There are four Laravel 7 projects, Project A going to use project B some routes. For sure project B needs to check if the request is authenticated or not. So there is a need to have a specific auth server for that handle authentication of all 4 projects.
Any idea or implementation of this pattern?
This is a very opinion based question, but what you're describing is similar to a microservice or distributed architecture for your projects.
If you need the ability to have one login or one point to authorise all of these projects, then one OAuth service sounds like the right way to go.
I am currently working across 12+ projects at the moment that are all interconnected, Laravel based, microservices.
We use one Laravel OAuth service to handle logins across all of the projects.
To login to Project A, you are directed to the Login service to login and redirected back.
For API or machine-to-machine communication, we still use the Login service. To do so, we setup a personal access client inside the Login service and then create user accounts for the projects. We then generate a personal access token for each project.
When Project A makes an API request to Project B, Project B takes the authentication token and verifies it is valid against the Login service by calling an api/user endpoint.
If they don't all need one single login, then you may be able to get away without one and just generate local API tokens that consuming projects need to use to authenticate against that project.
Related
I have two web applications that I've worked on in ASP.net Web API for a client. The two applications are hosted on the same domain, but in different virtual directories.
Each application has a data source with information about authorized users. I've been asked to set up a separate app that will allow a user to log in and issue an OAuth bearer token that can be used to access each web application.
I have the Owin stuff set up in one of my two apps with something of a boilerplate setup and it automagically makes the tokens work for me, but that's only within a single web application.
I have a general idea of how the OAuth tokens work: a user is authenticated and they receive a token that contains a user ID, a time stamp, and a cryptographic signature that says that the token came from my client's server and not somewhere else, and the Web API framework automatically checks that token for me when I decorate my API endpoints with the [Authorize] attribute.
I want to know how I can set this up so that the token is issued by one web app and consumed by another. I know that at a minimum, I need to make the two web apps work with the same key pair for the cryptographic signatures.
Can I please get some direction about how to make this happen?
For a Project I am working on I require some sort of SSO mechanism between applications. It doesn't have to be complex and it's not required that I use OAuth2. Right now I am unsing Laravel and am thinking of Sanctum. I know that Laravel Passport exists and am considering it too. This question is not really about the technology I need to use, but about understanding how a clean SSO workflow might look loke.
The concept I came up with looks like this:
I have a web-application A which has a database with users. This application is supposed to handle all authentication and provides an input mask.
I have a web-application B which wants to ask for authentication. The application B redirects to the input mask of application A.
If the user correctly enters his user data into this input mask on application A, the login is preserved on application A (probably through cookie) and a token is generated and passed to application B.
Later, when I create an application C, I can redirect to the same inpit mask on application An again, where the login is still preserved, so application A can pass an authorization token to application C without the user having to re-enter his login data.
I see one weird problem with this structure: In step 3, after the user inserts his login data, the token must be passed between two projects, and I can not guarantee that they are on the same domain or can share a database. So it seems the only way to pass the token, is to put it as URL parameter when project A on domain A redirects back to project B on domain B. Project B could then take the token from the URL.
But this seems like dirty programming to me, but I can't think of a better solution.
I have a Rest API developed with Spring Boot and neo4j as a database. There is no Frontend in the Spring Boot App. It only serves as a Backend. The Frontend is developed in Flutter.
In my app, the end user has to sign up and login with theis user credentials. The user management is currently handled with Spring Security and JWT, generating and storing the tokens with AuthenticationProvider, UserDetailsService and so on.
Now, we are migrating our whole infrastructure to Microsoft Azure. We already managed to get the DB, the Backend (as the Spring Boot App) and the Frontend there.
The question now is whether it makes sense to migrate the User Management to Azure Active Directory. Is this the right use case for that, or is Azure Active Directory actually there for other use cases?
Also, I want to use my Login and Signup Forms built with Flutter. I only found solutions so far where you get redirected to this Microsoft Login Form. I want to signup/login directly from my Flutter App, and then use the token for my requests in the Spring Boot App.
Does this even make sense? If yes, how can I realize that? I was searching for hours but I didn't find any proper solutions.
If you use AAD you will have to use the OAuth redirect based Microsoft login experience. There is no way around that.
