parsed by lots of standard libraries.
This information is not seen by end-users casually browsing a site, and doesn't affect your URL patterns.
Assuming you're using OAuth 2 bearer tokens, you might craft your HTTP Authorization header as follows (assuming you're representing it as a JSON blob):
{"Authorization": "Bearer myaccesstokenhere"}
Now, lastly, if you're looking for a good implementation of the above practices, I actually wrote and maintain one of the more popular auth libraries in Node: stormpath-express.
It handles all of the use cases above in a clean, well audited way, and also provides some convenient middlewares for handling authentication automatically.
Here's a link to the middleware implementations (you might find these concepts useful): https://github.com/stormpath/stormpath-express/blob/master/lib/authentication.js
The apiAuthenticationRequired middleware, itself, is pretty nice, as it will reject a user's request if they're not authenticating properly via API authentication (either HTTP Basic Auth or OAuth2 with Bearer tokens + JWTs).
Hopefully this helps!
I am developing an express app which serves as a REST api with web client and may be future mobile clients. I am using a Oauth 2.0 token authentication for both clients. This gives a good deal of security against CSRF. I want to know How to provide security against XSS.
*I made the tokens validity period very less, requiring the client to request with refresh_tokens and other client details for access_tokens. This makes it a bit safe but not entirely*.
I am concerned with the with client_id and client_secret being stolen since its present in the front-end javascript code and it being used by other client to validate. I am thinking of using a JWT for the client authentication, will this be helpful?
Data Sanitisation is another which I am confused about. There are modules like validator, express-validator which give regex validation. According to this blog post JSON Schema validations are fast. In the REST Api JSON will used for data exchange so I was wandering why can't I use modules like tv4 or any other JSON Schema validators for data validations?? I am not asking for suggestions to use what, I just want to know the basic difference in the kind of validations each provide and specially from a point of view of XSS protection and sanitisation.
So you have three separate questions here:
1) How to protect against XSS: As long as you use JSON to share data between the client & server and use standard libraries/methods for encoding/decoding JSON, you are mostly protected. After this, you only need to worry about DOM Based XSS, which is harder to be protected. But basically you need to be careful for not using any user supplied input that can be interpreted as anything other than "string" you intended. (please visit https://www.owasp.org/index.php/DOM_Based_XSS for more information)
2) client_id and client_secret being stolen: This does not seem to be possible in the way you require. In your scenario (where you distribute clientid&secret in javascript code) there is no way on server side to know whether the request is coming from your client or a fake one.
3) Data Sanitisation: I see two levels of sanitisation in the libraries you & blogpost mentioned. validator or express-validator is mostly used to validate individual data fields. Whereas others can validate a JSON object structure in addition to what "validator" does. If you require all exchanged data is in JSON format (as suggested for XSS protection as well) then you can use json object validators like tv4. (the only drawback of tv4 seems to be allowing latest json spec, which should not be a problem for you)
BTW: It would be easier if you specified your client application is purely client-side javascript (angularjs). I could not understand your question until I found this info in comments.
I have developed Restful Authentication System same as your case with NodeJS, MongoDB, ExpressJS in order to provide flexible authentication system for multiple clients like web, mobile. Let me summarize you the important points.
I have used html5 localstorage to keep user token after first time login by using login form. When user click login button, username and password sent to server and validated. After successfull validation, unique access token sent to client and stroed in local sotrage. If you have vulnerability on your client application, anyone can get your access token and make request by using your token. In order to prevent this, you need to use ssl connection for your app. This problem does not exists only restful auth systems, this can be happen in server side session storage. Let me explain this. I am using PHP for session. When user logs in, user session saved in to temp file on server and that session id sent to client browser. Somehow, if I can get that id, I can make request with header that contains someone's session id. When you compare, restful auth seems more flexible to me. I suggest you to ;
Use SSL connection prevent your access_token from to be stolen
Generate access token with powerfull encryption methods(SHA-256)
Small expire time for access_token the better
Implement a middleware for token validation for backend service usage. I mean make your requests like;
/use/update/{userid}
with custom headers contains your user token.
Design 5 attempt failed system for your backend. If user cannot success at 5 time try, this means someone tries to send random tokens in order to get in to system. Detect and block that IP
You can also deny requests other than browser clients.
