Kubernetes Hands-on: Pods, Deployments & Self-Healing

Lab 16 of 20 · Pods & Deployments · 5 hands-on activities · bilingual (English + සිංහල)

පිටුව 16: Kubernetes ප්‍රායෝගික පුහුණුව: Pods, Deployments සහ Self-Healing ක්‍රියාවලිය

Pod එකක් කියන්නේ මොකක්ද? Multi-container pods, Deployments සහ ReplicaSets, සහ Pod එකක් මැරුණම තත්පරයෙන් අලුත් එකක් හදන Self-Healing බලමු.

Architectural Foundation & Systems Diagram

In Kubernetes, you almost never deploy individual containers directly. The smallest deployable unit in the Kubernetes object model is the Pod.

• 🥜 What is a Pod? — A Pod wraps one or more tightly coupled containers that share the exact same network namespace (IP address) and can communicate with each other over localhost!

• 🚀 Deployments & ReplicaSets — A Deployment manages a ReplicaSet, which guarantees that a specified number of identical pod replicas are running at all times. If a pod crashes, the ReplicaSet replaces it instantly.

• 🩺 Pod Lifecycle & Probes — • Liveness Probe: Detects if an app is deadlocked; restarts container.
  • Readiness Probe: Detects if app is ready to receive network traffic.

Step-by-step practical lab

Step 01: Launch Imperative Test Pod Running Nginx

සිංහලෙන් — Kubernetes තුළ තනි Pod එකක් ධාවනය කරමු.

Launch an imperative test Pod running lightweight Alpine Nginx to verify pod scheduling and network allocation.

kubectl run nginx-pod --image=nginx:alpine

• run : තනි Pod එකක් ඉක්මනින් පරීක්ෂා කිරීමට භාවිතා කරයි.
• --image=nginx:alpine : භාවිතා කරන container image එක දක්වයි.

Expected output

pod/nginx-pod created

Step 02: Query Active Pods with IP & Node Assignment

සිංහලෙන් — දුවන Pods, ඒවායේ IPs සහ දුවන Worker Node එක පරීක්ෂා කරමු.

View active pods, their ready status, restart count, node assignment, and assigned internal pod IP.

kubectl get pods -o wide

• -o wide : Pod එකට ලැබුණු අභ්‍යන්තර IP ලිපිනය සහ එය ක්‍රියාත්මක වන Worker Node එක පෙන්වයි.

Expected output

NAME                        READY   STATUS    RESTARTS   AGE   IP           NODE
nginx-pod                   1/1     Running   0          4m    10.244.1.14  k8s-worker-01
web-app-76565f9797-2kz9w    1/1     Running   0          2m    10.244.1.15  k8s-worker-01
web-app-76565f9797-9xj4p    1/1     Running   0          2m    10.244.2.18  k8s-worker-02
web-app-76565f9797-w8p2m    1/1     Running   0          2m    10.244.2.19  k8s-worker-02

Step 03: Deep Inspect Pod Lifecycle Events & Logs

සිංහලෙන් — Pod එකක් startup වීමේදී සිදු වූ events සහ තොරතුරු පරීක්ෂා කරමු.

Inspect low-level container image pulling, startup timing, and lifecycle events to diagnose CrashLoopBackOff.

kubectl describe pod nginx-pod

• describe : Pod එකේ metadata, container state සහ scheduler events සවිස්තරාත්මකව පෙන්වයි.

Expected output

Name:             nginx-pod
Status:           Running
IP:               10.244.1.14
Events:
  Normal  Scheduled  4m  default-scheduler  Successfully assigned default/nginx-pod to k8s-worker-01
  Normal  Pulling    4m  kubelet            Pulling image "nginx:alpine"
  Normal  Pulled     4m  kubelet            Successfully pulled image in 5.82s
  Normal  Created    4m  kubelet            Created container nginx-pod
  Normal  Started    4m  kubelet            Started container nginx-pod

Step 04: Create Declarative 3-Replica Deployment

සිංහලෙන් — ස්වයංක්‍රීයව Self-Healing වන Replicas 3ක් සහිත Deployment එකක් හදමු.

Create a managed Deployment with 3 resilient replica pods that automatically self-heals if any pod crashes.

kubectl create deployment web-app --image=nginx:alpine --replicas=3

• create deployment : ReplicaSet එකක් මඟින් පාලනය වන Deployment එකක් සාදයි.
• --replicas=3 : නිරන්තරයෙන් Pods 3ක් ක්‍රියාත්මකව තැබීමට නියෝග කරයි.

Expected output

deployment.apps/web-app created

Hands-on practice activities

Work through these yourself — tick them off as you go.

• Activity 76: Launch Imperative Standalone Pod (Drill) — Launch an imperative test Pod running lightweight Alpine Nginx.
• Activity 77: Query Active Pods with IP Details (Drill) — View active pods, their ready status, restart count, node assignment, and assigned internal pod IP.
• Activity 78: Deep Inspect Pod Manifest and Events (Drill) — Inspect low-level container image pulling, startup timing, and lifecycle events.
• Activity 79: Create Declarative 3-Replica Deployment (Drill) — Create a managed Deployment with 3 resilient replica pods.
• Activity 80: Smallest Deployable Unit Quiz (Quiz) — What is the smallest deployable compute unit in the Kubernetes object model? Answer: Pod.

Keep going: ← Lab 15 · Lab 17 → · Track your progress in the Self Guide