Getting started with Minikube: deploying application on local Kubernetes cluster

Minikube is a great tool for local Kubernetes setup. In this article, we are going to learn:

  • What is minikube?
  • Minikube installation
  • Interacting with minikube
  • Accessing a service

What is Minikube?

Minikube allows you to run Kubernetes locally. It creates a single-node Kubernetes cluster on your local machine. It’s a great tool for starting local Kubernetes and do a quick prototype.

Minikube installation

To install minikube locally you need to have a container or virtual machine installed. For installation, you can check out official documentation at the minikube page. The simplest thing would be to install minikube with Docker.

Starting Minikube

Once you have minikube installed, you can start it by running command:

minikube start

One of the great tool that comes with minikube is the dashboard. You can start dashboard by running command:

minikube dashboard


Kubectl is the Kubernetes command-line tool. You can use kubectl to deploy applications, inspect and manage cluster resources, view logs, and many more things. To install, check the official documentation.

You can check kubectl reference documentation for details about each command and supported flags. For now, you mostly need to understand the commands listed below:

  • kubectl get – list resources (example kubectl get pods)
  • kubectl describe – show detailed information about a resource (example kubectl describe pod <pod_name>)
  • kubectl logs – print the logs from a container in a pod (example kubectl logs <pod_name>)
  • kubectl exec – execute a command on a container in a pod (example kubectl exec --stdin --tty <pod_name> -- /bin/bash)

You can interact with the minikube cluster using kubectl. For example, to list all pods in all namespace run command:

kubectl get pods -A

This return output like:

NAMESPACE              NAME                                        READY   STATUS    RESTARTS   AGE
kube-system            coredns-74ff55c5b-7cd2g                     1/1     Running   0          22h
kube-system            etcd-minikube                               1/1     Running   0          22h
kube-system            kube-apiserver-minikube                     1/1     Running   0          22h
kube-system            kube-controller-manager-minikube            1/1     Running   0          22h
kube-system            kube-proxy-xm2nd                            1/1     Running   0          22h
kube-system            kube-scheduler-minikube                     1/1     Running   0          22h
kube-system            storage-provisioner                         1/1     Running   2          22h
kubernetes-dashboard   dashboard-metrics-scraper-f6647bd8c-nj6g5   1/1     Running   0          22h
kubernetes-dashboard   kubernetes-dashboard-968bcb79-lk5dg         1/1     Running   0          22h

To check details about Kubernetes cluster run command kubectl cluster-info. This returns output like:

Kubernetes control plane is running at
KubeDNS is running at

To check details about node run command kubectl get nodes. This outputs:

NAME       STATUS   ROLES                  AGE   VERSION
minikube   Ready    control-plane,master   22h   v1.20.7

Creating a POD

Pods are the smallest deployable computing unit that we can create and manage in Kubernetes. A pod contains one or more containers, with shared storage and network resources and specification about how to run containers.

To create a pod, save the below code in a YAML file, say pod.yaml, and run the command kubectl apply -f pod.yaml

apiVersion: v1
kind: Pod
  name: product-svc
    app: product
    - name: product
      imagePullPolicy: IfNotPresent
      image: techdozo/product-svc:1.0.0

To validate if the pod is created successfully run command kubectl get pods

product-svc   1/1     Running   0          2m33s

You can see the pod in the minikube dashboard. You can also check the log from the Kebab menu of the pod.

Accessing application using port forwarding

You can use kubectl port-forwarding to forward the local port to the port of pod in the cluster. Run command kubectl port-forward product-svc 8080:8080 to forward local port 8080 to the pod product-svc port 8080. The product-svc is a microservice that exposes two end-points POST /products/ and GET /products/{productID}. Now, you can access the pod locally using curl or Postman.

curl --location --request POST 'http://localhost:8080/products/' \
--header 'Content-Type: application/json' \
--header 'userId:' \
--data-raw '{
    "name": "Apple iPhone 12",
    "description": "Apple iPhone 12 Pro Max 128 GB, Pacific Blue",
    "price": 1700

Creating a Deployment

You can use Kubernetes deployment to create multiple replicas of identical pods. Using deployment pods can be rolled out predictably (rolling update strategy) so that none of the users of your application experience downtime.

To create a pod, save the below code in a YAML file, say deployment.yaml, and run the command kubectl apply -f deployment.yaml

apiVersion: apps/v1
kind: Deployment
  name: product
    app: product
  replicas: 3
      app: product
        app: product
        - name: product
          imagePullPolicy: IfNotPresent
          image: techdozo/product-svc:1.0.0

You can validate deployment by running the command kubectl get deployment

product   3/3     3            3           3h12m

You can validate pods created by deployment by running command kubectl get pods

NAME                      READY   STATUS    RESTARTS   AGE
product-7b865c97d-bfxwt   1/1     Running   1          3h14m
product-7b865c97d-zd8gc   1/1     Running   1          3h14m
product-7b865c97d-zlrtv   1/1     Running   1          3h14m

Creating a Service

A Kubernetes service is an abstraction that defines a logical set of pods (usually determined by a selector) and a policy to access them. There are different types of services in the Kubernetes. For more information, you can check the Kubernetes service.

To create a Kubernetes service of type NodePort, copy the following code in service-nodeport.yaml and run command kubectl apply -f service-nodeport.yaml

apiVersion: v1
kind: Service
  name: product-service
  type: NodePort
    - port: 80
      targetPort: 8080
      nodePort: 32000
    app: product

This creates a service called product-service. You can check the service section of the minikube dashboard. You can check this article to know the details about other types of services in Kubernetes

The NodePort service defined above, maps pods port 8080 to port 32000 on the host machine.

You can also forward the local port to the service port 80 as kubectl port-forward service/product-service 8080:80 and access the application as described above.

Another way to access the application is by using minikube service by running the command minikube service --url product-service. This output something like

 Starting tunnel for service product-service.
 | NAMESPACE |      NAME       | TARGET PORT |          URL           |
 | default   | product-service |             | |
 ! Because you are using a Docker driver on windows, the terminal needs to be open to run it. 

On windows, using docker, accessing nodeport using host machine doesn’t work. This issue was discussed here with no solution.

Common operations

Viewing all services

You can view the URLs for the services in your local cluster by calling minikube service list.

Minikube logs

To troubleshoot you can check minikube log by calling the command minikube logs -f

Minikube Addons

Minikube comes with many addons. You can view the list of addons by running command minikube addons list. You can enable addons by running command minikube addons enable <add_on_name>.


Minikube is a great way to start development on local Kubernetes cluster. You can deploy application using Kubernetes deployment. The application can be accessed using NodePort service. There are some challenges associated with starting Ingress controller on docker and windows 10 setup.

Social Share !

Software Architect @ Schlumberger ``` Cloud | Microservices | Programming | Kubernetes | Architecture | Machine Learning | Java | Python ```