Opentelemetry Docs

OpenTelemetry

Overview

OpenTelemetry (OTel) is a vendor-neutral, open-source observability framework for instrumenting, generating, collecting, and exporting telemetry data — traces, metrics, and logs. It is an industry standard supported by 90+ observability vendors, integrated by many libraries/services/apps, and adopted by numerous end users.

OpenTelemetry was created as a merger of OpenTracing and OpenCensus.

Key Principles

1. You own the data you generate — no vendor lock-in
2. Single set of APIs and conventions — learn once, use everywhere

Main Components

• Specification — cross-language requirements for all implementations (API, SDK, Data/OTLP, Semantic Conventions)
• OTLP Protocol — defines the shape of telemetry data transport (gRPC + HTTP/Protobuf)
• Semantic Conventions — standard naming scheme for common telemetry data types
• Language SDKs — implement the specification, APIs, and export of telemetry data (Go, .NET, Java, JavaScript, Python, PHP, Ruby, Rust, Swift, Kotlin, C++)
• Instrumentation Libraries — auto-instrument common libraries and frameworks
• Zero-code Instrumentation — generate telemetry without code changes (Go, .NET, Java, JavaScript, Python, PHP)
• OpenTelemetry Collector — vendor-agnostic proxy that receives, processes, and exports telemetry data
• OpenTelemetry Operator for Kubernetes — injects zero-code instrumentation, manages Collector deployments
• Helm Charts — Collector, Demo, Operator charts for Kubernetes

Signals

| Signal | Description | |--------|-------------| | Traces | Path of a request through your application | | Metrics | Measurements captured at runtime | | Logs | Recordings of events | | Baggage | Contextual information passed between signals | | Profiles | Resource usage at code level (in development) |

Quick Reference

Core Concepts

• Observability — understanding internal state of a system from its outputs
• Context Propagation — mechanism that moves context between services/processes (W3C TraceContext default)
• Instrumentation — code emitting signals (code-based or zero-code)
• Resources — entity producing telemetry (service name, pod, container, etc.)
• Sampling — reducing telemetry volume while maintaining representativeness

Collector Components

• Receivers — collect telemetry from various sources/formats
• Processors — transform, filter, enrich telemetry data
• Exporters — send telemetry to observability backends
• Connectors — join two pipelines (act as both exporter and receiver)
• Extensions — additional capabilities (health checks, pprof, zpages)

Deployment Patterns

• Agent — Collector runs alongside application (sidecar/DaemonSet)
• Gateway — Collector runs as standalone service (cluster/datacenter/region level)
• Agent-to-Gateway — two-tiered: agents forward to central gateway

Documentation Structure

| File | Content | |------|---------| | getting-started.md | What is OTel, observability primer, getting started paths, language SDKs overview | | concepts.md | Signals (traces, metrics, logs, baggage), context propagation, instrumentation, resources, sampling, semantic conventions, components, glossary | | collector.md | Collector overview, quick start, configuration, components (receivers/processors/exporters/connectors/extensions), deployment patterns, scaling, distributions | | api.md | OTLP protocol specification, language SDK APIs, zero-code instrumentation, migration guides, semantic conventions reference | | languages.md | Per-language SDK detailed reference: Go, Python, Java, JavaScript, .NET, PHP, Ruby, Rust, Swift, Kotlin, C++ — status, installation, SDK setup, manual/zero-code instrumentation, code examples, repositories, cross-language common patterns | | platforms.md | Kubernetes (Operator, Helm Charts, deployment patterns, target allocator, HPA, auto-instrumentation injection), FaaS (AWS Lambda, Azure, GCP), client-side apps (Android, iOS, Web), building custom Collector with OCB, OpAMP, OpenTelemetry Demo | | specification.md | OTel specification v1.59.0 detailed reference: API spec (context, tracing, metrics, logs), SDK spec (tracing, metrics, logs, resource, configuration), data spec (OTLP, metrics/logs data models, semantic conventions), compatibility (OpenCensus, OpenTracing, Prometheus), semantic conventions v1.43.0 with all domains and attribute tables | | collector-components.md | Complete Collector components reference: all receivers (90+), processors (35+), exporters (45+), connectors (15), extensions (25+) with descriptions, configuration structure, file inclusion, env vars, auth, certificates, component stability levels, distribution inclusion | | getting-started-guides.md | Per-language getting started tutorials: Go (SDK init, otel.go, custom spans), Python (opentelemetry-distro, auto-instrument, manual), Java (javaagent, env vars, manual), JavaScript/Node.js (sdk-node, auto-instrumentations, manual), JavaScript/Browser (sdk-trace-web, document-load, parcel), .NET (NuGet packages, Program.cs setup, OTLP export), cross-language patterns, environment variables table |

Official Documentation

• Docs home: https://opentelemetry.io/docs/
• Specification: https://opentelemetry.io/docs/specs/otel/
• OTLP: https://opentelemetry.io/docs/specs/otlp/
• Semantic Conventions: https://opentelemetry.io/docs/specs/semconv/
• Collector: https://opentelemetry.io/docs/collector/
• Languages: https://opentelemetry.io/docs/languages/
• Zero-code: https://opentelemetry.io/docs/zero-code/
• Registry: https://opentelemetry.io/ecosystem/registry/

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: pledgeandgrow
• Source: pledgeandgrow/pledge-skills
• License: MIT

Install and usage instructions live in the source repository linked above.