🚀 Flask API Deployment on Amazon EKS (Kubernetes)

This project demonstrates the deployment of a simple Flask API (simulating a Feed/Post service) on an Amazon Elastic Kubernetes Service (EKS) cluster, using Terraform to provision all the AWS infrastructure and Kubernetes manifests and use a token to post using Kubernetes secrets.

The architecture uses ECR to store the Docker image and a Load Balancer (ELB/ALB) to expose the service publicly.

You can acess the live project here: Link

---

🎯 Project Objective

1. Infrastructure as Code (IaC): Provision a VPC, an EKS Cluster, and a Node Group using Terraform.

2. Security: Use Kubernetes Secrets to inject sensitive environment variables into the Pod.

3. Deployment: Automate the application deployment and service exposure (Load Balancer type Service).

---

🏗️ Deployment Architecture

Terraform manages the following stack:

1. AWS VPC: Isolated network with public subnets (for Load Balancer/NAT GW) and private subnets (for Workers).

2. AWS ECR: Private repository to store the Docker image flask-app-aws:latest.

3. AWS EKS: Kubernetes cluster (flask-project-eks).

4. EKS Node Group: Two EC2 Workers (t2.small) in the private subnets.

5. Kubernetes Deployment: Ensures that 2 replicas of the Flask container are always running.

6. Kubernetes Service (LoadBalancer): Provisions an AWS Load Balancer that distributes external traffic on port 80 to port 5000 of the Pods. ---

🧪 Flask Application (Endpoints)

The Flask API is a simple microservice with two main endpoints:

| Endpoint | Method | Description | Security Requirement |

| :--- | :--- | :--- | :--- |

| /post | POST | Returns the post feed (simulated) | (N/A)

| :--- | :--- | :--- | :--- |

| /feed | GET | Create a new post | Requires the Secret Token |

| :--- | :--- | :--- | :--- |

📋 Project Structure (Files and Folders)

.
├── app
│ ├── app.py `Flask application core`
│ ├── posts.json `Post database`
│ └── templates `Template folder (feed and index)`
│ ├── feed.html `Feed page template`
│ └── index.html `Main page template`
├── deployment.yaml `File for deploying the Docker image on EKS`
├── Dockerfile `Standard Dockerfile for creating the Docker image with the project in the app folder`
├── ecr `Separate folder for creating the ECR (where Docker images will be stored)`
│ ├── ecr.tf `File to apply ECR via Terraform`
│ └── terraform.tfstate
├── eks.tf `File to create ECK via Terraform`
├── iam.tf `File to apply different functions to EKS via Terraform`
├── output.tf `File to check if EKS is working after running the (terraform apply) command`
├── readme.md
├── requirements.txt `Libraries to be automatically installed in Docker via the Dockerfile`
├── secret.yaml `File to apply secrets via kubectl`
├── service.yaml `File to apply services via kubectl (Public URL to access the application)`
├── terraform.tfstate
└── vpc.tf `File to apply VPC via Terraform`

🔒 The Kubernetes Secret (kubernetes_secret_v1)

The /feed endpoint is protected and requires an access token. This token is securely stored in EKS using a Secret called flask-project-token.

The Secret is injected directly into the container as an environment variable, ensuring that the sensitive value (FEED_SECRET_TOKEN) is never exposed in the code or deployment manifest.

---

⚙️ How to Run the Project

Prerequisites

• AWS CLI configured (administrator credentials and permissions).

• Terraform (v1.0+) installed.

• Docker installed (to build and push the image to ECR).

• SSH key (mykeys-2) in the us-east-1 region (referenced in main.tf).

1. Building and Deploying the Docker Image

Before running Terraform, you must build your Flask image and deploy it to the ECR:

# 1. Log in to the ECR (replace with your AWS account ID and region)

aws  ecr  get-login-password  --region  us-east-1 | docker  login  --username  AWS  --password-stdin  `YOUR_AWS_ACCOUNT_ID`.dkr.ecr.us-east-1.amazonaws.com

  

# 2. Build the image

docker  build  -t  flask-app-aws  .

  

# 3. Tag the image

docker  tag  flask-app-aws:latest [`YOUR_AWS_ACCOUNT_ID`.dkr.ecr.us-east-1.amazonaws.com/flask-app-aws:latest](https://`YOUR_AWS_ACCOUNT_ID`.dkr.ecr.us-east-1.amazonaws.com/flask-app-aws:latest)

  

# 4. Push to the ECR (the ECR is created in 'terraform apply')

docker  push [`YOUR_AWS_ACCOUNT_ID`.dkr.ecr.us-east-1.amazonaws.com/flask-app-aws:latest](https://`YOUR_AWS_ACCOUNT_ID`.dkr.ecr.us-east-1.amazonaws.com/flask-app-aws:latest)