This project demonstrates how to use Cloudflare Workers' Streaming Tail API to export worker execution traces to OpenTelemetry (OTEL) and visualize them using Jaeger.

• Real-time trace streaming from Cloudflare Workers
• Complete OTEL conversion of Cloudflare trace events
• Jaeger visualization with comprehensive span relationships
• Type-safe implementation using Cloudflare's official types
• Local development friendly setup with Docker

The solution consists of:

1. tailStream handler - Converts Cloudflare trace events to OTEL format
2. Jaeger - OTEL-compatible trace visualizer
3. Docker setup - Simple local development environment

npm install

docker compose up -d

This starts Jaeger with:

• UI: http://localhost:16686
• OTLP HTTP: http://localhost:4318

npx wrangler dev

1. Make requests to your worker: http://localhost:8787
2. View traces in Jaeger UI: http://localhost:16686
3. Select "cloudflare-worker" service to see traces

The tailStream handler processes all Cloudflare trace events:

• onset - Worker invocation start with context
• spanOpen - New span creation
• spanClose - Span completion with outcome
• attributes - Dynamic span attribute addition
• log - Console log capture
• exception - Error and exception tracking
• return - Response information
• outcome - Final execution metrics
• diagnosticChannel - Diagnostic events

Each Cloudflare event is mapped to OTEL spans with:

• Proper span relationships (parent-child)
• Rich metadata (HTTP details, performance metrics)
• Structured logging (console logs as span events)
• Error tracking (exceptions with stack traces)

• Fetch events: HTTP method, URL, status codes, CF properties
• Scheduled events: Cron expressions, execution times
• Queue events: Queue names, batch sizes
• Performance: CPU time, wall time metrics
• Script metadata: Version, tags, execution model

# Replace Jaeger with Zipkin
docker run -d -p 9411:9411 --name zipkin openzipkin/zipkin

# Update environment
OTEL_ENDPOINT=http://localhost:9411/api/v2/spans

# docker-compose.yml with OTEL Collector
version: '3.9'
services:
  otel-collector:
    image: otel/opentelemetry-collector:latest
    command: ["--config=/etc/otel-collector-config.yml"]
    volumes:
      - ./otel-collector-config.yml:/etc/otel-collector-config.yml
    ports:
      - "4317:4317"   # OTLP gRPC receiver
      - "4318:4318"   # OTLP HTTP receiver

  jaeger:
    image: jaegertracing/all-in-one:latest
    ports:
      - "16686:16686"

Modify the converter to use dynamic service names:

// In extractTags method
tags['service.name'] = onset.scriptName || 'cloudflare-worker';
tags['service.version'] = onset.scriptVersion?.id || 'unknown';
tags['service.namespace'] = onset.dispatchNamespace || 'default';