Month_End_Automator

Azure Deployment Guide for Month-End Automator

This guide will help you deploy the Month-End Automator application to Azure App Service.

Prerequisites

1. Azure Account: You need an active Azure subscription
2. Azure CLI: Install Azure CLI
3. Git: For deployment via Git (optional)
4. Python 3.9+: Locally installed for testing

Deployment Options

Option 1: Deploy via Azure Portal (Recommended for Beginners)

1. Create Azure App Service:
   • Go to Azure Portal
   • Click “Create a resource” → Search for “Web App”
   • Click “Create”
2. Configure the App Service:
   • Subscription: Select your subscription
   • Resource Group: Create new or use existing
   • Name: month-end-automator (must be globally unique)
   • Publish: Code
   • Runtime stack: Python 3.9 or 3.10
   • Operating System: Linux (recommended) or Windows
   • Region: Choose closest to your users
   • App Service Plan: Create new or use existing (Basic B1 minimum recommended)
3. Deployment:
   • Go to “Deployment Center” in your App Service
   • Choose deployment source:
     • GitHub: Connect your GitHub repository
     • Local Git: Use local Git repository
     • ZIP Deploy: Upload a ZIP file of your project
4. Configure Startup Command:
   • Go to “Configuration” → “General settings”
   • Startup Command:

     gunicorn main:app --workers 2 --worker-class uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000 --timeout 600
     

Option 2: Deploy via Azure CLI

1. Login to Azure:

   az login
   

2. Create Resource Group:

   az group create --name monthEndAutomator-rg --location eastus
   

3. Create App Service Plan:

   az appservice plan create --name monthEndAutomator-plan --resource-group monthEndAutomator-rg --sku B1 --is-linux
   

4. Create Web App:

   az webapp create --resource-group monthEndAutomator-rg --plan monthEndAutomator-plan --name month-end-automator --runtime "PYTHON|3.9"
   

5. Configure Startup Command:

   az webapp config set --resource-group monthEndAutomator-rg --name month-end-automator --startup-file "gunicorn main:app --workers 2 --worker-class uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000 --timeout 600"
   

6. Deploy Code (via ZIP):

   # Create a ZIP file (exclude __pycache__, .git, etc.)
   zip -r deploy.zip . -x "*.git*" "*__pycache__*" "*.pyc" "*.csv" "*.xlsx" "test_*" "*.md"
   
   # Deploy
   az webapp deployment source config-zip --resource-group monthEndAutomator-rg --name month-end-automator --src deploy.zip
   

Option 3: Deploy via GitHub Actions (CI/CD)

1. Create GitHub Secrets:
   • Go to your GitHub repository → Settings → Secrets
   • Add:
     • AZURE_WEBAPP_NAME: Your app service name
     • AZURE_WEBAPP_PUBLISH_PROFILE: Download from Azure Portal (Get publish profile)
     • AZURE_RESOURCE_GROUP: Your resource group name
2. Create .github/workflows/deploy.yml:

   name: Deploy to Azure App Service
   
   on:
     push:
       branches: [ main ]
   
   jobs:
     build-and-deploy:
       runs-on: ubuntu-latest
       steps:
       - uses: actions/checkout@v2
          
       - name: Set up Python
         uses: actions/setup-python@v2
         with:
           python-version: '3.9'
          
       - name: Install dependencies
         run: |
           pip install -r requirements.txt
          
       - name: Deploy to Azure Web App
         uses: azure/webapps-deploy@v2
         with:
           app-name: $
           publish-profile: $
   

Configuration

Environment Variables

Set these in Azure Portal → Configuration → Application settings:

• PORT: 8000 (optional, Azure sets this automatically)
• PYTHONPATH: /home/site/wwwroot
• WEBSITE_TIME_ZONE: UTC (or your preferred timezone)

File Storage Considerations

Important: Azure App Service has ephemeral storage. Files stored in data/uploads will be lost on restart.

Solutions:

1. Use Azure Blob Storage (Recommended):
   • Create Azure Storage Account
   • Upload files to Blob Storage instead of local filesystem
2. Use Azure Files:
   • Mount Azure Files as persistent storage
3. Use External Database:
   • Consider migrating in-memory stores to Azure SQL Database or Cosmos DB

Post-Deployment Steps

1. Verify Deployment:
   • Visit https://your-app-name.azurewebsites.net
   • Should see login page
2. Check Logs:
   • Azure Portal → App Service → Log stream
   • Or: az webapp log tail --name month-end-automator --resource-group monthEndAutomator-rg
3. Monitor:
   • Azure Portal → App Service → Application Insights (enable if needed)

Troubleshooting

Common Issues:

1. 502 Bad Gateway:
   • Check startup command is correct
   • Verify gunicorn is in requirements.txt
   • Check App Service logs
2. Module Not Found:
   • Ensure all dependencies are in requirements.txt
   • Check Python version matches locally
3. Files Not Persisting:
   • Use Azure Blob Storage for file uploads
   • Or configure Azure Files mount
4. Timeout Errors:
   • Increase timeout in startup command
   • Consider upgrading App Service Plan

Security Recommendations

1. HTTPS: Enabled by default on Azure App Service
2. Authentication: Consider enabling Azure AD authentication
3. Secrets: Store credentials in Azure Key Vault
4. Environment Variables: Use Application Settings for configuration

Scaling

• Vertical Scaling: Upgrade App Service Plan tier
• Horizontal Scaling: Enable auto-scaling in App Service Plan
• Load Balancing: Automatically handled by Azure App Service

Cost Estimates

• Basic B1 Plan: ~$13/month
• Standard S1 Plan: ~$73/month (recommended for production)
• Blob Storage: ~$0.01/GB/month

Support

For issues:

1. Check Azure Portal → App Service → Logs
2. Review Application Insights
3. Check FastAPI documentation

Next Steps

After deployment:

1. Set up monitoring and alerts
2. Configure custom domain (optional)
3. Set up backup strategy
4. Implement CI/CD pipeline
5. Consider migrating to Azure Blob Storage for file persistence

This site is open source. Improve this page.