Kubernetes Networking, Services, Scaling & Rolling Updates

Lab 17 of 20 · K8s Services & Scaling · 5 hands-on activities · bilingual (English + සිංහල)

පිටුව 17: Kubernetes Networking, Services, Autoscaling සහ Zero-Downtime Rolling Updates

Pods වල IPs නිතර වෙනස් වෙන ප්‍රශ්නය විසඳන Services (ClusterIP, NodePort, LoadBalancer), CoreDNS, සහ Users ලට නොදැනී Update කරන Rolling Updates.

Architectural Foundation & Systems Diagram

Pods are ephemeral. When a pod crashes or is rescheduled, it receives a brand new IP address. You cannot hardcode pod IPs! To route traffic reliably, Kubernetes provides Services.

• 🌐 Service Types — • ClusterIP: Default internal IP accessible only inside the cluster.
  • NodePort: Exposes port (30000-32767) on every node.
  • LoadBalancer: Provisions a real AWS/GCP external Load Balancer.

• 📈 Scaling Replicas (HPA) — Scale workloads with kubectl scale deployment web-app --replicas=10, or configure the Horizontal Pod Autoscaler (HPA) to dynamically scale based on CPU/RAM saturation.

• 🔄 Zero-Downtime Rolling Updates — When deploying a new version, Kubernetes spins up new version pods before terminating old pods. Customers experience 0 seconds of downtime! If a bug is caught, roll back instantly: kubectl rollout undo.

Step-by-step practical lab

Step 01: Expose Deployment as a NodePort Service

සිංහලෙන් — Deployment එකට පිටතින් ඇතුල් විය හැකි NodePort Service එකක් සාදමු.

Create a stable network service abstraction with a dedicated NodePort accessible from outside the cluster.

kubectl expose deployment web-app --port=80 --target-port=80 --type=NodePort

• expose : Pods සඳහා ස්ථාවර IP එකක් සහ Port එකක් ලබාදෙන Service එකක් සාදයි.
• --type=NodePort : Cluster එකේ සෑම Node එකකම 30000-32767 අතර port එකක් විවෘත කරයි.

Expected output

service/web-app exposed

Step 02: Inspect Cluster Services and Assigned NodePorts

සිංහලෙන් — Cluster එකේ ඇති Services සහ Port අංක පරීක්ෂා කරමු.

Display all registered services, assigned virtual ClusterIPs, and exposed external ports.

kubectl get svc

• get svc : Services ලැයිස්තුව, ClusterIPs සහ mapped ports පෙන්වයි.

Expected output

NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1                443/TCP        14d
web-app      NodePort    10.104.182.45            80:31842/TCP   45s

Step 03: Scale Deployment Replicas Under High Load

සිංහලෙන් — පාරිභෝගික තදබදය වැඩි වූ විට Pods ගණන 3 සිට 5 දක්වා ක්ෂණිකව වැඩි කරමු.

Scale deployment replica count from 3 up to 5 pods in real time to absorb sudden traffic spikes.

kubectl scale deployment web-app --replicas=5

• scale : Deployment එකක replica count එක ක්ෂණිකව වැඩි හෝ අඩු කරයි.

Expected output

deployment.apps/web-app scaled to 5 replicas

Step 04: Execute Zero-Downtime Rolling Update

සිංහලෙන් — Users ලට නොදැනී සයිට් එකේ අලුත් Version එක දාන Zero-Downtime Rolling Update එකක් කරමු.

Trigger a zero-downtime rolling update to upgrade container image version from nginx:alpine to nginx:1.25.

kubectl set image deployment/web-app nginx=nginx:1.25

• set image : Pods එකින් එක යාවත්කාලීන කරමින් zero-downtime rolling update එකක් ආරම්භ කරයි.

Expected output

deployment.apps/web-app image updated

Step 05: Track Rolling Update Rollout Status

සිංහලෙන් — Rolling update එක සාර්ථකව අවසන් වූ බව තහවුරු කර ගනිමු.

Watch the rollout progress in real time and verify all 5 new replica pods are healthy.

kubectl rollout status deployment/web-app

• rollout status : Rolling update එකේ ප්‍රගතිය සජීවීව බලා සිටියි.

Expected output

deployment "web-app" successfully rolled out! (Zero-Downtime achieved)

Hands-on practice activities

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

• Activity 81: Expose Deployment as NodePort Service (Drill) — Create a stable network service abstraction with an external NodePort.
• Activity 82: Inspect Cluster Services and Ports (Drill) — Display all registered services, assigned virtual ClusterIPs, and exposed external ports.
• Activity 83: Scale Deployment Replicas (Drill) — Scale deployment replica count from 3 up to 5 pods to absorb increased customer traffic.
• Activity 84: Execute Zero-Downtime Rolling Update (Drill) — Trigger a zero-downtime rolling update to upgrade container image version.
• Activity 85: Default Service Type Quiz (Quiz) — What is the default Kubernetes Service type used for internal cluster communication? Answer: ClusterIP.

Keep going: ← Lab 16 · Lab 18 → · Track your progress in the Self Guide