What is Kubernetes - CLI Guide

Introduction

Kubernetes helps you run and manage many software containers across multiple computers. It solves the problem of keeping your apps running smoothly even if some computers fail or need updates.

When you want to run your app on many computers to handle more users without crashing

When you need to update your app without stopping it for users

When you want to automatically restart parts of your app if they stop working

When you want to easily move your app to a different computer or cloud service

When you want to manage many small parts of your app separately but still have them work together

Commands

Check the installed Kubernetes command-line tool version to ensure it is ready to use.

Terminal

kubectl version --client

Expected Output

Client Version: v1.27.3

Show information about the Kubernetes cluster you are connected to, confirming access and cluster status.

Terminal

kubectl cluster-info

Expected Output

Kubernetes control plane is running at https://127.0.0.1:6443

CoreDNS is running at https://127.0.0.1:6443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

List all the computers (nodes) in your Kubernetes cluster to see where your apps can run.

Terminal

kubectl get nodes

Expected Output

NAME           STATUS   ROLES           AGE   VERSION

worker-node-1   Ready    <none>          10d   v1.27.3

worker-node-2   Ready    <none>          10d   v1.27.3

Key Concept

If you remember nothing else from this pattern, remember: Kubernetes helps you run and manage many app parts across many computers automatically and reliably.

Common Mistakes

Trying to use Kubernetes commands without a running cluster or proper access.

Commands fail because there is no cluster to talk to or you lack permission.

Set up a Kubernetes cluster first or configure access to an existing one before running commands.

Confusing Kubernetes nodes with containers or pods.

Nodes are the computers, while pods are the app parts running inside containers on those nodes.

Learn the difference: nodes are machines, pods are groups of containers running your app.

Summary

Kubernetes manages app containers across many computers to keep apps running smoothly.

Use kubectl commands to check your tool version, cluster info, and nodes in the cluster.

Kubernetes helps apps update, recover, and scale automatically without manual intervention.

Practice

1. What is the main purpose of Kubernetes?

easy

A. To design websites with HTML and CSS

B. To manage and run many copies of programs across multiple computers

C. To write code for mobile applications

D. To create databases for storing data

Solution

1. Step 1: Understand Kubernetes role

   Kubernetes is a tool that helps run many copies of programs on many computers.

2. Step 2: Compare options

   Only To manage and run many copies of programs across multiple computers describes managing programs across computers, which matches Kubernetes' purpose.

3. Final Answer:

   To manage and run many copies of programs across multiple computers -> Option B

4. Quick Check:

   Kubernetes = Manage many program copies [OK]

Hint: Kubernetes manages many program copies on many computers [OK]

Common Mistakes:

• Confusing Kubernetes with programming languages
• Thinking Kubernetes is for website design
• Mixing Kubernetes with database creation

2. Which command is used to interact with Kubernetes from the command line?

easy

A. kubectl

B. docker

C. git

D. npm

Solution

1. Step 1: Identify Kubernetes command-line tool

   The tool to control Kubernetes is called kubectl.

2. Step 2: Check other options

   Docker manages containers, git manages code versions, npm manages JavaScript packages, so they are not correct.

3. Final Answer:

   kubectl -> Option A

4. Quick Check:

   Kubernetes CLI = kubectl [OK]

Hint: Use kubectl to control Kubernetes from terminal [OK]

Common Mistakes:

• Using docker instead of kubectl for Kubernetes
• Confusing git with Kubernetes commands
• Thinking npm controls Kubernetes

3. What happens if a program managed by Kubernetes crashes on one computer?

medium

A. Kubernetes deletes all copies of the program

B. The program stays crashed and Kubernetes does nothing

C. Kubernetes automatically restarts the program on the same or another computer

D. You must manually restart the program yourself

Solution

1. Step 1: Understand Kubernetes fault handling

   Kubernetes watches programs and restarts them automatically if they crash.

2. Step 2: Evaluate options

   Only Kubernetes automatically restarts the program on the same or another computer describes automatic restart, which matches Kubernetes behavior.

3. Final Answer:

   Kubernetes automatically restarts the program on the same or another computer -> Option C

4. Quick Check:

   Kubernetes auto-restarts crashed programs [OK]

Hint: Kubernetes auto-restarts crashed programs [OK]

Common Mistakes:

• Thinking Kubernetes ignores crashed programs
• Believing Kubernetes deletes programs on crash
• Assuming manual restart is always needed

4. You run kubectl get pods but get an error saying "connection refused". What is the likely problem?

medium

A. Kubernetes cluster is not running or unreachable

B. The kubectl command is misspelled

C. Your computer has no internet connection

D. You typed the wrong pod name

Solution

1. Step 1: Analyze error message

   "Connection refused" means kubectl cannot reach the Kubernetes cluster.

2. Step 2: Check options

   Misspelling would cause command not found, no internet is unrelated if cluster is local, wrong pod name gives different error.

3. Final Answer:

   Kubernetes cluster is not running or unreachable -> Option A

4. Quick Check:

   Connection refused = cluster unreachable [OK]

Hint: Connection refused means cluster is down or unreachable [OK]

Common Mistakes:

• Assuming internet is required for local cluster
• Thinking typo causes connection refused
• Confusing pod name errors with connection errors

5. You want to keep a program running with 3 copies at all times using Kubernetes. Which Kubernetes object should you create?

hard

A. Namespace

B. Service

C. ConfigMap

D. Deployment

Solution

1. Step 1: Identify object for managing program copies

   A Deployment manages multiple copies (replicas) of a program and keeps them running.

2. Step 2: Review other objects

   Service exposes programs to network, ConfigMap stores config data, Namespace groups resources, none manage replicas.

3. Final Answer:

   Deployment -> Option D

4. Quick Check:

   Multiple copies managed by Deployment [OK]

Hint: Use Deployment to keep multiple program copies running [OK]

Common Mistakes:

• Confusing Service with Deployment
• Using ConfigMap to manage replicas
• Thinking Namespace controls program copies