Since this topic became very extensive, I decided to split up the blog into 4 parts. To keep blog lengths manageable. Here is the split up
But don’t worry, all these 4 parts will be released within the same week so that those people that are eager to process it in one go, do not need to wait a long time before the series is published.
As the demand for secure and efficient authentication and authorization mechanisms grows, JSON Web Tokens (JWT) have emerged as a favored choice for developers. JWT tokens provide a modern approach to verifying user identity and defining access privileges within web applications. In this blog post, we will delve into the usage of JWT tokens across various frameworks, namely Spring Boot, Quarkus, Jakarta, and Kotlin Ktor. By comparing their implementation approaches, we aim to provide insights into how JWT tokens are utilized within each framework and help you make a transition from one to another easier.
Understanding the Basics of JWT Tokens
At the core of JWT tokens lies a simple yet powerful structure that encompasses all the necessary information for secure authentication and authorization. Let’s dive into the basics of JWT tokens and explore their three essential components: the header, the body, and the signature.
The header of a JWT token contains metadata about the token itself and the algorithms used to secure it. It typically consists of two parts: the token type, which is always “JWT,” and the signing algorithm employed, such as HMAC, RSA, or ECDSA. This header is Base64Url encoded and forms the first part of the JWT token.
2. Body (Payload):
The body, also known as the payload, carries the actual data within the JWT token. It contains the claims, which are statements about the user and additional metadata. Claims can include information like the user’s ID, name, email, or any other relevant data. The payload is also Base64Url encoded and forms the second part of the JWT token.
The signature is the crucial component that ensures the integrity and authenticity of the JWT token. It is created by combining the encoded header, the encoded payload, and a secret key known only to the server. The signature is used to verify that the token has not been tampered with during transmission or storage. It acts as a digital signature and prevents unauthorized modifications to the token. The signature is appended as the third part of the JWT token.
Self-Contained and Secure
One of the significant advantages of JWT tokens is their self-contained nature. Since all the necessary information is embedded within the token itself, there is no need for additional database queries or session lookups during authentication and authorization processes. This inherent characteristic contributes to improved performance and scalability.
To verify the authenticity and integrity of a JWT token, the recipient needs access to the public key or shared secret used to generate the signature. By retrieving the public key or shared secret, the recipient can verify the token’s signature and ensure that no tampering or unauthorized modifications have occurred. This mechanism provides a robust security layer, assuring that the token’s contents can be trusted.
User Roles in JWT Tokens
JWT tokens can also include user roles as part of their payload. User roles define the permissions and privileges associated with a particular user. By including this information in the JWT token, applications can determine the user’s authorization level and grant or restrict access to specific resources or functionalities accordingly. This granular approach to authorization allows for fine-grained control over user permissions within the application.
In the upcoming sections, we will explore how different frameworks incorporate these fundamental JWT token concepts into their authentication and authorization workflows. Understanding the core principles behind JWT tokens sets the stage for a comprehensive comparison, enabling us to evaluate the strengths and nuances of each framework’s implementation.
The same example application is made with different runtimes. It contains a couple of endpoints, they all require a valid token before they should be executed. One of the endpoints requires that the token contains the role of administrator.
GET /protected/user -> Hello
GET /protected/admin -> Protected Resource; Administrator Only
The tokens utilised in our example are sourced from Keycloak, a reliable and widely adopted Authorization provider. Keycloak offers various standard flows for obtaining these tokens, catering to diverse authentication scenarios.
One of the commonly employed flows is the authorization code flow, which involves user interaction through dedicated screens provided by the Authorization provider. Users are prompted to log in and provide their credentials, following which Keycloak generates the necessary tokens for authentication and authorization purposes.
Alternatively, Keycloak supports a username and password-based approach where users can submit their credentials to a designated endpoint. This method allows Keycloak to validate the provided information and issue the relevant tokens required for subsequent authentication and authorization processes.
For our example, a custom realm with a configuration that is suitable for all our runtimes is created by setup_jwt_example.py and can be found in the directory https://github.com/rdebusscher/Project_FF/tree/main/jwt/keycloak. The script prepares the realm and a OpenId Connect client so that in response to a valid user name and password combination, a JWT token with the roles of the user is returned. It creates also two users, one of them having the admin role.
The Python script test_jwt_example.py can be used to test out the solution in each of the runtimes. It calls both endpoints with the two users that are defined. And so, one of the calls will result in an error since the non-administrator user is not allowed to call the administrator endpoint.
The different runtimes are discussed in part 2 and part 3 of this series.
Training and Support
Do you need a specific training session on Jakarta EE, Quarkus, Kotlin or MicroProfile? Have a look at the training support that I provide on the page https://www.atbash.be/training/ and contact me for more information.