peter (admin)   -  



Installing kubectl on Linux

We have now installed the main prerequisites, installed Docker, and installed Minikube.

Docker allows us to work with containers, while Minikube allows us to create and run a local Kubernetes cluster.

However, one very important tool is still missing:

kubectl

kubectl is the main command-line tool used to communicate with and manage a Kubernetes cluster.

In this section, we will download, install, and verify kubectl on our Linux system.


What Is kubectl?

kubectl is the Kubernetes command-line interface.

It allows us to send commands to the Kubernetes API server and interact with resources inside the cluster.

For example, we can use kubectl to:

  • Create resources

  • View Pods

  • View Deployments

  • View Services

  • Delete resources

  • Scale applications

  • Examine logs

  • Troubleshoot Pods

  • Apply YAML configuration files

  • Inspect cluster information

Some common commands include:

kubectl get pods
kubectl get nodes
kubectl get services
kubectl get deployments

kubectl is therefore one of the tools we will use most frequently when working with Kubernetes.


How kubectl Communicates with Kubernetes

kubectl does not directly communicate with individual worker nodes or containers.

Instead, it communicates with the Kubernetes API server.

The basic relationship looks like this:

User
 |
 v
kubectl
 |
 v
Kubernetes API Server
 |
 +----------------------+
 |          |           |
 v          v           v
Pods    Services    Deployments

When we enter a kubectl command, kubectl sends an API request to the Kubernetes API server.

Kubernetes processes the request and returns the appropriate response.


Download kubectl

The recommended approach is to download kubectl from the official Kubernetes download location:

dl.k8s.io

We can use curl to determine the current stable Kubernetes version and download the appropriate kubectl executable.

For a 64-bit Intel/AMD Linux system, run:

curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"

The command contains two parts.

First:

curl -L -s https://dl.k8s.io/release/stable.txt

retrieves the current stable Kubernetes version.

For example, it may return something similar to:

v1.xx.x

The outer curl command then uses that version to construct the download URL for kubectl.


Understanding linux/amd64

The download path contains:

linux/amd64

linux tells us that the executable is intended for the Linux operating system.

amd64 refers to the common 64-bit x86 architecture used by most Intel and AMD desktop and server processors.

You can check your system architecture with:

uname -m

On an AMD64-compatible system, you will normally see:

x86_64

If you are using an ARM-based machine, you would need the appropriate ARM version instead.


Check the Download

After the download has completed, check that the file exists:

ls -lh kubectl

You should see a file called:

kubectl

You can also check what type of file it is:

file kubectl

The output should identify it as a Linux executable.


Install kubectl

Now install the downloaded executable into:

/usr/local/bin

Run:

sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl

This command both copies the file and sets the appropriate ownership and permissions.


Understanding the Permissions

The important part of the command is:

-m 0755

This sets the permissions of the kubectl executable.

0755 means:

Owner:   read + write + execute
Group:   read + execute
Others:  read + execute

This allows users on the system to execute kubectl while only the owner can modify the file.

The permissions can also be represented as:

rwxr-xr-x

You can check them after installation with:

ls -l /usr/local/bin/kubectl

Why Install into /usr/local/bin?

We install kubectl into:

/usr/local/bin

because this directory is normally included in the Linux system's PATH.

You can check your PATH with:

echo $PATH

Because /usr/local/bin is in the PATH, we can simply enter:

kubectl

instead of:

/usr/local/bin/kubectl

This makes kubectl available from any directory in the terminal.


Verify the Installation Location

We can check where Linux finds kubectl with:

which kubectl

The expected result is:

/usr/local/bin/kubectl

You can also use:

command -v kubectl

Again, you should see:

/usr/local/bin/kubectl

This confirms that kubectl has been installed correctly and is available through the system PATH.


Check the kubectl Version

We can now verify that kubectl works correctly.

Run:

kubectl version --client

The --client option is important at this stage.

It tells kubectl:

Display information about the kubectl client itself without requiring information from a Kubernetes cluster.

The output will contain the installed kubectl version.

Depending on the version, you may also see information relating to Kustomize, which is built into kubectl and can be used to customize Kubernetes configuration.


What Is Kustomize?

Modern versions of kubectl include support for Kustomize.

Kustomize allows Kubernetes YAML configurations to be customized without directly modifying the original YAML files.

For example, you may eventually encounter:

kubectl apply -k .

For now, we do not need to understand Kustomize in detail.

The important point is that if the kubectl version output includes something such as:

Client Version: ...
Kustomize Version: ...

this is normal.


Test kubectl

We can also display the available commands:

kubectl --help

This will show the large number of operations available through kubectl.

Some of the commands we will frequently use include:

kubectl get
kubectl describe
kubectl apply
kubectl create
kubectl delete
kubectl logs
kubectl exec
kubectl scale

These commands will become very familiar as we begin working with Kubernetes.


kubectl and Minikube

Once we start our Minikube cluster, kubectl will become our main way of interacting with it.

The relationship between the tools will look something like this:

            Linux System
                 |
       +---------+---------+
       |                   |
       v                   v
    Minikube            kubectl
       |                   |
       v                   |
Kubernetes Cluster <-------+
       |
       +---- Pods
       |
       +---- Deployments
       |
       +---- Services
       |
       +---- ConfigMaps
       |
       +---- Secrets

Minikube creates and manages our local Kubernetes environment.

kubectl allows us to communicate with the Kubernetes API and manage resources inside that environment.


Example kubectl Commands

Once our cluster is running, we can check the nodes with:

kubectl get nodes

We can view Pods with:

kubectl get pods

We can see all Pods across all namespaces with:

kubectl get pods -A

We can inspect a particular Pod with:

kubectl describe pod 

We can view container logs with:

kubectl logs 

And one of the most important commands when working with YAML files is:

kubectl apply -f application.yaml

This submits our YAML configuration to Kubernetes.


Our Kubernetes Toolkit

At this stage, we have installed three important tools:

Docker
   |
   +---- Build and run containers
   |
   v
Minikube
   |
   +---- Create a local Kubernetes cluster
   |
   v
kubectl
   |
   +---- Communicate with Kubernetes
   |
   v
Kubernetes Resources

Each tool has a different responsibility.

Tool Purpose
Docker Build, download, and run containers
Minikube Create and manage a local Kubernetes cluster
kubectl Communicate with and manage the Kubernetes cluster

kubectl Installation Complete

We have now successfully:

  • Downloaded kubectl from the official Kubernetes source

  • Installed it into /usr/local/bin

  • Configured the correct executable permissions

  • Verified its installation location

  • Checked the kubectl client version

  • Confirmed that the command is ready to use

Our basic Kubernetes toolset is now in place.

With Docker, Minikube, and kubectl installed, we are ready to start our local Kubernetes cluster and begin working with Kubernetes resources directly.

 


Comments (0)

There are no comments yet. Be the first to comment.

Log in or register to leave a comment. all comments are subject to admin approval