Deployment Options¶

This document outlines various ways to deploy the FinanceQuery API service. This only covers Docker, AWS Lambda, and Render deployments, but the service can also be deployed on any platform that supports Python and FastAPI.

Dependencies¶

For the full list of dependencies, see requirements.txt in the project root.

Docker Deployment¶

Docker provides a straightforward way to self-host the FinanceQuery server. A Dockerfile is included in the repository that installs all dependencies and cythonizes project files for optimal performance.

Build the Docker Image¶

docker build -t finance-query .

Run the Docker Container¶

docker run -p 8000:8000 finance-query

FinanceQuery will now be available on port 8000 at http://localhost:8000.

Docker with Logging Configuration¶

Configure logging for different environments:

DevelopmentProductionRuntime Configuration

# Build and run with development logging
docker build \
  --build-arg LOG_LEVEL=DEBUG \
  --build-arg LOG_FORMAT=text \
  --build-arg PERFORMANCE_THRESHOLD_MS=500 \
  -t finance-query .

docker run -p 8000:8000 finance-query

# Build with production settings
docker build \
  --build-arg LOG_LEVEL=INFO \
  --build-arg LOG_FORMAT=json \
  --build-arg PERFORMANCE_THRESHOLD_MS=2000 \
  -t finance-query .

# Run with additional production configuration
docker run -p 8000:8000 \
  -e REDIS_URL=redis://redis:6379 \
  -e USE_SECURITY=true \
  -e ADMIN_API_KEY=your-secure-key \
  finance-query

# Override logging settings at runtime
docker run -p 8000:8000 \
  -e LOG_LEVEL=WARNING \
  -e LOG_FORMAT=json \
  -e PERFORMANCE_THRESHOLD_MS=5000 \
  finance-query

AWS Lambda Deployment¶

Manual Deployment¶

Follow the AWS Lambda Deployment Guide.
Configure environment variables in the Lambda console:

LOG_LEVEL=INFO
LOG_FORMAT=json
PERFORMANCE_THRESHOLD_MS=3000
REDIS_URL=your-redis-endpoint
USE_SECURITY=true
ADMIN_API_KEY=your-secure-admin-key

Note: Lambda has higher latency, so consider using PERFORMANCE_THRESHOLD_MS=3000 or higher.

Automated Deployment Workflow¶

You can use the AWS Deployment Workflow from the repository:

Provide repository secrets for:
AWS_SECRET_ID
AWS_SECRET_KEY
Edit the following values in the workflow file:
AWS_REGION
ECR_REPOSITORY
FUNCTION_NAME

Render Deployment¶

Manual Deployment¶

Follow the Render Deployment Guide.
The deployment should use the Dockerfile included in the repository.
Be sure to override the CMD in the Dockerfile in your Render project settings to:
```
python -m uvicorn src.main:app --host 0.0.0.0 --port $PORT
```
Configure environment variables in Render dashboard:

LOG_LEVEL=INFO
LOG_FORMAT=json
PERFORMANCE_THRESHOLD_MS=2000
REDIS_URL=your-redis-url
USE_SECURITY=true
ADMIN_API_KEY=your-secure-admin-key

Automated Deployment Workflow¶

You can use the Render Deployment Workflow from the repository:

Provide the repository secret:
RENDER_DEPLOY_HOOK_URL
The deploy hook URL can be found in the settings of your Render project.

Monitoring and Logging¶

Production Logging Best Practices¶

For production deployments, use these recommended logging settings:

# Balanced logging for production
LOG_LEVEL=INFO
LOG_FORMAT=json
PERFORMANCE_THRESHOLD_MS=2000

Log Monitoring¶

CloudWatch (AWS Lambda)¶

AWS Lambda automatically sends logs to CloudWatch. View logs in the AWS Console:

# View recent logs
aws logs tail /aws/lambda/your-function-name --follow

# Filter for errors
aws logs filter-log-events --log-group-name /aws/lambda/your-function-name \
  --filter-pattern "ERROR"

Docker Logging¶

For Docker deployments, configure log drivers:

# Using JSON file driver with rotation
docker run -p 8000:8000 \
  --log-driver json-file \
  --log-opt max-size=10m \
  --log-opt max-file=3 \
  finance-query

# View logs
docker logs -f container-name

Render Logging¶

Render provides built-in log streaming in the dashboard. You can also access logs via CLI:

# Install Render CLI
npm install -g @render-services/cli

# View logs
render logs --service your-service-name --tail

Troubleshooting¶

Common Issues¶

High CPU/Memory Usage: - Set LOG_LEVEL=WARNING to reduce log volume - Increase PERFORMANCE_THRESHOLD_MS to reduce warnings

Missing Performance Data: - Ensure LOG_FORMAT=json for structured data - Lower PERFORMANCE_THRESHOLD_MS for more monitoring

External API Issues: - Monitor logs for "External API FAILED" messages - Check correlation IDs to trace failed requests

For detailed logging configuration and troubleshooting, see the Logging Documentation.