I've built an application that uses jQuery and JSON to consume an ASP.NET .asmx web service to perform crud operations. The application and .asmx are on the same domain. I dont mind people consuming the read operations of the .asmx remotely but dont want people randomly deleting stuff!!!
I can split the methods i'd like to be publicly accessible and the 'hidden' ones into 2 web services. How can I lock calls to the 'hidden'.asmx web service to the same domain that its hosted in?
Thanks in advance.
Edit:
Can someone comment on this, seems plausible ( source: http://www.slideshare.net/simon/web-security-horror-stories-presentation ):
Ajax can set Http headers, normal forms cant.
Ajax requests must be from the same domain.
So "x-requested-with" "XMLHttpRequest" requests must be from the same domain.
There are two scenarios you need to secure with web services:
Is the user authenticated?
Is the action coming from my page?
The authentication piece is already taken care of if you're using Forms Authentication. If your web service sits in a Forms Authentication-protected area of the site, nobody will be able to access your web services unless they're logged in.
The second scenario is a slightly trickier story. The attack is known as CSRF or XSRF (Cross Site Request Forgery). This means that a malicious website performs actions on behalf of your user while they're still logged in to your site. Here's a great writeup on XSRF.
Jeff Atwood sort of sums it all up in the link above, but here is XSRF protection in four steps:
Write a GUID to your user's cookie.
Before your AJAX call, read this value out of the cookie and add it
to the web service POST.
On the server side, compare the FORM value with the cookie value.
Because sites cannot read cookies from another domain, you're safe.
In AJAX the browser makes the calls, so even if you were to check that the domain is the same it wouldnt be secure enough because it can easily be faked.
You need to use some sort of authetication/autharization tokens (preferably with a time out) to keep things safe.
Quick and dirty solution would be to use IP address restrictions to allow only your domain's IP address access via IIS.
Probably better would be using HTTP authentication. There are many ways to do this, I found Authentication in ASP.NET Web Services a helpful overview.
Related
These are the criteria:
My users are unauthenticated.
My app is a single page web app (written in JavaScript) that uses RESTful web API endpoints to fetch data from my own server.
Objective: I'd like to constrain the web API to only allow requests from my own app, without requiring users to authenticate. Are there existing techniques for this?
The issue is that since the web app's code and the requests made by it are transparent to the client it can't hold a secret for authenticating with the server.
I thought about creating single-use access tokens on the server side, then use those from the client side for the API calls. Now this would make it harder to do requests from the outside, but still you could make requests if you fetch this token from the original website first (even if you'd have to do it for every request).
If it matters I'd implement this with an ASP.NET MVC website and ASP.NET Web API endpoints.
Similar to this question but not entirely the same.
Thanks in advance.
I asked pretty much the same question a couple of years ago, and for unauthenticated users, some kind of token is pretty much going to be your only option. It doesn't make it impossible to get unauthorized access to your API, but does make it a little more painful.
I took a slightly different approach in my solution, using 2 cookies to protect some anonymous GET requests. One of them was the anonymous identification cookie, which was used as an encryption salt for another custom cookie given to the page that delivered the javascript. Only when the API request received both of these cookies and the decrypted result was satisfactory did I allow the WebAPI to respond. But like I mentioned, this only makes it more painful to gain access outside of the javascript app, not impossible.
Here is a reference for using tokens in WebAPI: http://codebetter.com/johnvpetersen/2012/04/02/making-your-asp-net-web-apis-secure/
I am very new in web api security. I have used form authentication technique. when user logs in, a token is created and stored as a cookie in user's web browser. On each request the token is varified and if user is authenticated and authorized user is given access to the service.
but I think this approach does nothing in web api security. cookies can easily be copied and pasted in other browser and anyone can get the service.
I am thinking to use App key and secret along with form authentication. I am not suggested to use third party service like Oauth for authentication. I am not Sure about the Implementation of app key and secret that how it exactly works.
Please provide better way to secure my web api wihtout using third party services and to prevent cookie hijacking etc. What actions are performed to build a strengthly secure web api.
The forms authentication is good enough. You can also do the following:
Use anti-forgery (antifrogery) tokens. Check this or this
It will also be great if on sensitive actions you check if the call to the function was made from the same site or not.You can implement your own action filter for this. (check if the referral site is your site, or the expected site)
Edited:
Thanks guys for your comments. I guess you are right. Well authentication cookies in ASP are created as httpOnly cookies which means even if the site had some XSS vulnerabilities it will still be safe and cant be stolen. I would also suggest to use https everywhere if the site is used for sensitive operations (like a bank) to make sure the cookies are perfectly safe.
