Skip to content

Latest commit

 

History

History
148 lines (120 loc) · 5.84 KB

DEPLOYMENT.md

File metadata and controls

148 lines (120 loc) · 5.84 KB

Comprehensive Deployment Guide for AWS EC2 (Free Tier)

This detailed guide walks you through deploying CredibilityCheckerRAG on an AWS EC2 t2.micro instance (completely Free Tier eligible).

Prerequisites

  • An active AWS Account (Sign up here).
  • Your GitHub/GitLab Credentials (Username and Personal Access Token or Password) ready for cloning.
  • Your Mistral/OpenAI API Keys ready.

Part 1: Launch an EC2 Instance (Virtual Server)

  1. Log in to the AWS Management Console.
  2. In the search bar at the top, type EC2 and select EC2 from the results.
  3. On the EC2 Dashboard, click the orange Launch instance button.
  4. Name and tags:
    • In the "Name" field, type: CredibilityChecker-Server
  5. Application and OS Images (Amazon Machine Image):
    • Select Ubuntu.
    • Ensure "Free tier eligible" is listed under the version (e.g., Ubuntu Server 22.04 LTS). This is the safest default.
  6. Instance type:
    • Select t2.micro (1 vCPU, 1 GiB Memory). This is Free Tier eligible.
  7. Key pair (login):
    • Click Create new key pair.
    • Key pair name: credibility-key
    • Key pair type: RSA
    • Private key file format: .pem (for Mac/Linux/Powershell) or .ppk (for older PuTTY). .pem is recommended.
    • Click Create key pair. The file will download automatically. Keep this safe; you cannot download it again.
  8. Network settings:
    • Firewall (security groups): Select "Create security group".
    • Check Allow SSH traffic from. Ideally set to My IP for security, or Anywhere (0.0.0.0/0) for convenience.
    • Check Allow HTTP traffic from the internet.
    • Check Allow HTTPS traffic from the internet.
    • Crucial Step: We need to open port 5000 (Flask default). We will do this after launch to be precise.
  9. Configure storage:
    • Default is 8 GiB gp2/gp3. This is fine. You can increase to 20 GiB (Free Tier limit is 30GB total) if you want more space.
  10. Click the orange Launch instance button on the right.
  11. Click View all instances to see your new server initializing.

Part 2: Configure Security Group (Open Port 5000)

By default, standard web ports (80/443) are open, but Flask runs on 5000.

  1. In the EC2 Dashboard -> Instances, check the box next to your CredibilityChecker-Server.
  2. Click the Security tab in the bottom pane.
  3. Click the link under Security groups (it looks like sg-01234abcd...).
  4. In the new window identifying the security group, click the Inbound rules tab.
  5. Click Edit inbound rules.
  6. Click Add rule (bottom left).
    • Type: Select Custom TCP.
    • Port range: Type 5000.
    • Source: Select Anywhere-IPv4 (0.0.0.0/0).
    • Description (optional): "Flask App Port".
  7. Click Save rules.

Part 3: Connect to your Instance

The easiest way is using the browser (no keys needed).

  1. Go back to Instances.
  2. Select your instance CredibilityChecker-Server.
  3. Click the Connect button (top right).
  4. Select the EC2 Instance Connect tab.
  5. Click Connect.
  6. A new browser tab will open with a black terminal screen. You are now "inside" your server!

Part 4: Deploy the Code

Run these commands in the black terminal window.

1. Clone the Repository

Replace <YOUR_REPO_URL> with your actual repository URL.

Note: If your repo is private, it will ask for Username and Password. For Password, you essentially need a GitHub Personal Access Token.

# Example: git clone https://github.com/yourusername/CredibilityCheckerRAG.git
git clone <YOUR_REPO_URL>

Once cloned, move into the directory:

cd CredibilityCheckerRAG/v3
# OR whatever the folder name is. Type 'ls' to see the folder name.
# e.g., cd CredibilityCheckerRAG

2. Run the Setup Script

I have included a helper script to do all the heavy lifting (installing Python, creating a swap file to prevent crashing, setting up virtual environment).

# Make the script executable
chmod +x deploy_ec2.sh

# Run it
./deploy_ec2.sh

Wait for this to finish. It may take 2-5 minutes.

3. Setup Environment Variables (.env)

You need to create the .env file on the server.

  1. Open the text editor nano: 
    nano .env
  2. Now, paste your API keys. (Right-click in the terminal window usually pastes). 
    MISTRAL_API_KEY=your_actual_api_key_here
# Add any other keys from your local .env
  3. To Save and Exit nano:
    • Press Ctrl + X (to exit).
    • Press Y (to say Yes to saving).
    • Press Enter (to confirm the filename).

4. Start the Application

Run the start script created by the deploy script:

./start_app.sh

You should see output saying [INFO] Starting gunicorn... and Listening at: http://0.0.0.0:5000.

Part 5: Access Your App

  1. Go back to the AWS Console -> Instances.
  2. Find your instance and look for Public IPv4 address (e.g., 54.123.45.67).
  3. Open a new browser tab on your computer.
  4. Go to: http://<YOUR_PUBLIC_IP>:5000
    • Example: http://54.123.45.67:5000

Success! You should see your application running.

Troubleshooting

  • "This site can't be reached":
    • Did you mistakenly use https://? Make sure it is http:// (unless you set up SSL).
    • Did you do Part 2 (Security Groups) correctly? Verify Port 5000 is open to 0.0.0.0/0.
  • "MemoryError" or "Killed" during installation:
    • The t2.micro has only 1GB RAM. Ensure ./deploy_ec2.sh ran successfully—it creates a "swap file" that acts as extra RAM.
  • Deployment script permission denied:
    • Did you run chmod +x deploy_ec2.sh?