@opentelemetry/sdk-node

What is @opentelemetry/sdk-node?

The @opentelemetry/sdk-node package is part of the OpenTelemetry project, which aims to provide a set of APIs, libraries, agents, and instrumentation to provide observability for applications. Specifically, the @opentelemetry/sdk-node package is designed for Node.js applications and allows developers to collect traces and metrics, and export them to various observability backends for analysis. This helps in monitoring applications, understanding performance bottlenecks, and troubleshooting issues.

What are @opentelemetry/sdk-node's main functionalities?

Tracing

This feature allows the collection of trace data, which represents the lifecycle of a request through the system. The code sample initializes a tracer provider, sets up a span processor with a Jaeger exporter, and registers the provider. This setup enables the tracing of operations and requests, facilitating the monitoring and debugging of distributed systems.

const { NodeTracerProvider } = require('@opentelemetry/sdk-node');
const { SimpleSpanProcessor } = require('@opentelemetry/tracing');
const { JaegerExporter } = require('@opentelemetry/exporter-jaeger');

const provider = new NodeTracerProvider();
provider.addSpanProcessor(new SimpleSpanProcessor(new JaegerExporter()));
provider.register();

Metrics

This feature enables the collection and export of metrics data, such as counters, gauges, and histograms. The code sample demonstrates how to set up a MeterProvider with a Prometheus exporter, which collects metrics data and makes it available for scraping by Prometheus at a specified port. This is useful for monitoring application performance and resource usage.

const { MeterProvider } = require('@opentelemetry/sdk-node');
const { PrometheusExporter } = require('@opentelemetry/exporter-prometheus');

const meterProvider = new MeterProvider({
  exporter: new PrometheusExporter({ port: 9464 }),
  interval: 1000
});
meterProvider.start();

Other packages similar to @opentelemetry/sdk-node

jaeger-client

Jaeger-client is a Node.js library for reporting tracing data to Jaeger, a distributed tracing system. While it is focused specifically on integration with Jaeger, @opentelemetry/sdk-node provides a more flexible approach, allowing for the export of traces to multiple backends, not just Jaeger.

prom-client

Prom-client is a Node.js client library for Prometheus metrics. It is similar to the metrics functionality provided by @opentelemetry/sdk-node but is dedicated solely to Prometheus metrics collection and exposition. In contrast, @opentelemetry/sdk-node supports multiple metrics backends and integrates with the broader OpenTelemetry ecosystem for observability.

README

OpenTelemetry SDK for Node.js

Note: This is an experimental package under active development. New releases may include breaking changes.

This package provides the full OpenTelemetry SDK for Node.js including tracing and metrics.

Quick Start

To get started you need to install @opentelemetry/sdk-node, a metrics and/or tracing exporter, and any appropriate instrumentation for the node modules used by your application.

Installation

$ # Install the SDK
$ npm install @opentelemetry/sdk-node

$ # Install exporters and plugins
$ npm install \
    @opentelemetry/exporter-jaeger \ # add tracing exporters as needed
    @opentelemetry/exporter-prometheus \ # add metrics exporters as needed
    @opentelemetry/instrumentation-http # add instrumentations as needed

$ # or install all officially supported core and contrib plugins
$ npm install @opentelemetry/auto-instrumentations-node

Note: this example is for Node.js. See examples/opentelemetry-web for a browser example.

Initialize the SDK

Before any other module in your application is loaded, you must initialize the SDK. If you fail to initialize the SDK or initialize it too late, no-op implementations will be provided to any library which acquires a tracer or meter from the API.

This example uses Jaeger and Prometheus, but exporters exist for other tracing backends. As shown in the installation instructions, exporters passed to the SDK must be installed alongside @opentelemetry/sdk-node.

const opentelemetry = require("@opentelemetry/sdk-node");
const { JaegerExporter } = require("@opentelemetry/exporter-jaeger");
const { PrometheusExporter } = require("@opentelemetry/exporter-prometheus");
const {
  getNodeAutoInstrumentations,
} = require("@opentelemetry/auto-instrumentations-node");

const jaegerExporter = new JaegerExporter();
const prometheusExporter = new PrometheusExporter();

const sdk = new opentelemetry.NodeSDK({
  // Optional - if omitted, the tracing SDK will be initialized from environment variables
  traceExporter: jaegerExporter,
  // Optional - If omitted, the metrics SDK will not be initialized
  metricReader: prometheusExporter,
  // Optional - you can use the metapackage or load each instrumentation individually
  instrumentations: [getNodeAutoInstrumentations()],
  // See the Configuration section below for additional  configuration options
});

sdk.start();

