腾讯mini项目-【指标监控服务重构】2023-08-24

今日已办

Jeager

功能

1. 监控分布式工作流程并排除故障
2. 识别性能瓶颈
3. 追踪根本原因
4. 分析服务依赖关系

部署

• 部署 Deployment --- Jaeger documentation (jaegertracing.io)

• 支持 clickhouse jaegertracing/jaeger-clickhouse: Jaeger ClickHouse storage plugin implementation (github.com)

• 使用 prometheus 监控 Service Performance Monitoring (SPM) --- Jaeger documentation (jaegertracing.io)

• 使用 elasticsearch docker - How to configure Jaeger with elasticsearch? - Stack Overflow

• github issue jaeger-collector: Failed to init storage factory · Issue #931 · jaegertracing/jaeger (github.com)

version: "3"

services:
  proxy:
    image: traefik:v3.0
    container_name: proxy
    hostname: proxy
    networks:
      - elastic-jaeger
    ports:
      - "80:80"
      - "8080:8080"
    command:
      - --ping=true
      - --api.dashboard=true
      - --api.insecure=true
      - --providers.file.directory=/etc/traefik
      - --providers.file.watch=true
      - --entrypoints.web-entrypoint.address=:80
      - --entrypoints.kafka-entrypoint.address=:9092
      - --accesslog=true
      - --metrics.openTelemetry=true
      - --metrics.openTelemetry.address=jaeger-collector:4317
      - --metrics.openTelemetry.grpc=true
      - --metrics.openTelemetry.insecure=true
      - --tracing.openTelemetry=true
      - --tracing.openTelemetry.address=jaeger-collector:4317
      - --tracing.openTelemetry.grpc=true
      - --tracing.openTelemetry.insecure=true
      - --log.level=WARN # DEBUG, INFO, WARN, ERROR, FATAL, PANIC
    healthcheck:
      test: [ "CMD-SHELL", "traefik healthcheck --ping" ]
      interval: 5s
      timeout: 3s
      retries: 3
    volumes:
      - ./config/traefik:/etc/traefik

  elasticsearch:
    image: elasticsearch:7.17.12
    container_name: elasticsearch
    networks:
      - elastic-jaeger
    ports:
      - "127.0.0.1:9200:9200"
      - "127.0.0.1:9300:9300"
    restart: on-failure
    environment:
      - cluster.name=jaeger-cluster
      - discovery.type=single-node
      - http.host=0.0.0.0
      - transport.host=127.0.0.1
      - ES_JAVA_OPTS=-Xms512m -Xmx512m
      - xpack.security.enabled=false
    volumes:
      - esdata:/usr/share/elasticsearch/data

  jaeger-collector:
    container_name: jaeger-collector
    image: jaegertracing/jaeger-collector
    ports:
      - "14269:14269"
      - "14268:14268"
      - "14267:14267"
      - "14250:14250"
      - "9411:9411"
      - "4317:4317"
    networks:
      - elastic-jaeger
    restart: on-failure
    environment:
      - SPAN_STORAGE_TYPE=elasticsearch
    command: [
      "--es.server-urls=http://elasticsearch:9200",
      "--es.num-shards=1",
      "--es.num-replicas=0",
      "--log-level=error"
    ]
    depends_on:
      - elasticsearch

  jaeger-agent:
    container_name: jaeger-agent
    image: jaegertracing/jaeger-agent
    hostname: jaeger-agent
    command: [ "--reporter.grpc.host-port=jaeger-collector:14250" ]
    ports:
      - "5775:5775/udp"
      - "6831:6831/udp"
      - "6832:6832/udp"
      - "5778:5778"
    networks:
      - elastic-jaeger
    restart: on-failure
    environment:
      - SPAN_STORAGE_TYPE=elasticsearch
    depends_on:
      - jaeger-collector

  jaeger-query:
    container_name: jaeger-query
    image: jaegertracing/jaeger-query
    environment:
      - SPAN_STORAGE_TYPE=elasticsearch
      - no_proxy=localhost
    ports:
      - "16686:16686"
      - "16687:16687"
    networks:
      - elastic-jaeger
    restart: on-failure
    command: [
      "--es.server-urls=http://elasticsearch:9200",
      "--span-storage.type=elasticsearch",
      "--log-level=debug"
    ]
    depends_on:
      - jaeger-agent

volumes:
  esdata:
    driver: local

networks:
  elastic-jaeger:
    driver: bridge

• Service Performance Monitoring (SPM) --- Jaeger documentation (jaegertracing.io)

可以看到指标了

jaeger的 trace 展示与 grafana，signoz 不一致

出现异常，为修改相关代码，先前可以在Prometheus观测到traefik的指标【已修复】

Otel-collector 的 Pipeline

理解了整个 otel-collector 的 Pipeline 的流程和各个组件的功能

• spanmetrics 是一个 connector
• 它可以作为一个 receiver 【可以接收上游 trace pipeline 的 spanmetrics - 它作为一个 exporter】来开启一个metric 的 pipeline
• 它可以作为一个 exporter 【存储 trace pipeline 的 span 指标】
• spanmetrics 定义为 processer，可以在 trace 的 pipeline 中将 span的指标导出到 Prometheus 里

可以观测到 traefik、venus、profile 上报的 metrics！

明日待办

1. 压测 jaeger
2. 测试替换 jaeger 的数据库为 es