OpenTelemetry Sample Apps

A curated set of minimal, runnable samples that show how to instrument applications with OpenTelemetry across multiple languages and frameworks. Each sample is self-contained with clear run instructions and uses OTLP exporters by default.

Repository layout

• templates/ – shared templates and checklists for new samples.
• golang/, python/, nodejs/, flask/, fastapi/, dot-net/, java/ – language/framework-specific samples organized by app folder.
• Every sample folder includes its own README.md with setup, run, and instrumentation notes.

Suggested tree (what we have now):

examples/
├── README.md
├── templates/
│   └── SAMPLE_README.md
├── golang/
├── python/
├── nodejs/
├── flask/
├── fastapi/
├── dot-net/
└── java/

Conventions for samples

• Prefer OTLP exporters with env var config (OTEL_EXPORTER_OTLP_ENDPOINT, OTEL_EXPORTER_OTLP_HEADERS).
• Always set OTEL_SERVICE_NAME and useful resource attributes (e.g., deployment.environment).
• Keep samples minimal: one service, in-memory state, no external DB unless the sample is about DB instrumentation.
• Show both manual spans and auto-instrumentation when possible; note which is enabled by default.
• Include docker compose or a local OTLP/collector hint when practical.
• Add a short “What this demonstrates” section and expected telemetry (traces/metrics/logs).

How to add a new sample

1. Pick the language/framework folder (or create a new one) and add an app directory with a descriptive name.
2. Copy templates/SAMPLE_README.md into the app folder and fill it out.
3. Keep runtimes pinned (e.g., Go 1.22, Node 20, Python 3.11) and list them.
4. Provide a one-command start (make run, npm run start, go run main.go, etc.).
5. Add minimal validation instructions (curl the endpoint, view spans/metrics/logs).