// You can also use the shutdown method to gracefully shut down the SDK before process shutdown
// or on some operating system signal.
const process = require("process");
process.on("SIGTERM", () => {
  sdk
    .shutdown()
    .then(
      () => console.log("SDK shut down successfully"),
      (err) => console.log("Error shutting down SDK", err)
    )
    .finally(() => process.exit(0));
});

Configuration

Below is a full list of configuration options which may be passed into the NodeSDK constructor;

autoDetectResources

Detect resources automatically from the environment using the default resource detectors. Default true.

contextManager

Use a custom context manager. Default: AsyncHooksContextManager

textMapPropagator

Use a custom propagator. Default: CompositePropagator using W3C Trace Context and Baggage

metricReader

Add a MetricReader that will be passed to the MeterProvider. If metricReader is not configured, the metrics SDK will not be initialized and registered.

views

A list of views to be passed to the MeterProvider. Accepts an array of View-instances. This parameter can be used to configure explicit bucket sizes of histogram metrics.

instrumentations

Configure instrumentations. By default none of the instrumentation is enabled, if you want to enable them you can use either metapackage or configure each instrumentation individually.

resource

Configure a resource. Resources may also be detected by using the autoDetectResources method of the SDK.

resourceDetectors

Configure resource detectors. By default, the resource detectors are [envDetector, processDetector, hostDetector]. NOTE: In order to enable the detection, the parameter autoDetectResources has to be true.

If resourceDetectors was not set, you can also use the environment variable OTEL_NODE_RESOURCE_DETECTORS to enable only certain detectors, or completely disable them:

• env
• host
• os
• process
• serviceinstance (experimental)
• all - enable all resource detectors above
  • NOTE: future versions of @opentelemetry/sdk-node may include additional detectors that will be covered by this scope.
• none - disable resource detection

For example, to enable only the env, host detectors:

export OTEL_NODE_RESOURCE_DETECTORS="env,host"

sampler

Configure a custom sampler. By default, all traces will be sampled.

spanProcessor

Deprecated, please use spanProcessors instead.

spanProcessors

An array of span processors to register to the tracer provider.

traceExporter

Configure a trace exporter. If an exporter is configured, it will be used with a BatchSpanProcessor. If an exporter OR span processor is not configured programmatically, this package will auto setup the default otlp exporter with http/protobuf protocol with a BatchSpanProcessor.

spanLimits

Configure tracing parameters. These are the same trace parameters used to configure a tracer.

serviceName

Configure the service name.

Disable the SDK from the environment

Disable the SDK by setting the OTEL_SDK_DISABLED environment variable to true.

Configure log level from the environment

Set the log level by setting the OTEL_LOG_LEVEL environment variable to enums:

• NONE,
• ERROR,
• WARN,
• INFO,
• DEBUG,
• VERBOSE,
• ALL.

The default level is INFO.

Configure Trace Exporter from environment

This is an alternative to programmatically configuring an exporter or span processor. This package will auto setup the default otlp exporter with http/protobuf protocol if traceExporter or spanProcessor hasn't been passed into the NodeSDK constructor.

Exporters

Environment variable	Description
OTEL_TRACES_EXPORTER	List of exporters to be used for tracing, separated by commas. Options include otlp, jaeger, zipkin, and none. Default is otlp. none means no autoconfigured exporter.

OTLP Exporter

Environment variable	Description
OTEL_EXPORTER_OTLP_PROTOCOL	The transport protocol to use on OTLP trace, metric, and log requests. Options include grpc, http/protobuf, and http/json. Default is http/protobuf.
OTEL_EXPORTER_OTLP_TRACES_PROTOCOL	The transport protocol to use on OTLP trace requests. Options include grpc, http/protobuf, and http/json. Default is http/protobuf.
OTEL_EXPORTER_OTLP_METRICS_PROTOCOL	The transport protocol to use on OTLP metric requests. Options include grpc, http/protobuf, and http/json. Default is http/protobuf.

Additionally, you can specify other applicable environment variables that apply to each exporter such as the following:

• OTLP exporter environment configuration
• Zipkin exporter environment configuration
• Jaeger exporter environment configuration

Useful links

• For more information on OpenTelemetry, visit: https://opentelemetry.io/
• For more about OpenTelemetry JavaScript: https://github.com/open-telemetry/opentelemetry-js
• For help or feedback on this project, join us in GitHub Discussions

License

Apache 2.0 - See LICENSE for more information.