分布式追踪示例

# Jaeger Configuration File
# Complete setup for production distributed tracing

version: '3.8'

services:
  # Jaeger Collector
  jaeger-collector:
    image: jaegertracing/jaeger-collector:1.45
    environment:
      - SPAN_STORAGE_TYPE=elasticsearch
      - ES_SERVER_URLS=http://elasticsearch:9200
      - ES_USERNAME=elastic
      - ES_PASSWORD=changeme
      - ES_INDEX_PREFIX=jaeger
      - COLLECTOR_ZIPKIN_HOST_PORT=:9411
      - COLLECTOR_OTLP_ENABLED=true
    ports:
      - "14269:14269"
      - "14268:14268"
      - "14250:14250"
      - "9411:9411"
    depends_on:
      - elasticsearch

  # Jaeger Agent
  jaeger-agent:
    image: jaegertracing/jaeger-agent:1.45
    command: ["--reporter.grpc.host-port=jaeger-collector:14250"]
    ports:
      - "6831:6831/udp"
      - "6832:6832/udp"
    depends_on:
      - jaeger-collector

  # Jaeger Query Service
  jaeger-query:
    image: jaegertracing/jaeger-query:1.45
    environment:
      - SPAN_STORAGE_TYPE=elasticsearch
      - ES_SERVER_URLS=http://elasticsearch:9200
      - ES_USERNAME=elastic
      - ES_PASSWORD=changeme
      - ES_INDEX_PREFIX=jaeger
      - JAEGER_UI_CONFIG_JSON=/etc/jaeger-ui-config.json
    ports:
      - "16686:16686"
      - "16687:16687"
    volumes:
      - ./jaeger-ui-config.json:/etc/jaeger-ui-config.json
    depends_on:
      - elasticsearch

  # Elasticsearch Backend
  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.8.0
    environment:
      - discovery.type=single-node
      - ES_JAVA_OPTS=-Xms1g -Xmx1g
      - xpack.security.enabled=false
    ports:
      - "9200:9200"
    volumes:
      - elasticsearch_data:/usr/share/elasticsearch/data

volumes:
  elasticsearch_data:

# OpenTelemetry Auto-Instrumentation Configuration
# Example configuration for Java applications

# Environment Variables
export OTEL_SERVICE_NAME=my-service
export OTEL_RESOURCE_ATTRIBUTES=service.name=my-service,service.version=1.0.0
export OTEL_EXPORTER_OTLP_ENDPOINT=http://otel-collector:4317
export OTEL_TRACES_SAMPLER=traceidratio
export OTEL_TRACES_SAMPLER_ARG=1.0

# Java Agent Configuration
-javaagent:/app/opentelemetry-javaagent.jar -Dotel.service.name=my-service -Dotel.resource.attributes=service.name=my-service,service.version=1.0.0 -Dotel.exporter.otlp.endpoint=http://otel-collector:4317 -Dotel.traces.sampler=traceidratio -Dotel.traces.sampler.arg=1.0 -Dotel.javaagent.extensions=io.opentelemetry.extension.noopapi.NoOpApiExtension

# Custom Span Creation (Java Example)
import io.opentelemetry.api.OpenTelemetry;
import io.opentelemetry.api.trace.Span;
import io.opentelemetry.api.trace.Tracer;

Tracer tracer = OpenTelemetry.getGlobalTracer("my-service");
Span span = tracer.spanBuilder("custom-operation")
    .setAttribute("user.id", "12345")
    .setAttribute("operation.type", "business")
    .startSpan();

// Do work here...
span.end();

