Skip to content

QUICK_DEPLOY.md

Latest commit

 

History

History
96 lines (74 loc) · 2.98 KB

QUICK_DEPLOY.md

File metadata and controls

96 lines (74 loc) · 2.98 KB

AWS App Runner Deployment via GitHub Actions

Quick Solution for GitHub Connection Issues

Since AWS App Runner GitHub connection is experiencing loading issues, we can use GitHub Actions + ECR deployment as an alternative.

Step 1: Setup AWS Credentials in GitHub

  1. Get AWS Access Keys:

    # In AWS Console → IAM → Users → Your User → Security Credentials
# Create Access Key for CLI/SDK usage

  2. Add GitHub Secrets:

    • Go to: https://github.com/hk-dev13/project-permit-api/settings/secrets/actions
    • Add these secrets:
      • AWS_ACCESS_KEY_ID: Your AWS Access Key
      • AWS_SECRET_ACCESS_KEY: Your AWS Secret Key

Step 2: Run GitHub Action

  1. Go to GitHub Actions:

    • URL: https://github.com/hk-dev13/project-permit-api/actions
    • Click on "Deploy to AWS App Runner" workflow
    • Click "Run workflow" → "Run workflow"

  2. The action will:

    • Build Docker image using Dockerfile.apprunner
    • Push to Amazon ECR
    • Provide image URI for App Runner

Step 3: Create App Runner Service with ECR

After GitHub Action completes:

  1. In AWS App Runner Console:

    • Create service
    • Source: "Container registry"
    • Provider: Amazon ECR
    • Container image URI: (from GitHub Action output)
    • Port: 8000

  2. Service Configuration:

    Service name: permit-api-service
CPU: 1 vCPU
Memory: 2 GB
Port: 8000
Environment variables:
  FLASK_ENV: production
  FLASK_DEBUG: 0
  PORT: 8000

Alternative: Manual ECR Push (Local Deployment)

Prerequisites

  1. Install Docker Desktop (if not installed)
  2. AWS CLI Setup: 
    # AWS CLI already installed via pip in venv
.\.venv\Scripts\python.exe -m awscli configure

Manual Deployment Steps

# Run the automated deployment script
.\deploy-ecr.ps1

# Or manual steps:
# 1. Test AWS credentials
.\.venv\Scripts\python.exe -m awscli sts get-caller-identity

# 2. Create ECR repository
.\.venv\Scripts\python.exe -m awscli ecr create-repository --repository-name permit-api --region ap-southeast-2

# 3. Get ECR login
$AWS_ACCOUNT_ID = (.\.venv\Scripts\python.exe -m awscli sts get-caller-identity --query Account --output text)
$LOGIN_PASSWORD = (.\.venv\Scripts\python.exe -m awscli ecr get-login-password --region ap-southeast-2)
echo $LOGIN_PASSWORD | docker login --username AWS --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.ap-southeast-2.amazonaws.com"

# 4. Build and push
docker build -f Dockerfile.apprunner -t permit-api .
docker tag permit-api "$AWS_ACCOUNT_ID.dkr.ecr.ap-southeast-2.amazonaws.com/permit-api:latest"
docker push "$AWS_ACCOUNT_ID.dkr.ecr.ap-southeast-2.amazonaws.com/permit-api:latest"

Next Steps

  1. Setup GitHub Secrets (AWS credentials)
  2. Run GitHub Action to build and push Docker image
  3. Create App Runner service using ECR image URI
  4. Test deployment once service is running

This approach bypasses the GitHub connection loading issue completely!