06-rolling-updates.md

Replicas and Rolling update:

Create Deployment

Recreate the nginx deployment that we did earlier:

$ kubectl create deployment nginx --image=nginx:1.7.9

And, expose the pod using a load balancer service (remember that it might take a few minutes for the cloud infrastructure to deploy the load balancer, i.e. the external IP might be shown as pending):

$ kubectl expose deployment nginx --port 80 --type LoadBalancer

Note down the loadbalancer IP from the services command:

$ kubectl get service

In case, you are doing this excercise on your local kubernetes clsuter (minikube, kubeadm, etc), then you can simply expose this service as NodePort and use the worker-node-name/IP:nodeport to achieve the same.

$ kubectl expose deployment nginx --port 80 --type NodePort

$ kubectl get services
NAME         TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.32.0.1     <none>        443/TCP        1h
nginx        NodePort    10.32.254.6   <none>        80:30900/TCP   7m
$ 

Increase the replicas to four:

$ kubectl scale deployment nginx --replicas=4

From another terminal on your machine check (using load balancer IP) which version is currently running and to see changes when rollout is happening:

$ while true; do  curl -sI 35.205.60.29  | grep Server; sleep 2; done

On local kubernetes cluster, it would be:

$ while true; do  curl -sI k8s-worker-node:30900  | grep Server; sleep 1; done

Update Deployment

Rollout an update to the image:

$ kubectl set image deployment nginx nginx=nginx:1.9.1 --record

Check the rollout status:

$ kubectl rollout status deployment nginx

Investigate rollout history:

$ kubectl rollout history deployment nginx

Try rolling out other image version by repeating the set image command from above. Suggested image versions are 1.12.2, 1.13.12, 1.14.1, 1.15.2.

Try also rolling out a version that does not exist:

$ kubectl set image deployment nginx nginx=nginx:100.200.300 --record

what happened - do the curl operation still work? Investigate the running pods with:

$ kubectl get pods

You should see ImagePullBackOff under STATUS of some of the pods.

Undo Update

The rollout above using a non-existing image version caused some pods to be non-functioning. Next, we will undo this faulty deployment. First, investigate rollout history:

$ kubectl rollout history deployment nginx

Undo the rollout and restore the previous version:

$ kubectl rollout undo deployment nginx

Investigate the running pods:

$ kubectl get pods

Clean up

Delete deployments and services as follow:

$ kubectl delete deployment nginx
$ kubectl delete service nginx