Those are the informations that I have learnt while implementing the project.
tl;dr
I am considering a webservice design model which consist of several services/subdomains, each of which may be implemented in different platforms and hosted in different servers.
The main issue is authentication. If a request for jane's resources came in, can a split system authenticate that request as her's?
All services access the same DB layer, of course. So I have in mind a single point of truth each service can use to authenticate each request.
For example, jane accesses www.site.com, which renders stuff in her browser. The browser may send a client-side request to different domains of site.com, with requests like:
from internalapi.site.com fetch /user/users_secret_messages.json
from imagestore.site.com fetch /images/list_of_images
The authentication issue is: another user (or an outsider) can craft a request that can fool a subdomain into giving them information they should not access.
So I have in mind a single point of truth: a central resource accessible by each service that can be used to authenticate each request.
In this pseudocode, AuthService.verify_authentication() refers the central resource
//server side code:
def get_user_profile():
auth_token=request.cookie['auth_token']
user=AuthService.verify_authentication(auth_token)
if user=Null:
response.write("you are unauthorized/ not logged in")
else:
response.write(json.dumps(fetch_profile(user)))
Question: What existing protocols, software or even good design practices exist to enable flawless authentication across multiple subdomains?
I seen how OAuth takes the headache out of managing 3rd-party access and wonder if something exists for such authentication. I also got the idea from Kerberos and TACACS.
This idea was the result of teamthink, as a way to simplify architecture (rather than handle heavy loads).
I built a system that did this a little while ago. We were building shop.megacorp.com, and had to share a login with www.megacorp.com, profile.megacorp.com, customerservice.megacorp.com, and so on.
The way it worked was in two parts.
Firstly, all signon was handled through a set of pages on accounts.megacorp.com. The signup link from our pages went there, with a return URL as a parameter (so https://accounts.megacorp.com/login?return=http://shop.megacorp.com/cart). The login process there would redirect back to the return URL after completion. The login page also set an authentication cookie, scoped to the whole of the megacorp.com domain.
Secondly, authentication was handled on the various sites by grabbing the cookie from the request, then forwarding it via an internal web service to accounts.megacorp.com. We could have done this is a straightforward SOAP or REST query, with the cookie as a parameter, but actually, what we did was send a HTTP request, with the cookie added to the headers (sort of as if the user had sent the request directly). That URL would then come back as a 200 if the cookie was valid, serving up some information about the user, or a 401 or something if it wasn't. We could then deal with the user accordingly.
Needless to say, we didn't want to make a request to accounts.megacorp.com for every user request, so after a successful authentication, we would mark the user's session as authenticated. We'd store the cookie value and a timestamp, and if subsequent requests had the same cookie value, and were within some timeout of the timestamp, we'd treat them as authenticated without passing them on.
Note that because we pass the cookie as a cookie in the authentication request, the code to validate it on accounts.megacorp.com is exactly the same as handling a direct request from a user, so it was trivial to implement correctly. So, in response to your desire for "existing protocols [or] software", i'd say that the protocol is HTTP, and the software is whatever you can use to validate cookies (a standard part of any web container's user handling). The authentication service is as simple as a web page which prints the user's name and details, and which is marked as requiring a logged-in user.
As for "good design practices", well, it worked, and it decoupled the login and authentication processes from our site pretty effectively. It did introduce a runtime dependency on a service on accounts.megacorp.com, which turned out to be somewhat unreliable. That's hard to avoid.
And actually, now i think back, the request to accounts.megacorp.com was actually a SOAP request, and we got a SOAP response back with the user details, but the authentication was handled with a cookie, as i described. It would have been simpler and better to make it a REST request, where our system just did a GET on a standard URL, and got some XML or JSON describing the user in return.
Having said all that, if you share a database between the applications, you could just have a table, in which you record (username, cookie, timestamp) tuples, and do lookups directly in that, rather than making a request to a service.
The only other approach i can think of is to use public-key cryptography. The application handling login could use a private key to make a signature, and use that as the cookie. The other applications could have the corresponding public key, and use that to verify it. The keys could be per-user or there could just be one. That would not involve any communication between applications, or a shared database, following the initial key distribution.