https://drive.google.com/file/d/1oUzRfeaB33aht95JwtJKTa7A0d7VrXtB/view?usp=sharing

🔍 YAML Breakdown

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nnwebserver
spec:
  selector:
    matchLabels:
      run: nnwebserver          # ← Must match Pod template labels
  replicas: 2                   # ← Run 2 identical Pods
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1             # ← Allow 1 extra Pod during update
      maxUnavailable: 0       # ← Keep all Pods available during update
  template:
    metadata:
      labels:
        run: nnwebserver       # ← Labels for Pods (must match selector!)
    spec:
      containers:
        - name: nnwebserver
          image: lovelearnlinux/webserver:v2
          livenessProbe:       # ← Health check
            exec:
              command: ["cat", "/var/www/html/index.html"]
            initialDelaySeconds: 10
            periodSeconds: 20
          resources:
            requests:
              cpu: "100m"
              memory: "128Mi"
            limits:
              cpu: "200m"
              memory: "256Mi"
          ports:
            - containerPort: 80

📌 Why Use a Deployment (vs Bare Pod)?

✅ Deployments manage ReplicaSets, which manage Pods → 3-layer controller pattern.

🔑 Key Concepts Explained

1. selector.matchLabels

• Must exactly match template.metadata.labels
• Immutable after creation → don’t change it!

2. RollingUpdate Strategy

• maxSurge: 1 → During update, allow 1 extra Pod (total = 3)
• maxUnavailable: 0 → Zero downtime (always keep 2 Pods running)

🎯 Result:

• Update starts → 3 Pods (2 old + 1 new)
• After new Pod is ready → kill 1 old → 3 Pods (1 old + 2 new)
• Repeat until all updated

3. Liveness Probe

• Checks if app is alive
• If fails → restart container
• Uses exec to check if index.html exists