Lab 2: Create the Lab Operator Project

Create the Lab Operator Project

In this part of the lab we will create a demo Ansible operator and deploy it to our Cluster.

Create the ansible-operator-frontend directory

cd
mkdir ansible-operator
cd ~/ansible-operator

Create the ansible-operator-frontend Project

operator-sdk new ansible-operator-frontend --type=ansible --api-version=ansiblelab.ibm.com/v1beta1 --kind=MyAnsibleLabDemo 
   
> INFO[0000] Creating new Ansible operator 'ansible-operator-frontend'. 
> INFO[0000] Created deploy/service_account.yaml          
> INFO[0000] Created deploy/role.yaml                     
> INFO[0000] Created deploy/role_binding.yaml             
> INFO[0000] Created deploy/crds/lab_v1beta1_MyAnsibleLabDemo_crd.yaml       
...

You can ignore the errors concerning git

Change to the ansible-operator-frontend directory

cd ~/ansible-operator/ansible-operator-frontend

The basic file structure for the Ansible Operator has been created

ll
   
> total 36
> drwxr-x--- 7 training training 4096 Jul  1 17:29 ./
> drwxrwxr-x 3 training training 4096 Jul  1 17:29 ../
> drwxr-x--- 3 training training 4096 Jul  1 17:29 build/
> drwxr-x--- 3 training training 4096 Jul  1 17:29 deploy/
> drwxrwxr-x 7 training training 4096 Jul  1 17:29 .git/
> drwxr-x--- 5 training training 4096 Jul  1 17:29 molecule/
> drwxr-x--- 3 training training 4096 Jul  1 17:29 roles/
> -rw-r--r-- 1 training training  140 Jul  1 17:29 .travis.yml
> -rw-r--r-- 1 training training  121 Jul  1 17:29 watches.yaml

Create the Lab Operator API

With the above, the API has already been created (unlike for GO operators) and added to the new Custom Resource (CR), with APIVersion ansiblelab.ibm.com/v1beta1 and Kind MyAnsibleLabDemo.

Add the deployment.image field to the Custom Resource

Training VM

gedit ~/ansible-operator/ansible-operator-frontend/deploy/crds/ansiblelab.ibm.com_v1beta1_myansiblelabdemo_cr.yaml &

Cloud/Standalone

nano ~/ansible-operator/ansible-operator-frontend/deploy/crds/ansiblelab.ibm.com_v1beta1_myansiblelabdemo_cr.yaml

Add the last two lines from the following at the end of the file and save to make it look the same

apiVersion: ansiblelab.ibm.com/v1beta1
kind: MyAnsibleLabDemo
metadata:
  name: example-MyAnsibleLabDemo
spec:
  --> Add fields here
  size: 3
  demo:
    image: niklaushirt/k8sdemo:1.0.0

Save and quit

**We have now finished setting up the API and CRDs **

Create the Lab Operator Controller

By default, an Ansible role executes the tasks defined at roles/tasks/main.yml. For defining our deployment we will use the k8s module of Ansible.

1) Define the Lab Operator Controller

Edit the ansible-operator-frontend Controller

Training VM

gedit ~/ansible-operator/ansible-operator-frontend/roles/myansiblelabdemo/tasks/main.yml &

Cloud/Standalone

nano ~/ansible-operator/ansible-operator-frontend/roles/myansiblelabdemo/tasks/main.yml	```

Replace the content with the following:

- name: Create the k8sdemo deployment
  community.kubernetes.k8s:
    definition:
      apiVersion: apps/v1
      kind: Deployment
      metadata:
        name: k8sdemo
        namespace: default
      labels:
        app: k8sdemo
      spec:
        selector:
         matchLabels:
           app: k8sdemo
        replicas: 1
        template:
          metadata:
            labels:
              app: k8sdemo
          spec:
            containers:
            - name: k8sdemo
              image: "{{demo.image}}"
              imagePullPolicy: IfNotPresent 
              ports:
              - containerPort: 3000
              env:
                - name: PORT
                  value : "3000"
                - name: APPLICATION_NAME
                  value: k8sdemo
                - name: BACKEND_URL
                  value: http://k8sdemo-backend-service.default.svc:3000/api
   
- name: Create the k8sdemo service
  community.kubernetes.k8s:
    definition:
      apiVersion: v1
      kind: Service
      metadata:
        name: k8sdemo-service
        namespace: default
      spec:
        selector:
          app: k8sdemo
        ports:
          - protocol: TCP
            port: 3000
            targetPort: 3000
            nodePort: 32123
        type: NodePort

This will ensure that the Pod will be created with the Image information defined in the Custom Resource (CR) definition ("{{deployment.image}}"). This picks up the value defined in the CR.

