Vercel Log Drain for Google Cloud Logging

This project implements a Vercel Log Drain that forwards logs from Vercel deployments to Google Cloud Logging. It's designed to be deployed as a Google Cloud Function or on Google Cloud Run.

Setup

1. Clone the repository:

   git clone https://github.com/Ingiq/vercel-log-drain-gcp.git
   cd vercel-log-drain-gcp

2. Install dependencies:

   npm install

3. Environment Variables:

   Create a .env file based on .env.example and populate it with your values:

   VERCEL_VERIFICATION_KEY=your-verification-key-here
   VERCEL_LOG_DRAIN_SECRET=your-secret-key-here
   GOOGLE_CLOUD_PROJECT=your-gcp-project-id
   # GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account-key.json # Optional: for local development with service account
   

   • VERCEL_VERIFICATION_KEY: A unique key for Vercel to verify ownership of the log drain.
   • VERCEL_LOG_DRAIN_SECRET: A secret key used to sign log payloads from Vercel.
   • GOOGLE_CLOUD_PROJECT: Your Google Cloud Project ID where logs will be sent.

Configuration Options

Feature Flags (Optional)

The log drain includes configurable parsing features that can be controlled via environment variables:

# Optional: Disable JSON message parsing (default: enabled)
ENABLE_JSON_PARSING=false

# Optional: Disable Lambda execution log parsing (default: enabled)  
ENABLE_LAMBDA_PARSING=false

ENABLE_JSON_PARSING (default: true)

• When enabled: Parses JSON-formatted log messages to extract structured data and meaningful messages from fields like msg, message, or error
• When disabled: Uses the raw JSON string as the log message without parsing
• Use case: Disable if you prefer raw JSON messages or experience parsing issues

ENABLE_LAMBDA_PARSING (default: true)

• When enabled: Parses AWS Lambda execution logs (START/END/REPORT format) to extract performance metrics like duration, memory usage, and request IDs
• When disabled: Treats Lambda execution logs as plain text messages
• Use case: Disable if you don't need Lambda metrics or want to reduce processing overhead

Both features are enabled by default. To disable a feature, explicitly set the environment variable to 'false'. Any other value (including unset) keeps the feature enabled.

4. Google Cloud Authentication:

   Ensure your Google Cloud environment is authenticated. For local development, you might need:

   gcloud auth login
   gcloud config set project YOUR_PROJECT_ID
   gcloud auth application-default login

Running Locally

To run the function locally:

npm run dev

This will build the TypeScript code and start the function on http://localhost:8080 (or the port specified by the PORT environment variable).

Deployment to Google Cloud

This project is designed for deployment to Google Cloud Run or Google Cloud Functions.

To deploy to Google Cloud Run (ensure gcloud CLI is configured and authenticated):

npm run deploy

This command uses env-cmd to pick up environment variables from your .env file and deploys the service to Google Cloud Run.

Testing

Run unit tests using Vitest:

npm test

Run integration tests against the local server:

npm run dev
npm run test:integration

Run integration tests against Cloud Run:

npm run dev
npm run test:integration https://url-to-the-cloud-run.run.app