Multi Cloud Network Architecture

Building Networking Skills in the age of the Cloud

Multi Cloud Network Architecture

Google Kubernetes Engine (GKE) 101

Share on linkedin
Share on twitter
Share on email

In this post, I’m switching focus a little to an area that I am very interested in and get questions constantly from customers: Kubernetes.

This post isn’t meant to be a primer on K8s, but rather a “get your hands dirty introduction”. I guarantee you, that just by going through the motions of deploying a basic application, you will improve your chances of learning this technology greatly.

Let’s have a look at the application we are going to deploy.

shop front

Stans Robot Shop is a microservice e-commerce application leveraging a number of backend technologies and is ideal as a more complete “Hello World” application.

Step 1 – Preparation

There are a few tools that you’re going to need to have installed before we go any further.

GCloud CLI
GKE Auth Plugin
GitGoogle for your OS link
PythonGoogle for your OS link

You are of course going to need to have a GCP account and an active project created where you will deploy your K8s cluster.

Step 2 – Create the Kubernetes hosting VPC in GCP

First up, log into your GCP account, and let’s begin by creating a VPC to host the K8s cluster.

Add the subnet of your choice:

Firewall rules and other settings can remain as default.

A note on IP addressing: the VPC subnet range will be used for the worker nodes as they are deployed and is the address used to reach the Kubernetes API.

When creating the cluster, we will be prompted for the pod and service address ranges, which will not be in the VPC CIDR range.

Step 3 – Create the Kubernetes cluster in GCP

Next up, we’ll deploy the cluster itself:

For this example, I have chosen the autopilot configuration:

Give your cluster a name:

Ensure you deploy it in the same region as your VPC subnet:

Choose the VPC you created in step 2 above. You can keep the default settings for the cluster and service address ranges.

In a future post I’ll go into more detail on the CIDR ranges.

Hit create.

You will be able to follow the deployment on Gcloud CLI as seen here:

gcloud container operations list

You can see that the “create” process is currently running.

As soon as it has finished, you can move on to deploying the application.

Step 4 – Deploy the application

We’ll begin by authorizing gcloud CLI to manage the cluster:

Click on the cluster and then click on connect.

Copy this command into your local terminal:

You can test this by running a few kubectl commands from your local machine:

kubectl get pods -A

Next we will clone the application locally on our machine as this is where the helm chart is stored:

git clone

Navigate to the /robot-shop/K8s/helm directory.

Now you are ready to use Helm to deploy your application.

Begin by creating a new namespace for your application:

kubectl create ns robot-shop

helm install robot-shop --namespace robot-shop .

Now we should see our pods getting deployed in GCP:

And we should also check if our load balancer got deployed:

We can see that the web component is exposed to the internet. As soon as the web pod is up, we should be able to access it:

Step 5 -Cleanup

At this stage, we have deployed our application. You can use this environment to test some common K8s commands:

kubectl get pods --namespace=robot-shop

kubectl get nodes

kubectl get svc --namespace=robot-shop

When finished, you can use Helm to uninstall the deployed application:

helm uninstall robot-shop --namespace robot-shop .

Then we can delete the cluster:

gcloud container clusters list

Use the information from the above to delete the cluster:

gcloud container clusters delete autopilot-cluster-1-stans-robot --region=europe-west1

And that’s it for this post.

You can use this environment to test further.

I brushed over quite a few fundamentals in this post. The objective was to pique your curiosity so that you can go off and read up on the nuances that we didn’t go into detail on – helm charts, cluster configuration, control plane verification, etc.

In a future post, I will look at some advanced GKE options, troubleshooting, and observability.

A sneak peak into a future post:

Check our the K9s project for more powerful cluster management features:

Thanks for reading.

Share if you liked it
Share on linkedin
Share on twitter
Share on email
0 0 votes
Article Rating
Notify of
Inline Feedbacks
View all comments

20 year veteran of the networking industry currently specialising in Cloud Networking and Security.

CCIE #16661 (R&S, SP)


I am currently an employee of Aviatrix. All opinions, views and statements are my own and do not reflect that of my employer. Any errors are mine and mine alone. Any ignorance is mine, though I do believe my parents and the public school system should shoulder some of that blame. 

Recent Posts