vendure/docs/docs/reference/core-plugins/telemetry-plugin/index.md

---

title: "TelemetryPlugin" isDefaultIndex: false

generated: true

import MemberInfo from '@site/src/components/MemberInfo'; import GenerationInfo from '@site/src/components/GenerationInfo'; import MemberDescription from '@site/src/components/MemberDescription';

TelemetryPlugin

The TelemetryPlugin is used to instrument the Vendure application and collect telemetry data using OpenTelemetry.

Installation

npm install @vendure/telemetry-plugin

Configuration

The plugin is configured via the TelemetryPlugin.init() method. This method accepts an options object which defines the OtelLogger options and method hooks.

Example

import { VendureConfig } from '@vendure/core';
import { TelemetryPlugin, registerMethodHooks } from '@vendure/telemetry-plugin';

export const config: VendureConfig = {
  // ...
  plugins: [
    TelemetryPlugin.init({
      loggerOptions: {
        // Log to the console at the verbose level
        console: LogLevel.Verbose,
      },
    }),
  ],
};

Preloading the SDK

In order to use the OpenTelemetry SDK, you must preload it before the Vendure server is started. This is done by using the --require flag when starting the server with a custom preload script.

Create a file named instrumentation.ts in the root of your project and add the following code:

import { OTLPLogExporter } from '@opentelemetry/exporter-logs-otlp-proto';
import { OTLPTraceExporter } from '@opentelemetry/exporter-trace-otlp-http';
import { BatchLogRecordProcessor } from '@opentelemetry/sdk-logs';
import { NodeSDK } from '@opentelemetry/sdk-node';
import { BatchSpanProcessor } from '@opentelemetry/sdk-trace-base';
import { getSdkConfiguration } from '@vendure/telemetry-plugin/preload';

// In this example we are using Loki as the OTLP endpoint for logging
process.env.OTEL_EXPORTER_OTLP_ENDPOINT = 'http://localhost:3100/otlp';
process.env.OTEL_LOGS_EXPORTER = 'otlp';
process.env.OTEL_RESOURCE_ATTRIBUTES = 'service.name=vendure-dev-server';

// We are using Jaeger as the OTLP endpoint for tracing
const traceExporter = new OTLPTraceExporter({
    url: 'http://localhost:4318/v1/traces',
});
const logExporter = new OTLPLogExporter();

// The getSdkConfiguration method returns a configuration object for the OpenTelemetry Node SDK.
// It also performs other configuration tasks such as setting a special environment variable
// to enable instrumentation in the Vendure core code.
const config = getSdkConfiguration({
    config: {
        // Pass in any custom configuration options for the Node SDK here
        spanProcessors: [new BatchSpanProcessor(traceExporter)],
        logRecordProcessors: [new BatchLogRecordProcessor(logExporter)],
    },
});

const sdk = new NodeSDK(config);

sdk.start();

The server would then be started with the following command:

node --require ./dist/instrumentation.js ./dist/server.js

or for development with ts-node:

npx ts-node --require ./src/instrumentation.ts ./src/server.ts

class TelemetryPlugin {
    static options: TelemetryPluginOptions = {};
    constructor(methodHooksService: MethodHooksService, options: TelemetryPluginOptions)
    init(options: TelemetryPluginOptions) => ;
}

### options TelemetryPluginOptions`} /> ### constructor TelemetryPluginOptions) => TelemetryPlugin`} /> ### init TelemetryPluginOptions) => `} />