Monitoring and Observability

WunderGraph exposes several metrics that allow you to get a better understanding of your application's performance. The following metrics are available:

• Served HTTP requests (tagged per operation and status code)
• Size of incoming HTTP requests
• Size of outgoing HTTP responses
• Duration of received HTTP requests
• Outgoing HTTP requests sent to upstream APIs (tagged per host and status code)
• Duration of outgoing HTTP requests (tagged per host and status code)

By default, Prometheus is enabled and serves its metrics on http://<host>:8881/metrics. To disable it or to use a custom port, use the prometheus field when calling configureWunderGraphApplication():

// wundergraph.config.ts

configureWunderGraphApplication({
  options: {
    prometheus: {
      enabled: true, // Set to false to disable
      port: 8881, // Defaults to 8881
    },
  },
});

WunderGraph supports OpenTelemetry out of the box. OpenTelemetry is a collection of tools, APIs, and SDKs used to instrument, generate, collect, and export telemetry data (metrics, logs, and traces) for analysis in order to understand your software's performance and behavior. With the help of OpenTelemetry, you can monitor your WunderGraph nodes and get a better understanding of your application's performance. We instrument all outgoing HTTP requests and inner service calls with OpenTelemetry. This includes:

• All Hooks
• TypeScript functions
• GraphQL Operations
• Webhooks
• Calls made with the Operations Client
• Origin requests

If you are not familiar with OpenTelemetry, you can read more about it here . The OpenTelemetry integration is disabled by default. To enable it, set enabled to true and optionally configure the endpoint and sampling rate:

// wundergraph.config.ts

configureWunderGraphApplication({
  options: {
    openTelemetry: {
      // Disabled by default
      enabled: true,
      // Defaults to 1 (every trace request). Samples a given fraction of traces. Must be a value between 0 and 1.
      // For example, a value of 0.1 means 10% of traces are sampled. Don't set this to 1 in production, unless you want to trace every request.
      sampler: 1,
      // Endpoint to the OTLP http endpoint. Defaults to http://localhost:4318
      exporterHttpEndpoint: 'http://localhost:4318',
    },
  },
});

We support any OpenTelemetry backend that supports the OTLP http protocol. This includes, but is not limited to:

• OpenTelemetry Collector
• Jaeger

You can also use the OpenTelemetry Collector to export your traces to any other backend that is supported by the collector.

If you have the need to authenticate with the OpenTelemetry endpoint, you can do so by providing an authToken. Currently, we only support JWT authentication in combination with our JWT authenticator plugin.

// wundergraph.config.ts

configureWunderGraphApplication({
  options: {
    openTelemetry: {
      enabled: true,
      exporterHttpEndpoint: 'https://your-collector-endpoint.com',
      authToken: 'jwt-token', // Used to authenticate with the OpenTelemetry endpoint in form of a Bearer token
    },
  },
});