I have a bunch of websites that are setup identically to use a WIF identity provider. I've recently moved the business logic out of the web applications and into a Web Api service application. This runs in a different virtual directory to the other sites. The idea being that browser will put the data into the page AJAXy.
The issue I have is with securing the web API. It seems that WIF single sign-on works okay with traditional sites. The user can access one website, get redirected to the identity provider, login and get redirected back to the website they wanted. When they access another site they also get redirected back to the identity provider but needn't log in as a FEDAUTH cookie exists so they automatically get authenticated and redirected to the second site.
This doesn't work for the Web Api scenario because when the browser perhaps makes a GET to it, the Api will return a redirect to the calling javascript when it should be expecting JSON.
Is it even possible to secure Web Api with WIF?
Not sure whether I got you right, but it seems like the main problem is that javascript/ajax does not support http redirects.
A possible solution could be to simulate the redirection with a sequence of seperate calls in ajax:
Check whether you are authenticated on your web api site (by a dummy ajax call).
If this is not the case:
Call your sts over ajax and grab the security token out of the "wresult" form field.
Call the login site on your web api site and pass the security token as "wresult" data.
Dominick Bayer wrote a few blog posts about securing rest services. For further reading have a look at http://www.leastprivilege.com/. (Especially http://leastprivilege.com/2009/09/11/adding-a-rest-endpoint-to-a-wif-token-service/ and
http://leastprivilege.com/2010/05/05/thinktecture-identitymodel-wif-support-for-wcf-rest-services-and-odata/).
The following presentation from TechDays might also be interesting: http://www.microsoft.com/showcase/sv/se/details/ffc61019-9756-4175-adf4-7bdbc6dee400 (starting at about ~ 30 minutes).
Scenario
I have an HTML/javascript website that uses javascriptSOAPClient communicate with an ASP.NET 1.1 web service in order to read/write to a SQL database. (http://www.codeproject.com/KB/ajax/JavaScriptSOAPClient.aspx). The database contains anonymous demographic information--no names, no credit cards, no addresses. Essentially the data collected is for data mining purposes.
The site is live, but we want to introduce a more secure communication between the javascript/ajax client and the wbe service for both this and future projects. Working as contractors in the financial industry, at some point we're going to get nailed with the question: is this website hackable? If we don't have a solution we could be out on our ears.
I am already following best practices such as communicating with the database via command parameters and stored procedures). However, currently anyone could browse to our web service description and figure out how to consume our exposed services.
Questions
With my hybrid solution (i.e. not end-to-end Microsoft) how should I go about authenticating client requests on the web service?
If I start passing a username/password or some other identifiable element into the web service as authentication, should I be concerned about how that key is generated/stored on the client side?
A few suggestions to consider:
List the threats, and compare each to your current setup.
Use SSL / HTTPS. This alleviates a whole class of vulnerabilities.
Use username/password, generated on the server side and sent out of band (in the post or by phone) to the user. (Hope this answers question 2).
Use 2-factor authentication. To do this, you can look at security tokens such as RSA's keyfob-type gizmos or look at Steve Gibson's Perfect Paper Passwords
The easiest solution from a programming standpoint is to use two way HTTPS. That is, the server presents a certificate to the client, and the client presents a certificate to the server. Then only clients with proper certs (issued by you) can connect.
That helps reassure clients that your site is not generally accessible, yet the security is transparent to the application and, once they've signed up and received a cert, to them. The downside is that you have admin overhead in issuing and tracking the user certs -- but that's probably less than you'd have dealing with username/password combos.
There are a few simple options:
SSL + Cookie
If the web app is also ASP.NET and hosted along with your web service, then you should have access to the User/Membership/Session of the web app inside your web service (essentially #1, but you get it without doing any work).
If the web app and web service are not on the same domain, then cookies are out due to cross-domain issues - so you can have the web app embed a GUID into a hidden form field, and use that GUID as a sort of cookie (and it will need to be passed as a parameter on all web service requests).
Can you incorporate a certificate authentication mechanism? So that only clients that have keys you can verify can communicate? That's how the product I work with has its managed devices communicate back to the core.
I am building a web based application written in ASP.NET and Flex. One of my biggest challenges is implementing security for the application in a flexible and maintainable way. This challenge is compounded when different technologies are involved. I'll try to describe what I have below.
The website is laid out as follows:
/mydomain.com/
Login.aspx
Default.aspx (hosts flex [.swf] application)
/Administration/
AddUsers.aspx
AddRoles.aspx
AddPermissions.aspx
etc...
