Daniela Field on August 25, 2014
Mendix applications are implemented by a large variety of companies to support numerous and varied business processes. All these different Mendix users share the critical need for their applications to be secure and accessible.
The Mendix App Platform handles key security aspects out of box. For example, measures against front-end security threats, such as Cross Site Scripting, and server-side security threats, such as SQL Injection and Code Execution, are provided by the platform.
Mendix developers do not need to take these technical security aspects into consideration when building Mendix apps, as the platform handles this as a service. Obviously, this does not mean that developers do not have to consider security at all. Application-level authorization and access rights need to be configured in the model by the developer. This blog post gives you a set of best practices to manage application-level security and do it right from the very start of your project.
Application security has never been easier to manage within the Mendix App Platform. With just a few clicks, users can see only their own relevant information and specific parts of the applications. Furthermore, with Mendix, security can be as granular as the business users need it to be.
As you may know, building an app in Mendix takes place in a project. In the project there are 3 types of security levels:
The Project security oversees the modules security and you configure the user roles that you wish to assign to the end users. Each user role will have a number of underlying module roles. Advantages to having module specific user roles include the ease of importing and exporting the modules between projects. Furthermore, it’s easier to focus on small parts of the app instead of having everything in together.
Within the access rules, there are options on what an end user can do for each entity. For the application in my example, we have a grants funds entity and 3 module roles of Administrator, User and Anonymous User. Then we have the access rights and XPath constraints for the entity.
For each user role, we can grant different access levels. For example, we can allow the anonymous users to create a new grant or be able to delete their own grant. For anonymous users, it is always important to add a restriction using XPath constraint, which allows them to edit their own grant only.
We can go down into more details and allow the user roles to only read or write for a specific attribute.
It is best practice to start applying security at the domain model for various reasons highlighted below:
Always add XPath constraints to determine which level of security should be applied to each object. For example, a business analyst should be able to see only the reports they created instead of the reports created by other company business analysts.
Whenever you are creating an administrator role, it is important for the administrator to ONLY serve as an administrator. That means any user that has the administrator role should be able to do “administrative” tasks such as creating new users, managing email settings, and monitoring an application. An administrator role should NOT have access to the application at large.
For example, if your company has sensitive data such as finance transactions in a financial app, or physician and patient information in a clinical trial- or healthcare app, the administrator should not be able to view that information.
As you are building your application, it is important to keep security in the back of your mind. Always think of who should have access to what. If you are always implementing the security as you are building the application, it will save time when testing. Furthermore, as the application gets more complicated, it can take a while to go back into each page, microflow and entity and determine the correct access and permissions. Again as you are building the domain model and your database – add the security.
As you build pages and microflows – add the security. Each page and microflow will have the Visible for property. Adding the user roles here allows for the users with the specific userrole to see the page or microflow. If the page does not have a user role assigned, the Mendix Business Modeler will let you know in the Errors log.
Again, always start from the bottom (the Domain Model) and work your way up and always be wary of the user roles. For more information on security, user roles and other implemtation questions use our Mendix forum and documentation and don’t hesitate to leave a comment below.
For more information on the security aspects of the Mendix App Platform, download our whitepaper, “Security for Cloud & On Premise Deployment.”
Receive Mendix platform tips, tricks, and other resources straight to your inbox every two weeks.