If you can't think of any way you or your users will benefit by migrating to AAD, then there's no reason to do that. You're doing a bunch of work, and incurring risk, for no real benefit.
I'm in throws of moving our security architecture from ASP.NET Core Identity to Azure AD V2 with MSAL.js. We used a lot of Roles with the ASP.NET Core Identity implementation and the information was managed in the database using the web application. The pattern I'm abandoning is similar to this one.
https://www.dotnetcurry.com/aspnet-core/role-based-security
Azure AD with MSAL is working. The tokens are being created and passed and the local Web API Endpoints decorated with the generic [Authorize] attribute are being honored as you would expect. Web API Endpoints decorated with [Authorize(Roles= "Fee, Foo, Fi, Fum")] are throwing a 401 unauthorized error.
I'm not sure where to go from here. Do I write a CustomAuthorize attribute override for Web API and go back to the database and grab the roles. (probably match the DB defined roles to the user based on email address)
OR
Is there a way to implement roles natively with Azure AD V2?
I'm not sure whats the best course of action from here. Documentation and Code samples seem limited. It would sure be nice to just throw a AD User in a Group and have the Group be respected as a Role in the Web API. On the other hand, It's nice to have Role delegation handled within the confines of the Web Application.
Any advice, experience or interest would be greatly appreciated.
Answer
Follow up to my question. #Marc , You're correct, after looking at the token the Roles are not present. Adding Roles to the token seems pretty straight forward. You need to Patch the graph schema to include them, Configure the roles and assign them to users as needed thru AAD.
Or that's how it looks at first glance. After digging a deeper, it requires a P1 or P2 Enterprise license which only costs an additional 6$ per month per user. This will literally double the cost of hosting email in the cloud for us.
Alternatively I wrote a CustomAuthAttribute for WebAPI and tied User & Roles together on the server backend. Roles can still be managed via the web application and users can still login using Active Directory Credentials.
I recall that the id token returned in implicit flow (the one you use with JS) does not include app roles (or groups). I cannot find any docs confirming that but see others who got around the issue (so the issue must be there) by using Graph to get the roles (or groups).
You can capture the token you receive from AAD and view it using https://jwt.ms to see whether roles are included in it.
We are having a hosted web application and it uses forms authentication.
This webapplication is accessed by users belong to different partner organizations.
Currently users belong to the partner organizations are accessing the application using the credentials that we give it to them.
Now, some partner organizations wants their users to access the application using their active directory credentials.
We are planning to use ADFS for these partner organizations, so the users will be authenticated using Active Directory within their network and claims will be sent to the webapp via the Authentication token cookie set by the ADFS. From the claims, we map the users to the internal userIds of the web application.
My questions are , if we make the web application ADFS enabled,
1)Is it possible to still allow the other partner organization users(who don't want to use ADFS) to login to the web application using the existing login page(forms authentication)?
2) Should every page in the ADFS enabled webapp be accessed through https?
Any solutions or pointers would be much appreciated.
Thanks
-arul
Your app needs to require claims that describe the user, regardless of where they login from. It should not handle authentication in either case; this should be delegated to a trusted issuer, an STS. This will allow it to interact w/ users in a uniform way irrespective of where and how they authenticate. This means that you're going to need to use ADFS in two roles: that of an Identity Provider (IP) STS and of a Federation Provider (FP) STS. For users of partner companies that don't want to maintain users themselves, you'll be the IP-STS; for those that do, you'll be an FP-STS. In the latter case, ADFS will redirect users from your realm back to the partner's site where their IP-STS will authenticate them and send them to your FP-STS. It will map your partner's user ID and claims into ones that make sense in your realm. This and other information about the user will be included in the set of claims that are issued from your FP-STS. As a result, your app, only trusts your STS regardless of which scenario is appropriate for different users. Note that in this scenerio, there will be two STSs: your ADFS FP-STS and your partner's IP-STS, which may or may not be ADFS. In the other case, there will only be one STS: your IP-STS.
Not every page on your ADFS Web app needs to be accessed via HTTPS; however, everyone that's used in the authentication process should be.
This is really a non-trivial undertaking. If you want to talk about it more, please feel free to get in touch w/ me.