Save and Quit

2) Build the Lab Operator Controller

Now let’s build the Operator container

operator-sdk build localhost:5000/ansible-operator-frontend:ansible
     
> INFO[0000] Building OCI image localhost:5000/ansible-operator-frontend:ansible
> Sending build context to Docker daemon  49.15kB
> Step 1/3 : FROM quay.io/operator-framework/ansible-operator:v0.10.0
>  ---> 168416e214f1
> Step 2/3 : COPY watches.yaml ${HOME}/watches.yaml
>  ---> Using cache
>  ---> 43f81409e05d
> Step 3/3 : COPY roles/ ${HOME}/roles/
>  ---> 0ad354c77a7a
> Successfully built 0ad354c77a7a
> Successfully tagged localhost:5000/ansible-operator-frontend:ansible
> INFO[0001] Operator build complete.

Where localhost:5000/ansible-operator-frontend:ansible is the name of the Docker image to be created.

And push the Operator container to the local registry

Training VM

First execute this in a separate Terminal Window/Tab in order to be able to access the private registry:

kubectl port-forward –namespace kube-system $(kubectl get po -n kube-system | grep kube-registry-v0 | \awk ‘{print $1;}') 5000:5000

> Forwarding from 127.0.0.1:5000 -> 5000
> Forwarding from [::1]:5000 -> 5000


 ![](../images/cloud.png) **Cloud/Standalone**
 
 ![](../images/important.png) **Cloud/Standalone** This does probably not work if you are using a standalone or cloud environment.
 You can ignore this as we will be using a version already provided on Docker hub.




5. And then push the image:

 ![](../images/vm.png) **Training VM**
 
 ```bash
 docker push localhost:5000/ansible-operator-frontend:ansible
 ```

 ![](../images/cloud.png) **Cloud/Standalone**
 
 ![](../images/important.png) **Cloud/Standalone** This does probably not work if you are using a standalone or cloud environment.
 You can ignore this as we will be using a version already provided on Docker hub.




