Read more about FF4J usage

• There are three different types of users which can be configured for feature toggle

  • Admin: These users can login into FF4J web console and perform all operations on feature toggles(create, update, delete, enable, disable and access auditing information using web console).
  • Editor: These users cannot access the FF4J web console but can perform operations such as create, update, delete, enable and disable feature toggle through FF4J REST API.
  • Read: These users can only read feature toggle values through FF4J REST API and cannot perform any other operations. They also cannot access web console.

💡 Currently feature toggle values can be read without authentication.

• Roles are dynamically assigned based on the configuration.

• Users needs to added to below section in application.yaml
• Depending on the section in which you add users they will be assigned roles.
• Username and password are set as environment variable through terraform configuration.

users:
  admins:
    - username: ${TEST_ADMIN_USERNAME}
      password: ${TEST_ADMIN_PASSWORD}

  editors:
    - username: ${TEST_EDITOR_USERNAME}
      password: ${TEST_EDITOR_PASSWORD}

• Create following secrets in Azure Key Vault named rpe-ft-api-{environment}:
  • admin-username-{servicename}
  • admin-password-{servicename}
  • editor-username-{servicename}
  • editor-password-{servicename}

Passwords needs to be in plain text in the vault and they will be hashed(BCrypt Password encryption) before saving it in database.

💡 Recommendation is to configure username using pattern [email protected] and a strong secured password. Usernames have to be unique hence do not use generic user names.

• Add below code in main.tf

  data "azurerm_key_vault_secret" "admin_username_{servicename}" {
     name       = "admin-username-{servicename}"
     vault_uri  = "${local.permanent_vault_uri}"
  }
  

  data "azurerm_key_vault_secret" "admin_password_{servicename}" {
    name       = "admin-password-{servicename}"
    vault_uri  = "${local.permanent_vault_uri}"
  }
  

  data "azurerm_key_vault_secret" "editor_username_{servicename}" {
     name       = "editor-username-{servicename}"
     vault_uri  = "${local.permanent_vault_uri}"
  }
  

  data "azurerm_key_vault_secret" "editor_password_{servicename}" {
    name       = "editor-password-{servicename}"
    vault_uri  = "${local.permanent_vault_uri}"
  }
  

    app_settings = {
      ADMIN_USERNAME_{SERVICENAME}  = "${data.azurerm_key_vault_secret.admin_username_{servicename}.value}"
      ADMIN_PASSWORD_{SERVICENAME}  = "${data.azurerm_key_vault_secret.admin-password_{servicename}.value}"
      EDITOR_USERNAME_{SERVICENAME} = "${data.azurerm_key_vault_secret.editor-username_{servicename}.value}"
      EDITOR_PASSWORD_{SERVICENAME} = "${data.azurerm_key_vault_secret.editor-password_{servicename}.value}"
    }  
  

• To make it work on AKS , Add the Azure Key Vault secret created in the previous steps along with the environment used to add users in bootstrap.yaml .

     spring:
     cloud:
       propertiesvolume:
         aliases:
           admin-username-{serviceName}: ADMIN_USERNAME_{SERVICENAME}
           admin-password-{serviceName}: ADMIN_PASSWORD_{SERVICENAME}
           editor-username-{serviceName}: EDITOR_USERNAME_{SERVICENAME}
           editor-password-{serviceName}: EDITOR_PASSWORD_{SERVICENAME}
  

• Also, Add secret key to values.template.yaml

       java:
         keyVaults:
           "s2s":
             secrets:
               - microservicekey-test-service
  

API and Web Console is restricted to certain user roles discussed above.

In order to see possible actions one should navigate to root (index) page where together with welcome message inspect the links to login/logout pages and ff4j web console.

NOTE. If logged in user does not have sufficient access levels to web console the server response will be parsed to 403 FORBIDDEN and, despite the fact of API allowance, user will only be able to logout.

API has Basic Authorisation configured which allows following action:

curl -X PUT -u username:password -d 'let us create some feature' localhost:8580/api/ff4j/store/features

The project uses Gradle as a build tool. It already contains ./gradlew wrapper script, so there's no need to install gradle.

To build the project execute the following command:

  ./gradlew build

Create the image of the application by executing the following command:

  ./gradlew assemble

Run the distribution (created in build/bootScripts/feature-toggle-api directory) by executing the following command:

  docker-compose up

This will start the API container exposing the application's port (set to 8580 in this template app).

In order to test if the application is up, you can call its health endpoint:

  curl http://localhost:8580/health

You should get a response similar to this:

  {"status":"UP","diskSpace":{"status":"UP","total":249644974080,"free":137188298752,"threshold":10485760}}

To skip all the setting up and building, just execute the following command:

./bin/run-in-docker.sh

For more information:

./bin/run-in-docker.sh -h

Script includes bare minimum environment variables necessary to start api instance. Whenever any variable is changed or any other script regarding docker image/container build, the suggested way to ensure all is cleaned up properly is by this command:

docker-compose rm

It clears stopped containers correctly. Might consider removing clutter of images too, especially the ones fiddled with:

docker images

docker image rm <image-id>

There is no need to remove postgres and java or similar core images.

This project is licensed under the MIT License - see the LICENSE file for details