riteshatri/deploymentrecreate repository overview

⁠🚀 Kubernetes Deployment – Recreate Strategy (Complete Practical Lab)

This document is a step-by-step, beginner-friendly lab guide to clearly understand how Deployment Recreate Strategy works in Kubernetes.

You will perform 3 image-based practicals (v1, v2, v3) and visually experience downtime when images are updated.

⁠🧠 Learning Objective

By the end of this lab, you will understand:

• What Recreate strategy is
• Why downtime happens
• How Kubernetes behaves when images are updated
• Why Recreate strategy is not recommended for production
• Why RollingUpdate exists

⁠🧩 Prerequisites

• Kubernetes cluster (AKS / EKS / GKE / Minikube)
• kubectl configured
• Docker installed
• Internet access (to pull images)
• Basic understanding of Pods & Services

⁠🐳 Docker Images Used (Already Pushed)

We will use 3 versions of the same application:

docker push riteshatri/deploymentrecreate:v1
docker push riteshatri/deploymentrecreate:v2
docker push riteshatri/deploymentrecreate:v3

Each image shows different UI text so changes are clearly visible in browser.

⁠📁 Files Used in This Lab

⁠📄 deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: deploymentrecreate
spec:
  # replicas: 10
  replicas: 2
  # replicas: 3
  # replicas: 50
  selector:
    matchLabels:
      humhai: recreate
  template:
    metadata:
      labels:
        humhai: recreate
    spec:
      containers:
        - name: ritcontainer
          image: riteshatri/deploymentrecreate:v1
          # image: riteshatri/deploymentrecreate:v2
          # image: riteshatri/deploymentrecreate:v3
          ports:
            - containerPort: 80
  strategy:
    type: Recreate

⁠📄 service.yaml

apiVersion: v1
kind: Service
metadata:
  name: ritservice
spec:
  selector:
    humhai: recreate
  ports:
    - port: 6541
      targetPort: 80
  type: LoadBalancer

⁠🔹 STEP 1: Deploy Version-1 (Initial Setup)

⁠Apply Deployment (v1 image)

kubectl apply -f deployment.yaml

Check pods:

kubectl get pods  
kubectl get deployment

➡️ Pods should be Running

⁠🔹 STEP 2: Create Service

kubectl apply -f service.yaml
kubectl get service

Copy EXTERNAL-IP and PORT

Open browser:

http://<EXTERNAL-IP>:<PORT>

✅ You will see Deployment Recreate – Version 1

⁠🔹 STEP 3: Update Image to Version-2 (Recreate in Action)

⁠Edit deployment.yaml

Change image:

image: riteshatri/deploymentrecreate:v2
strategy:
  type: Recreate

Apply update:

kubectl apply -f deployment.yaml

⁠⚠️ IMPORTANT OBSERVATION (VERY IMPORTANT)

Immediately run:

kubectl get pods

You will notice:

❌ All OLD pods TERMINATED
❌ For some time NO pods running
❌ Website goes DOWN
❌ Browser shows error

➡️ THIS IS DOWNTIME

⁠🔹 STEP 4: New Pods Come Up

After a few seconds:

kubectl get pods

➡️ New pods start running
➡️ Refresh browser

✅ Now Version-2 UI is visible

⁠🧠 CONCLUSION (Version-2)

⁠🔹 STEP 5: Update Image to Version-3

Edit deployment.yaml again:

image: riteshatri/deploymentrecreate:v3

Apply:

kubectl apply -f deployment.yaml

⁠💥 WHAT HAPPENS NOW?

• All pods die ☠️
• Website disappears ❌
• Users see nothing 😵
• After some time…
• New pods start
• Website comes back 🎉

Refresh browser → Version-3 UI visible

⁠📊 Complete Image Flow Summary

⁠🚨 Key Takeaways (EXAM / INTERVIEW READY)

✔ Recreate strategy deletes all pods first
✔ Guaranteed downtime
✔ Not safe for production
✔ Good only for learning
✔ Used rarely in real systems

⁠🧠 ONE-LINE MEMORY TRICK

⁠🔜 What Next?

👉 Learn RollingUpdate Strategy
👉 Zero downtime deployments
👉 Blue-Green & Canary deployments

⁠❤️❤️ Author ❤️❤️

RITESH SHARMA

Created for Kubernetes Hands-on Learning
Happy Learning ☸️🔥