example/

README

Playbook example

The following example shows a real use case of Blackbeard usage.

This example describe a technical stack composed of :

• 2 versions of an API;
• A front-end app.

Build my-app by yourself

If you want to fully test this example, you may want to build the docker images yourself. To do that, edit the my-app/Makefile file and change the following env var :

DOCKER_API_V1 = "seblegall/blackbeard-example-api:v1"
DOCKER_API_V2 = "seblegall/blackbeard-example-api:v2"
DOCKER_FRONT = "seblegall/blackbeard-example-front:v1"

Replace the docker images name using your own docker hub namespace.

Using blackbeard to deploy my-app

Requirement

You must have kubectl installed and configured and a Kubernetes cluster ready.

Tips: On MacOS and Windows, you may use the built-in Kubernetes cluster with docker-for-desktop

You must be located in the playbook directory :

cd my-app-playbook

Create a namespace using the v1 api

blackbeard create -n v1
blackbeard apply -n v1

Those commands will create a namespace called v1 and deploy the API (using the v1 version) and the front-end app in this namespace.

You may check that the api v1 is actually running with this command :

kubectl logs {api_pod_name} -n v1

Test api v2 in a different namespace

Now, if you'd like to run the API v2 in a different namespace, you may run :

blackbeard create -n v2

Edit the inventories/v2_inventory.json file and change the version value. This file now should look like :

{
    "namespace": "v2",
    "values": {
        "api": {
            "version": "v2"
        },
        "front": {
            "version": "v1"
        }
    }
}

Finally, apply the changes :

blackbeard apply -n v2

Now, you can check that v2 is actually deployed by running the following command :

kubectl logs {api_pod_name} -n v2