/Services/
SecurityService.asmx
MapService.asmx
PhotoService.asmx
etc...
I am currently using forms authentication to secure the resources on the website. All pages/resources not in the /Services/ folder require an authenticated user and will be redirected to Login.aspx if they are not yet authenticated. The .asmx pages allow unauthenticated users. To secure these resources I throw an exception in the SOAP method. This avoids redirecting pages for SOAP web services which is not supported by any SOAP web service clients I am aware of. Finally, SecurityService.asmx contains a Login method to allow the Flex application to Login without redirecting to the Login.aspx page should the cookie expire for any reason. Because the cookie established is sent with any request to the server, including requests coming from the Flex application, this seems to work pretty well.
However, this still feels like a bad approach securing web services. I feel like I am using Forms Authentication for something it was not intended for. Specifically, I am concerned about:
This model will not work when the services are separated from the core website. This is a newly discovered requirement and I believe that Forms Authentication is not going to work well (if at all) without a lot more modification and trickery.
Clients other the Flex may require access to the services. Some of these clients may not even be able use cookies. If so, this model immediately falls apart. This is not an immediate requirement but it is known that this is one of the long term goals.
We will eventually (hopefully sooner rather than later) move to a REST based architecture (vs. SOAP) so any solution needs to work for SOAP and REST.
So, my question is.
What are the best authentication and authorization mechanisms for securing an application built on ASP.NET, Flex, and SOAP or REST web services?
Note: I am actively looking into OAuth; however, I am having a difficult time finding complete examples from which to learn. Additionally, I need to be able to filter the data returned for a given user based on the permissions that user has, OAuth seems to remove the identity of the user from the token. As such, I am not sure how OAuth applies in a fine grained security model.
Others may disagree, but I actually don't see a huge problem with handling it the way you are now; that's probably how I'd handle myself, at least initially. One way or another, even down the road, you'll probably want to keep the Flex app aware of the authentication state of the session, so if that means checking the ASP.NET session token directly, or some other cookie you set when you set that one, it seems like a fine and reliable way to go.
I do see what you mean about the services redirecting, but even so, with forms auth, it's not so much the service specifically that's handling the redirecting so much as the ASP.NET app itself. Later, should you need to use a different authentication scheme, you can consider that scheme's specific implementation considerations. Unless you've got concerns about using forms auth in general, there's probably no need complicate your approach simply because of the Flex client and Web services.
I admit I don't work with web services much, but what about requiring an access key as a soap header parameter? Any client app which can communicate with a soap web service is likely to have a low level API to modify the soap request, and use of the access key allows you to (in theory) limit the use of the service. Google, Amazon, and several other providers use this type of authentication for their web services and it seems to work very well.
This article seems like it might be a good place to start...
The WCF Security Guide published on CodePlex may help you there, if you are using, or can use WCF.
There's also Microsoft's Web Services Enhancements (WSE) 3.0 which I believe implements some of the WS-* security specifications.
Hope that helps.
If you move your services to another place, then the standard ASP.net authentication cookie can be re-used if both web apps have the same machineKey in the web.config.
As far as I know, FLEX will honour the asp.net authentication cookies because it will make http requests through the browser, which will pass the http cookies (including the asp.net authentication ticket) like a normal http request.
Have you tried securing your website and services using normal asp.net authentication yet?
I think it's best to have independent authentication systems - even if there are relations between the user and the auth tokens on the back end. They are different beasts that have differing capabilities and demands.
Use the normal forms based auth for the flex portion. That is good.
For web services, you could create a login method that returns some auth token which is used by subsequent tasks to execute. Or add a few fields to your web services (posted in the header or as params) to use a userid/password combo for authentication each and every time.
A side note: I wouldn't rely on a soap exception to handle authentication problems. But you wouldn't need to worry about the redirection if you send an auth token or user/pass with the WS requests.
EDIT:
RE: Comment-
Ideally there is. There are products out there (Tivoli access manager) that service those needs, but they are expensive.
I gave this recommendation because it eases the pain of allowing access to alternative clients and as long as you designed the services correctly it's stateless. It also gives you finer grained control over data level access on the service side of things.
See Web Services authentication - best practices?
Dave Dunkin wrote:
The easiest way to handle it across a
variety of platforms is to use HTTP
basic authentication and HTTPS for the
transport layer. WS-Security would be
good if your needs go beyond simple
username/password but the support is
going to vary quite a bit between
platforms. HTTP authentication is
supported by every decent SOAP
implementation.
and my Custom HTTP Basic Authentication for ASP.NET Web Services on .NET 3.5/VS 2008.