Create Task Definition & Service

In this final step for the frontend, we define how our application container should run (Task Definition) and deploy it as a scalable service (ECS Service) connected to our Load Balancer.

1. Create Task Definition

The Task Definition serves as a blueprint for our application.

1. Navigate to Amazon ECS > Task definitions > Create new task definition.
2. Task definition configuration:
   • Task definition family: bandup-frontend.
   • Launch type: AWS Fargate.
   • OS/Architecture: Linux/X86_64.
   • Task size: .5 vCPU and 1 GB Memory (Sufficient for our Next.js frontend).
   • Task Role & Task Execution Role: Select ecsTaskExecutionRole (Created in section 5.3.3).

3. Container details:
   • Name: bandup-fe-container.
   • Image URI: Enter the ECR URI we pushed earlier (e.g., .../band-up-frontend:v1.0.0).
   • Container port: 3000.
4. Click Create.

2. Create ECS Service

Now we deploy this blueprint into our Cluster.

1. Navigate to Clusters > Select bandup-cluster.
2. In the Services tab, click Create.

Step 1: Environment

• Compute options: Launch type -> FARGATE.
• Task definition: bandup-frontend (Revision 1).
• Service name: bandup-frontend-service.
• Desired tasks: 1.

Step 2: Networking

• VPC: band-up-vpc.
• Subnets: Select the Private Subnets (private-subnet-1, private-subnet-2).
• Security group: Select ecs-private-sg (This allows traffic from ALB).

Step 3: Load Balancing

• Load balancer type: Application Load Balancer.
• Load balancer: Select bandup-public-alb.
• Container to load balance: bandup-fe-container 3000:3000.

• Listener: Create a new listener on Port 80 (HTTP).
• Target group: Use an existing target group -> target-bandup-fe.

3. Click Create. The service will start deploying your container. Wait until the status changes to Active and the Task status is Running.

3. Verify Deployment

Once the service is stable, open your web browser and navigate to the DNS name of your Application Load Balancer.

You should see the IELTS BandUp landing page loading successfully, served from your container in the private subnet.