Next, we'll use the Ansible k8s module to leverage existing Kubernetes and OpenShift Resource files. Let's take the nginx deployment example from the Kubernetes docs.

Note: We've modified the resource file slightly as we will be deploying on OpenShift.

a. Copy the nginx deployment definition nginx-deployment.yml into example-role/templates, adding a .j2 extension

cp nginx-deployment.yml ./example-role/templates/nginx-deployment.yml.j2

$ cat ./example-role/templates/nginx-deployment.yml.j2

kind: DeploymentConfig
apiVersion: v1
metadata:
  name: nginx-deployment
spec:
  template:
    metadata:
      labels:
        name: nginx
    spec:
      containers:
        - name: nginx
          image: nginx:1.15.4
          ports:
          - containerPort: 80
  replicas: 3
  selector:
    name: nginx

b. Update tasks file example-role/tasks/main.yml to create the nginx deployment using the k8s module

---
- name: set test namespace to {{ state }}
  k8s:
   api_version: v1
   kind: Namespace
   name: test
   state: "{{ state }}"

- name: set nginx deployment to {{ state }}
  k8s:
   state: "{{ state }}"
   definition: "{{ lookup('template', 'nginx-deployment.yml.j2') }}"
   namespace: test
 

c. Run the Playbook to deploy nginx onto OpenShift

Running the Playbook with the command below will read the state variable defined in example-role/defaults/main.yml

ansible-playbook -i myhosts playbook.yml

d. Examine Playbook results

You can see the test namespace created and the nginx deployment created in the new namespace.

oc get all -n test

Next, let's make it possible to customize the replica count for our nginx deployment** by adding an nginx_replicas variable to the DeploymentConfig template and filling the variable value dynamically with Ansible.

a. Modify vars file example-role/defaults/main.yml, setting nginx_replicas: 2

---
state: present
nginx_replicas: 2

b. Modify nginx deployment definition nginx-deployment.yml.j2 to read replicas from the nginx_replicas variable

kind: DeploymentConfig
apiVersion: v1
metadata:
  name: nginx-deployment
spec:
  template:
    metadata:
      labels:
        name: nginx
    spec:
      containers:
        - name: nginx
          image: nginx:1.15.4
          ports:
          - containerPort: 80
  replicas: "{{ nginx_replicas }}"
  selector:
    name: nginx
 

c. Run the Playbook to change the nginx replica count

Running the Playbook again will read the variable nginx_replicas and use the provided value to customize the nginx DeploymentConfig.

ansible-playbook -i myhosts playbook.yml

d. Examine Playbook results

After running the Playbook, the cluster will scale down one of the nginx pods to meet the new requested replica count of 2.

oc get pods -n test

e. Tear down the nginx deployment

To remove the nginx deployment, we'll override the state variable to contain state=absent using the -e / --extra-vars flag.

ansible-playbook -i myhosts playbook.yml --extra-vars state=absent