RBAC is a popular way to enforce user access constraints because it allows for granular permission grants that match each individual’s responsibilities. Many systems have a requirement that some users can create data, others can review it, and administrators get uninhibited access. RBAC can implement all these requirements.
What Are RBAC Roles?
An RABC role is a collection of permissions. Permissions represent the unique actions that users can take in your system. They can be as specific as you need. Your codebase should gate protected endpoints using a permission check.
Roles aggregate permissions together so they can be more conveniently allocated to users. Here’s a quick example of the difference between a role and a permission:
Permissions
Create new article. Publish an article. Delete an article.
Roles
Author: Permission 1. Editor: Permission 1 and Permission 2. Admin: All permissions.
Users
User A: Author role. User B: Editor role.
This model means User A can only create articles while User B can create and publish.
In a real-world application you could have many more permissions. Assigning them directly to users would be tedious. The role concept neatly bridges between precise permission checks and clear expression of user responsibilities.
Most RBAC implementations allow users to have any number of roles. Roles can overlap when needed. If the same permission exists in two of the user’s roles, their account will still satisfy the in-code access control checks.
Implementing RBAC
RBAC can be relatively complicated to implement into a system. You need to assess how many permissions you need, the ways in which they’ll be enforced, and how roles will be created and allocated to users. You might be able to cope with a fixed number of roles to begin with but many larger applications will allow users to create their own roles for specific use cases.
It’s best to carefully plan out your requirements before you embark on integrating RBAC. Narrowing the scope of your implementation as far as possible can help to reduce complexity. Start with the bare minimum access controls that your application can’t function without, then layer in additional roles and controls over time.
Another side of RBAC is when roles apply with respect to a specific resource. As an example, a user might be allowed to submit new articles to the “Blog” category but not “Case Studies.” Now your permission checks will need to consider the category that the user’s interacting with. You could manage this flow by dynamically creating new permissions for each category and assigning them a predictable identifier.
You might be able to simplify your permission checks by using an external system for identity management and authorization. Policy-based access control systems that support RBAC, such as Auth0 and Cerbos, can help you set up complex authorization logic without extensively modifying your own code. These platform can also offer a simpler route to achieving resource-bound permission checks.
An external system is often overkill for smaller projects that are working with a limited number of roles and permissions. In these cases you can build an RBAC implementation from a couple of database tables: one that associates permissions with roles, and another that links roles to users. To check whether a user can perform an action, iterate over their roles and see if any role includes the required permission.
RBAC Drawbacks
RBAC provides increased security and more flexible permission grants when implemented correctly. However it also comes with some drawbacks that should be acknowledged before you use it to protect your system.
Arguably the most significant of these is the ease with which RBAC-based apps can become cluttered with unused roles and duplicate permissions. Roles should only be created when they satisfy a new requirement or mirror a distinct user responsibility. Having too many roles will make your system harder to maintain and reduce the visibility into the grants that each user requires.
It’s important to regularly review role-to-user allocations too. Roles should be precise so users can be given the minimum set of roles they require for their work. Allocating too many roles, or putting a large number of permissions inside each role, can cause accounts to become over-privileged. This increases the risk to your application if an account is compromised.
RBAC also requires upfront awareness of how your system will function and where boundaries between responsibilities occur. Trying to implement RBAC without this understanding will normally be sub-optimal because your permissions and roles will either be too broad or tediously precise. The size and shape of your RBAC solution should normally reflect how your internal operations work.
Summary
Role-Based Access Control is one of the most common forms of user access constraint in software applications. It lets you set up granular permissions that are then combined into roles for assignation to users. The approach provides a high degree of flexibility and customization but can also lead to bloat if you don’t actively audit which roles are used.
Alternatives to RBAC include access control lists, which act as rules that grant or deny access based on conditions, and attribute-based access control (ABAC) that protects access based on attributes of the requesting user. ABAC can provide even greater flexibility when users have many roles but RBAC is a better choice when you want your access control system to closely represent your organization’s structure.
