Kubernetes & Amazon EKS Training

English

한국어
中文
日本語
Español

English

한국어
中文
日本語
Español

Appearance

EKS Cluster Creation Lab Guide ​

Difficulty: Intermediate Estimated Time: 60 minutes Last Updated: February 11, 2026

Learning Objectives ​

  • Create an EKS cluster using eksctl
  • Access the cluster with kubectl and check its status
  • Deploy a sample application
  • Safely delete the cluster

Prerequisites ​

  • [ ] AWS account and AWS CLI configured (verify with aws sts get-caller-identity)
  • [ ] eksctl installed (verify with eksctl version)
  • [ ] kubectl installed
  • [ ] Completed EKS Cluster Creation learning

Cost Warning: Operating an EKS cluster incurs AWS costs. Be sure to delete the cluster after completing the lab.

Exercise 1: eksctl Configuration Verification ​

Steps ​

Step 1.1: Check tool versions

bash
aws --version
eksctl version
kubectl version --client

Step 1.2: Verify AWS credentials

bash
aws sts get-caller-identity

Expected output:

json
{
    "UserId": "AIDACKCEVSQ6C2EXAMPLE",
    "Account": "123456789012",
    "Arn": "arn:aws:iam::123456789012:user/your-user"
}

Step 1.3: Set default region

bash
export AWS_DEFAULT_REGION=ap-northeast-2
echo "Region: $AWS_DEFAULT_REGION"
Need a hint?
  • Use aws configure list to check current configuration
  • eksctl uses CloudFormation internally
  • The IAM user needs EKS, EC2, CloudFormation, and IAM permissions

Exercise 2: EKS Cluster Creation ​

Steps ​

Step 2.1: Write cluster configuration file

bash
cat > /tmp/eks-cluster.yaml << 'EOF'
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig
metadata:
  name: lab-cluster
  region: ap-northeast-2
  version: "1.31"
managedNodeGroups:
  - name: workers
    instanceType: t3.medium
    desiredCapacity: 2
    minSize: 1
    maxSize: 3
    volumeSize: 20
EOF

Step 2.2: Create cluster

bash
eksctl create cluster -f /tmp/eks-cluster.yaml

Cluster creation takes 15-20 minutes.

Step 2.3: Verify kubeconfig

bash
kubectl config current-context
kubectl cluster-info

Verification ​

bash
kubectl get nodes
# Should display 2 Ready nodes

Exercise 3: Cluster Exploration ​

Steps ​

Step 3.1: Check node information

bash
kubectl get nodes -o wide
kubectl describe node $(kubectl get nodes -o jsonpath='{.items[0].metadata.name}')

Step 3.2: Check system components

bash
kubectl get pods -n kube-system
kubectl get svc -n kube-system

Step 3.3: Check resource usage

bash
kubectl top nodes 2>/dev/null || echo "Metrics Server is not installed"

Exercise 4: Sample App Deployment ​

Steps ​

Step 4.1: Deploy Nginx

bash
kubectl create deployment nginx --image=nginx:1.25 --replicas=2
kubectl expose deployment nginx --port=80 --type=LoadBalancer
kubectl wait --for=condition=available deployment/nginx --timeout=120s

Step 4.2: Verify access

bash
# Check LoadBalancer External IP (ELB creation takes a few minutes)
kubectl get svc nginx -w
# Press Ctrl+C once EXTERNAL-IP is assigned

# Test access
ELB_URL=$(kubectl get svc nginx -o jsonpath='{.status.loadBalancer.ingress[0].hostname}')
echo "ELB URL: $ELB_URL"
curl -s "$ELB_URL" | head -5

Step 4.3: Scaling test

bash
kubectl scale deployment nginx --replicas=4
kubectl get pods -l app=nginx -o wide
Need a hint?
  • It may take a few minutes for the ELB URL to propagate through DNS
  • Use kubectl get svc -w to monitor EXTERNAL-IP assignment in real-time
  • You can also verify in AWS Console under EC2 > Load Balancers

Verification ​

bash
kubectl get deployment nginx -o jsonpath='{.status.readyReplicas}'
# Output: 4

Cleanup ​

Important: Be sure to delete the cluster to prevent ongoing costs.

bash
# 1. Clean up application (so LoadBalancer deletes the ELB)
kubectl delete svc nginx
kubectl delete deployment nginx

# 2. Wait for ELB deletion (about 1 minute)
sleep 60

# 3. Delete cluster
eksctl delete cluster -f /tmp/eks-cluster.yaml --wait

# 4. Clean up configuration file
rm -f /tmp/eks-cluster.yaml

Troubleshooting ​

Cluster creation fails
  • Check IAM permissions (AdministratorAccess or EKS-related policies required)
  • Check VPC/subnet limits (per-region default VPC count limits)
  • Get details with eksctl utils describe-stacks --region=ap-northeast-2 --cluster=lab-cluster
kubectl cannot connect to the cluster

Manually update kubeconfig:

bash
aws eks update-kubeconfig --name lab-cluster --region ap-northeast-2

Next Steps ​