repository·main·Indexed 21 days ago
https://github.com/open-telemetry/opentelemetry-js-contribA repository for community-maintained OpenTelemetry JavaScript contributions, including instrumentations, metapackages, propagators, and resource detectors that extend the core OpenTelemetry JS SDK. It provides examples and support for instrumenting various frameworks and libraries such as Express, Koa, GraphQL, MongoDB, MySQL, Bunyan, and React/Preact.
This example demonstrates how to use OpenTelemetry Redis Instrumentation to automatically collect trace data from Redis cache calls within an Express API. It is designed to showcase distributed tracing capabilities, including:
OpenTelemetry MySQL Instrumentation provides automatic collection of trace data and metrics for applications using the mysql npm module. It enables observability for distributed systems by capturing interactions with a MySQL backend and exporting them to backends like Zipkin or Grafana.
The instrumentation supports various connection methods:
opentelemetry-js-contrib is a repository for community-maintained OpenTelemetry JavaScript contributions that are not part of the core API or SDK. It provides additional capabilities to extend OpenTelemetry functionality in JavaScript environments.
Key types of components included in this repository:
instrumentation-*): Enable automatic collection of tracing data for various libraries and frameworks.auto-*): Bundled packages (often for Node.js or Web) that simplify the installation of multiple instrumentations.propagator-*): Tools for propagating context across service boundaries.resource-detector-*): Tools to automatically collect attributes about the entity producing telemetry (e.g., Kubernetes Pod names, namespaces, or Deployment names).For core API and SDK details, refer to the official OpenTelemetry JS documentation.
The OpenTelemetry MongoDB instrumentation automatically collects trace data for operations performed via the mongo npm module. This example demonstrates observability for the following query types:
The @opentelemetry/instrumentation-graphql package provides instrumentation for Node.js applications using GraphQL. Because it instruments the GraphQL core directly, it is compatible with various GraphQL server implementations, including:
apollo-server)express-graphql)This instrumentation allows you to capture spans for GraphQL operations. In the provided example, trace data is exported using the @opentelemetry/exporter-trace-otlp-http exporter.
The @opentelemetry/resource-detector-aws package provides several detectors to automatically populate OpenTelemetry resource attributes based on the AWS environment where your process is running. It uses @opentelemetry/semantic-conventions version 1.22+ (Semantic Convention Version 1.7.0).
Supported detectors include:
The @opentelemetry/instrumentation-bunyan module provides two primary capabilities:
Supported Bunyan versions: >=1.0.0 <2
const { trace } = require('@opentelemetry/api');
const { SimpleSpanProcessor, ConsoleSpanExporter } = require('@opentelemetry/sdk-trace');
const { SimpleLogRecordProcessor, ConsoleLogRecordExporter } = require('@opentelemetry/sdk-logs');
const { NodeSDK } = require('@opentelemetry/sdk-node');
const { BunyanInstrumentation } = require('@opentelemetry/instrumentation-bunyan');
const sdk = new NodeSDK({
spanProcessors: [
new SimpleSpanProcessor(new ConsoleSpanExporter()),
],
logRecordProcessors: [
new SimpleLogRecordProcessor({ exporter: new ConsoleLogRecordExporter() }),
],
instrumentations: [
new BunyanInstrumentation({}),
]
});
sdk.start();
const bunyan = require('bunyan');
const logger = bunyan.createLogger({name: 'example'});
// Log sending in action
logger.info('hi');
// Log correlation in action
const tracer = trace.getTracer('example');
tracer.startActiveSpan('manual-span', span => {
logger.info('in a span');
// Log record will include trace_id, span_id, and trace_flags
});The BaggageSpanProcessor reads entries stored in Baggage from the parent context and adds those keys and values as attributes to the span when it starts.
Important Security Warning: Do not put sensitive information in Baggage. Baggage entries are propagated via outgoing HTTP headers, meaning any data added to Baggage will be visible in those headers and sent to external services.
LogRecord is automatically associated with that context. The OpenTelemetry Logs SDK populates the LogRecord's spanContext (including traceId, spanId, and traceFlags), allowing exported logs to be correlated with their originating traces.The AWSXRayPropagator translates OpenTelemetry SpanContext into the AWS X-Amzn-Trace-Id header format.
Note: TraceState is currently not propagated.
An example header looks like this:
X-Amzn-Trace-Id: Root=1-5759e988-bd862e3fe1be46a994272793;Parent=53995c3f42cd8ad8;Sampled=1
The header consists of three parts:
(spec-version)-(timestamp)-(UUID).spec_version: Currently only 1 is valid.timestamp: A 32-bit number in base16 format (corresponds to the first 8 characters of the OpenTelemetry trace ID).UUID: A 96-bit random number in base16 format (corresponds to the last 10 characters of the OpenTelemetry trace ID).Span ID.0 and 1. (If the value is ?, a new trace will be started).