SimpleTimeService is a lightweight Python microservice that returns the current timestamp and the client IP address in JSON format.
The project demonstrates:
- A simple web service (Python + Flask)
- Containerization with Docker (multi-stage, distroless, non-root)
- Infrastructure as Code using Terraform
- Deployment to AWS ECS (Fargate) behind an Application Load Balancer
This repository is intended as a reference implementation for cloud-native application deployment and infrastructure automation.
GET /
Example response:
{
"timestamp": "2025-01-01T12:34:56Z",
"ip": "203.0.113.10"
}📁 Repository Structure
simple-time-service/
├── app.py # Python application
├── Dockerfile # Multi-stage, distroless Docker image
└── terraform/
├── main.tf # AWS infrastructure (VPC, ECS, ALB)
├── variables.tf # Terraform variables
├── output.tf # Terraform outputs
├── terraform.tfvars.example
└── .gitignore
🔧 Prerequisites:
The following tools are required to deploy this project. 1️⃣ Operating System Amazon Linux 2023
2️⃣ Install Git : https://git-scm.com/
sudo yum install git -y
3️⃣ Install Docker : https://docs.docker.com/get-started/get-docker/
sudo yum install docker -y
sudo systemctl start docker
sudo systemctl enable docker
sudo usermod -aG docker $USER
newgrp docker
verify if docker is installed correctly:
docker version
4️⃣ Install AWS CLI v2 (Amazon linux 2023 comes preinstalled with aws cli, so skipping this step) you can verify aws cli by -
aws --version
Link to install aws CLI if on other OS - https://docs.aws.amazon.com/cli/latest/userguide/install-cliv2.html
5️⃣ Install Terraform - https://developer.hashicorp.com/terraform/downloads
sudo yum install -y dnf-plugins-core
sudo yum config-manager --add-repo https://rpm.releases.hashicorp.com/AmazonLinux/hashicorp.repo
sudo yum install terraform -y
verify terraform installation by -
terraform version
🔐 AWS Credentials Configuration Configure an IAM role with below permissions so that EC2 can launch and create resources on your behalf (this is better option as compared to AWS configure option as anyone who has access to this machine can get your crednetials and misuse it, so its better to use IAM role attach it to instance from "modify IAM role" option available in ec2 console)
Attach below permission to IAM role-
AmazonEC2ContainerRegistryFullAccess
AmazonEC2FullAccess
AmazonECS_FullAccess
AmazonVPCFullAccess
ElasticLoadBalancingFullAccess
IAMFullAccess
🐳 Build Docker Image and Run Locally
docker build -t simple-time-service .
Run the container
docker run -p 5000:5000 simple-time-service
Test in browser http://public-ip-of-instance:5000
you should get below output -
{
"timestamp": "2025-12-19T15:39:15.408020Z",
"ip": "103.228.147.80"
}
📦 Push Image to Amazon ECR: Create ECR Repository
aws ecr create-repository \
--repository-name simple-time-service \
--region us-east-1
Authenticate Docker to ECR
aws ecr get-login-password --region us-east-1 \
| docker login --username AWS --password-stdin <ACCOUNT_ID>.dkr.ecr.us-east-1.amazonaws.com
Build, Tag, and Push image to ECR
docker build -t simple-time-service .
docker tag simple-time-service:latest <ACCOUNT_ID>.dkr.ecr.us-east-1.amazonaws.com/simple-time-service:latest
docker push <ACCOUNT_ID>.dkr.ecr.us-east-1.amazonaws.com/simple-time-service:latest
🏗️ Deploy Infrastructure with Terraform
1️⃣ Configure Terraform Variables Edit terraform.tfvars with image URI just created in previous step
image = "<ACCOUNT_ID>.dkr.ecr.us-east-1.amazonaws.com/simple-time-service:latest"
2️⃣ Initialize Terraform
cd terraform
terraform init
3️⃣ Apply Terraform
terraform apply
it will prompt for "yes" (24 resources will be created)
4️⃣ Access the Application After deployment, Terraform outputs the ALB DNS name:
terraform output
Open the URL in a browser: http://alb-dns-name
🧠 Design Considerations-->
ECS Fargate used to avoid managing servers Private subnets for application tasks Public ALB for internet access Distroless Docker image for security and size optimization Infrastructure fully reproducible via Terraform
🧹 Cleanup To destroy all AWS resources:
terraform destroy
📌 Notes: Terraform state is stored locally in this setup for simplicity. In a production environment, a remote backend (S3 + DynamoDB) is recommended. This project is intended for demonstration and learning purposes. IAM policies and security groups can be further restricted if required.