### 3) Prepare the Deployment for the Lab Operator Controller


The deployment manifest for the Operator (that was generated automatically) looks like this:

```yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: ansible-operator-frontend
spec:
replicas: 1
selector:
 matchLabels:
   name: ansible-operator-frontend
template:
 metadata:
   labels:
     name: ansible-operator-frontend
 spec:
   serviceAccountName: ansible-operator-frontend
   containers:
     - name: ansible
       command:
       - /usr/local/bin/ao-logs
       - /tmp/ansible-operator/runner
       - stdout
       --> Replace this with the built image name
       image: "{{ REPLACE_IMAGE }}"
       imagePullPolicy: "{{ pull_policy|default('Always') }}"
       volumeMounts:
       - mountPath: /tmp/ansible-operator/runner
         name: runner
         readOnly: true
     - name: operator
       --> Replace this with the built image name
       image: "{{ REPLACE_IMAGE }}"
       imagePullPolicy: "{{ pull_policy|default('Always') }}"
       volumeMounts:
       - mountPath: /tmp/ansible-operator/runner
         name: runner
       env:
         - name: WATCH_NAMESPACE
           valueFrom:
             fieldRef:
               fieldPath: metadata.namespace
         - name: POD_NAME
           valueFrom:
             fieldRef:
               fieldPath: metadata.name
         - name: OPERATOR_NAME
           value: "ansible-operator-frontend"
   volumes:
     - name: runner
       emptyDir: {}

Modify the YAML to use the image that we have pushed to the registry in the Operator deployment

cp ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml.bak
	
sed -i 's|REPLACE_IMAGE|niklaushirt/ansible-operator-frontend:ansible|g' ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml
sed -i 's|{{ pull_policy.* }}|Always|g' ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml

On Mac please use:
sed -i '' 's|REPLACE_IMAGE|niklaushirt/ansible-operator-frontend:ansible|g' ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml
sed -i '' 's|{{ pull_policy.* }}|Always|g' ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml

more ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml

To the attentive observer, we’re using the prepared image from the docker hub, due to some routing issues for Kubernetes to pull the image internally - thanks minikube. The Training VM also uses a slightly older

2) Deploy the Lab Operator

Deploy the ansible-operator-frontend Custom Resource Definition

kubectl apply -f  ~/ansible-operator/ansible-operator-frontend/deploy/crds/ansiblelab.ibm.com_myansiblelabdemos_crd.yaml
> customresourcedefinition.apiextensions.k8s.io/MyAnsibleLabDemos.ansiblelab.ibm.com created

Create the ansible-operator-frontend Service Account

kubectl apply -f  ~/ansible-operator/ansible-operator-frontend/deploy/service_account.yaml
kubectl apply -f  ~/ansible-operator/ansible-operator-frontend/deploy/role.yaml
kubectl apply -f  ~/ansible-operator/ansible-operator-frontend/deploy/role_binding.yaml
   
> serviceaccount/ansible-operator-frontend created
> role.rbac.authorization.k8s.io/ansible-operator-frontend created
> rolebinding.rbac.authorization.k8s.io/ansible-operator-frontend created

Create the ansible-operator-frontend Operator

kubectl apply -f  ~/ansible-operator/ansible-operator-frontend/deploy/operator.yaml
   
> deployment.apps/ansible-operator-frontend created

Check and wait for the Operator to be running

kubectl get pods
   
> NAME                                       READY   STATUS    RESTARTS   AGE
> ansible-operator-frontend-fd78bcf5-zxgws   1/1     Running   0          43m

Get the log of the Operator Pod (use the Pod name from the previous step)

 kubectl logs -c operator ansible-operator-frontend-fd78bcf5-zxgws
 
  > {"level":"info","ts":1593618304.1191652,"logger":"cmd","msg":"Go Version: go1.12.5"}
  > {"level":"info","ts":1593618304.1194193,"logger":"cmd","msg":"Go OS/Arch: linux/amd64"}
  > {"level":"info","ts":1593618304.1194592,"logger":"cmd","msg":"Version of operator-sdk: v0.8.0"}
  > {"level":"info","ts":1593618304.1194792,"logger":"cmd","msg":"Watching namespace.","Namespace":"default"}
  > {"level":"info","ts":1593618304.1632428,"logger":"leader","msg":"Trying to become the leader."}
  
  > {"level":"info","ts":1593618338.328806,"logger":"leader","msg":"Became the leader."}
  > {"level":"info","ts":1593618338.3861306,"logger":"proxy","msg":"Starting to serve","Address":"127.0.0.1:8888"}
  > {"level":"info","ts":1593618338.386716,"logger":"manager","msg":"Using default value for workers 1"}
  > {"level":"info","ts":1593618338.3867893,"logger":"ansible-controller","msg":"Watching resource","Options.Group":"ansiblelab.ibm.com","Options.Version":"v1beta1","Options.Kind":"MyAnsibleLabDemo"}
  > {"level":"info","ts":1593618338.387424,"logger":"kubebuilder.controller","msg":"Starting EventSource","controller":"myansiblelabdemo-controller","source":"kind source: ansiblelab.ibm.com/v1beta1, Kind=MyAnsibleLabDemo"}
  > {"level":"info","ts":1593618338.487868,"logger":"kubebuilder.controller","msg":"Starting Controller","controller":"myansiblelabdemo-controller"}
  > {"level":"info","ts":1593618338.5881243,"logger":"kubebuilder.controller","msg":"Starting workers","controller":"myansiblelabdemo-controller","worker count":1}

This means that the Operator is working and ready.

Lab 1: Kubernetes Operators Lab 3: Deploy the Lab Operator