How to Create a Kubernetes Cluster on Google Cloud using Kops

Hey yo! In recent times, there has been more adoption of Kubernetes. As a result, the possibilities are getting endless.
Kubernetes Operations(Kops) is a simple and easy way to get Kubernetes Cluster up and running. It is Production Ready, can be used for upgrades and management of your Kubernetes Cluster(s). In this article, we will learn how to deploy a Kubernetes Cluster to the Google Cloud Platform.

Previously, I have written an article on How to Create a Kubernetes Cluster on AWS with Kops and Netlify DNS. You should check it out if you need to deploy to AWS. Now let’s dive right in!

Prerequisites

• Kops
• A Google Cloud Platform Account
• Kubectl
• gcloud CLI

Step 1

Ensure you have kubectl, kops, and gcloud correctly installed; you can check that by running

kubectl

You can also check what version of kubectl you are running

kubectl version | grep -i GitVersion

You should see an output similar to this:

Client Version: version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.0", GitCommit:"cb303e613a121a29364f75cc67d3d580833a7479", GitTreeState:"clean", BuildDate:"2021-04-08T16:31:21Z", GoVersion:"go1.16.1", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.1", GitCommit:"5e58841cce77d4bc13713ad2b91fa0d961e69192", GitTreeState:"clean", BuildDate:"2021-05-21T23:01:33Z", GoVersion:"go1.16.4", Compiler:"gc", Platform:"linux/amd64"}

You can also check the version for kops

kops version

Check to see gcloud is installed properly too

gcloud

You can also check the version of gcloud

gcloud version

Step 2

Login to your gcloud account using the command below

gcloud auth login

Automatically, a browser tab will be opened to authenticate and authorize your access to the google cloud platform. It should look like this.

Go ahead and accept, and you will be redirected to the auth success page. It should look like this.

You have made some significant progress at this point! Let’s keep going

Step 3

Configure the Google cloud project where kops will deploy the resources.

gcloud config set project [name of the project-id]

Create a Google cloud bucket.

gsutil mb gs://hey-kubernetes

We are creating a bucket because kops needs to save the state configuration of the provisioned resources somewhere, and it helps track what has been completed, modified, or deleted. Think of this as version control

Export the following environmental variables in the terminal

export KOPS_FEATURE_FLAGS=AlphaAllowGCE # to unlock the GCE features
export KOPS_STATE_STORE=gs://hey-kubernetes

Step 4

Create a Virtual Private Cloud Network

gcloud compute networks create hey-kubernetes --subnet-mode=auto

The command above will create a vpc called hey-kubernetes where the kubernetes cluster resources will be assigned to while creating such compute engine, firewalls, load balancers, etc.

Step 5

Spin up a kubernetes cluster using Kops

kops create cluster heykubernetes.k8s.local --zones us-central1-a  --node-count 3 --node-size n1-standard-4  --master-size n1-standard-2 --vpc=hey-kubernetes

You should get a confirmation output that looks like this

Cluster configuration has been created.

Suggestions:
 * list clusters with: kops get cluster
 * edit this cluster with: kops edit cluster heykubernetes.k8s.local
 * edit your node instance group: kops edit ig --name=heykubernetes.k8s.local nodes-us-central1-a
 * edit your master instance group: kops edit ig --name=heykubernetes.k8s.local master-us-central1-a

Finally configure your cluster with: kops update cluster --name heykubernetes.k8s.local --yes --admin

Run the command below to configure the cluster for provisioning

kops update cluster --name heykubernetes.k8s.local --yes --admin

After a few minutes, you should get the final output that is similar to this

Cluster is starting.  It should be ready in a few minutes.

Suggestions:
 * validate cluster: kops validate cluster --wait 10m
 * list nodes: kubectl get nodes --show-labels
 * ssh to the master: ssh -i ~/.ssh/id_rsa ubuntu@api.heykubernetes.k8s.local
 * the ubuntu user is specific to Ubuntu. If not using Ubuntu please use the appropriate user based on your OS.
 * read about installing addons at: https://kops.sigs.k8s.io/operations/addons. 

You can validate to cluster to know when it is fully functional

kops validate cluster --wait 10m

You should see the output below when it is ready

INSTANCE GROUPS
NAME			ROLE	MACHINETYPE	MIN	MAX	SUBNETS
master-us-central1-a	Master	n1-standard-2	1	1	us-central1
nodes-us-central1-a	Node	n1-standard-4	3	3	us-central1

NODE STATUS
NAME				ROLE	READY
master-us-central1-a-p7qn	master	True
nodes-us-central1-a-1p8b	node	True
nodes-us-central1-a-9hdc	node	True
nodes-us-central1-a-snq9	node	True

Your cluster heykubernetes.k8s.local is ready.

Hurray! You have a Kubernetes Cluster running. If you reach here, you are fantastic. Let’s try something way cool; spinning up an Nginx image.

Step 6

Run an nginx image using kubectl

kubectl run hey  --image nginx

You should see output similar to this

pod/hey created

You can check to see if the Nginx is running using this command below

kubectl get pod 

You will get this output

NAME   READY   STATUS    RESTARTS   AGE
hey    1/1     Running   0          79s

You can delete the Nginx container by running

kubectl delete pod hey

Step 7

Run the following commands to destroy all the resources in the following order

Destroy the Kubernetes cluster

kops delete cluster heykubernetes.k8s.local --yes

Next, Destroy the VPC.

gcloud compute networks delete hey-kubernetes

Finally, Destroy the bucket.

gsutil rm -r gs://hey-kubernetes

Please, some resources take a few minutes to create or destroy. Do not delete the bucket when the Kubernetes cluster hasn’t been deleted completely.

Way to go!

Either you are a beginner at Backend, SRE/DevOps, Infrastructure engineering, this post was created for you to get your feet wet with Kubernetes. I can’t wait to see what you do with Kubernetes. Cheerio!

— Oct 19, 2021