{
  "trace_id": "550e8400-e29b-41d4-a716-446655440000",
  "spans": [
    {
      "trace_id": "550e8400-e29b-41d4-a716-446655440000",
      "span_id": "550e8400-e29b-41d4-a716-446655440001",
      "parent_span_id": null,
      "operation_name": "HTTP GET /api/orders",
      "service_name": "api-gateway",
      "start_time": 1701954645123000000,
      "end_time": 1701954645156000000,
      "duration_ms": 33,
      "tags": {
        "http.method": "GET",
        "http.url": "/api/orders",
        "http.status_code": "200",
        "component": "netty"
      },
      "logs": [
        {
          "timestamp": 1701954645140000000,
          "level": "info",
          "message": "Request authenticated successfully"
        }
      ]
    },
    {
      "trace_id": "550e8400-e29b-41d4-a716-446655440000",
      "span_id": "550e8400-e29b-41d4-a716-446655440002",
      "parent_span_id": "550e8400-e29b-41d4-a716-446655440001",
      "operation_name": "UserService.GetUser",
      "service_name": "user-service",
      "start_time": 1701954645130000000,
      "end_time": 1701954645145000000,
      "duration_ms": 15,
      "tags": {
        "user.id": "12345",
        "db.system": "postgresql",
        "operation.type": "query"
      }
    },
    {
      "trace_id": "550e8400-e29b-41d4-a716-446655440000",
      "span_id": "550e8400-e29b-41d4-a716-446655440003",
      "parent_span_id": "550e8400-e29b-41d4-a716-446655440001",
      "operation_name": "OrderService.GetOrders",
      "service_name": "order-service",
      "start_time": 1701954645132000000,
      "end_time": 1701954645152000000,
      "duration_ms": 20,
      "tags": {
        "user.id": "12345",
        "operation.type": "business"
      }
    }
  ]
}

# Performance Analysis Configuration
# Example setup for performance bottleneck detection

# Custom Span Attributes
attributes:
  - database.query.duration
  - cache.hit_ratio
  - memory.usage
  - cpu.utilization

# Performance Thresholds
thresholds:
  slow_query: 100ms
  high_memory: 80%
  high_cpu: 90%

# Alert Configuration
alerts:
  - name: Slow Database Queries
    condition: span.duration > 100ms
    severity: warning

# OpenTelemetry Collector Configuration
receivers:
  otlp:
    protocols:
      grpc:
      http:

processors:
  memory_limiter:
    check_interval: 5s
    limit_mib: 512
  batch:
    timeout: 5s
    send_batch_size: 1024
  attributes:
    actions:
      - key: environment
        value: production
        action: insert

exporters:
  otlp/jaeger:
    endpoint: jaeger-collector:4317
    tls:
      insecure: true
  logging:
    loglevel: info

service:
  pipelines:
    traces:
      receivers: [otlp]
      processors: [memory_limiter, batch, attributes]
      exporters: [otlp/jaeger, logging]

// Context propagation across services (Node.js + OpenTelemetry)
import { context, propagation, trace } from '@opentelemetry/api'

const tracer = trace.getTracer('orders-service')

export async function handleIncomingRequest(req, res) {
  // Extract trace context from inbound HTTP headers
  const parentContext = propagation.extract(context.active(), req.headers)

  await context.with(parentContext, async () => {
    const span = tracer.startSpan('handle-order-request')

    try {
      span.setAttribute('http.route', req.url)
      span.setAttribute('service.name', 'orders-service')

      // Inject context when calling downstream service
      const headers = {}
      const spanContext = trace.setSpan(context.active(), span)
      propagation.inject(spanContext, headers)

      const response = await fetch('http://inventory-service/check', {
        method: 'POST',
        headers: {
          ...headers,
          'content-type': 'application/json'
        },
        body: JSON.stringify({ sku: req.body.sku, quantity: req.body.quantity })
      })

      if (!response.ok) {
        span.setAttribute('error', true)
        span.setAttribute('http.status_code', response.status)
      }

      res.status(200).json({ ok: true })
    } catch (error) {
      span.recordException(error)
      span.setAttribute('error', true)
      res.status(500).json({ ok: false })
    } finally {
      span.end()
    }
  })
}