Kubernetes How It works

We run Docker, then we go to Kubernetes and turn on cluster in Docker — what exactly are we doing? Is it related to architecture?

Try to Understand this

Your App

↓

Docker Image

↓

Container Runtime (Docker / containerd)

↓

Kubernetes

↓

Cluster (Control Plane + Worker Nodes)

↓

Pods (running your containers)

Architecture of Kubernetes

When you enable Kubernetes in Docker Desktop:

Your laptop becomes BOTH master and worker.

When you enable Kubernetes:

You are moving from:

Single container management (Docker only)

To:

Cluster-based container orchestration (Kubernetes)

Docker Only	Docker + Kubernetes
Run 1 container	Run many containers
Manual restart	Auto restart
No auto scaling	Auto scaling
No self-healing	Self-healing
Local only	Cluster architecture

1. Docker builds the image

2. Kubernetes runs and manages containers from that image

When we run with Docker,

docker build -t flask-app .

docker run flask-app

That is simple container execution.

But in Kubernetes:

You create:

• Deployment YAML

• Service YAML

Then:

kubectl apply -f deployment.yaml

apiVersion: apps/v1

kind: Deployment

metadata:

name: flask-deployment

spec:

replicas: 3

selector:

matchLabels:

app: flask-app

template:

metadata:

labels:

app: flask-app

spec:

containers:

- name: flask-container

image: flask-app:latest

ports:

- containerPort: 5000

What happens internally?

1. API Server receives YAML

2. Scheduler selects node

3. 3 Pods are created

4. Each Pod runs container from image

5. kubelet starts container

Why we Use Serverice. Yaml

Pod-1 → 10.244.0.5

Pod-2 → 10.244.0.6

Pod-3 → 10.244.0.7

What Happens If Pod Crashes?

Imagine you have:

Replica = 1
Pod IP = 10.244.0.5

Now:

• Pod crashes ❌

• Kubernetes deletes it

• Kubernetes creates a NEW Pod

Now new Pod might get:

Pod IP = 10.244.0.9

Old IP (10.244.0.5) is gone.

That is what we mean by:

Pods have dynamic IPs.

like we did here ,

apiVersion: v1

kind: Service

metadata:

name: flask-service

spec:

type: NodePort

selector:

app: flask-app

ports:

- protocol: TCP

port: 80

targetPort: 5000

nodePort: 30007