MQL5-Google-Onedrive/pentagi/docker-compose-observability.yml

volumes:
  grafana-data:
    driver: local
  victoriametrics-data:
    driver: local
  clickhouse-data:
    driver: local

networks:
  observability-network:
    driver: bridge
    external: true
    name: observability-network
  langfuse-network:
    driver: bridge
    external: true
    name: langfuse-network
  pentagi-network:
    driver: bridge
    external: true
    name: pentagi-network

services:
  grafana:
    image: grafana/grafana:11.4.0
    restart: unless-stopped
    container_name: grafana
    hostname: grafana
    expose:
      - 3000/tcp
    ports:
      - ${GRAFANA_LISTEN_IP:-127.0.0.1}:${GRAFANA_LISTEN_PORT:-3000}:3000
    environment:
      GF_USERS_ALLOW_SIGN_UP: false
      GF_EXPLORE_ENABLED: true
      GF_ALERTING_ENABLED: true
      GF_UNIFIED_ALERTING_ENABLED: true
      GF_FEATURE_TOGGLES_ENABLE: traceToMetrics,alertingSimplifiedRouting,alertingQueryAndExpressionsStepMode
    volumes:
      - ./observability/grafana/config:/etc/grafana:rw
      - ./observability/grafana/dashboards:/var/lib/grafana/dashboards:rw
      - grafana-data:/var/lib/grafana:rw
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  node-exporter:
    image: prom/node-exporter:v1.8.2
    restart: unless-stopped
    command:
      - --path.procfs=/host/proc
      - --path.sysfs=/host/sys
      - --collector.filesystem.ignored-mount-points
      - ^/(sys|proc|dev|host|etc|rootfs/var/lib/docker/containers|rootfs/var/lib/docker/overlay2|rootfs/run/docker/netns|rootfs/var/lib/docker/aufs)($$|/)
    container_name: node_exporter
    hostname: node-exporter
    expose:
      - 9100/tcp
    volumes:
      - /proc:/host/proc:ro
      - /sys:/host/sys:ro
      - /:/rootfs:ro
    deploy:
      mode: global
    depends_on:
      otel:
        condition: service_started
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  cadvisor:
    image: gcr.io/cadvisor/cadvisor:v0.51.0
    restart: unless-stopped
    command:
      - --store_container_labels=false
      - --docker_only=true
      - --disable_root_cgroup_stats=true
    container_name: cadvisor
    hostname: cadvisor
    expose:
      - 8080/tcp
    volumes:
      - /:/rootfs:ro
      - /var/run:/var/run:rw
      - /sys:/sys:ro
      - /var/lib/docker/:/var/lib/docker:ro
    depends_on:
      otel:
        condition: service_started
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  otel:
    image: otel/opentelemetry-collector-contrib:0.116.1
    restart: unless-stopped
    entrypoint:
      - "/otelcol-contrib"
      - "--config"
      - "/etc/otel/config.yml"
      - "--set"
      - "service.telemetry.logs.level=warn"
    container_name: otel
    hostname: otelcol
    expose:
      - 8148/tcp
      - 4318/tcp
    ports:
      - ${OTEL_GRPC_LISTEN_IP:-127.0.0.1}:${OTEL_GRPC_LISTEN_PORT:-8148}:8148
      - ${OTEL_HTTP_LISTEN_IP:-127.0.0.1}:${OTEL_HTTP_LISTEN_PORT:-4318}:4318
    extra_hosts:
      - host.docker.internal:host-gateway
    volumes:
      - ./observability/otel:/etc/otel:rw
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network
      - langfuse-network
      - pentagi-network

  victoriametrics:
    image: victoriametrics/victoria-metrics:v1.108.1
    restart: unless-stopped
    command:
      - --storageDataPath=/storage
      - --graphiteListenAddr=:2003
      - --opentsdbListenAddr=:4242
      - --httpListenAddr=:8428
      - --influxListenAddr=:8089
      - --selfScrapeInterval=10s
    container_name: victoriametrics
    hostname: victoriametrics
    expose:
      - 8428/tcp
    volumes:
      - victoriametrics-data:/storage:rw
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  clickstore:
    image: clickhouse/clickhouse-server:24
    restart: unless-stopped
    container_name: clickstore
    hostname: clickstore
    expose:
      - 9000/tcp
    environment:
      CLICKHOUSE_DB: jaeger
      CLICKHOUSE_USER: clickhouse
      CLICKHOUSE_PASSWORD: clickhouse
    ulimits:
      nofile:
        hard: 262144
        soft: 262144
    volumes:
      - ./observability/clickhouse/prometheus.xml:/etc/clickhouse-server/config.d/prometheus.xml:ro
      - clickhouse-data:/var/lib/clickhouse:rw
    healthcheck:
      test: wget --no-verbose --tries=1 --spider http://localhost:8123/ping || exit 1
      interval: 5s
      timeout: 5s
      retries: 10
      start_period: 1s
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  loki:
    image: grafana/loki:3.3.2
    restart: unless-stopped
    command: -config.file=/etc/loki/config.yml
    container_name: loki
    hostname: loki
    expose:
      - 3100/tcp
    volumes:
      - ./observability/loki/config.yml:/etc/loki/config.yml:ro
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network

  jaeger:
    image: jaegertracing/all-in-one:1.56.0
    restart: unless-stopped
    entrypoint: >
      /bin/sh -c '
      if [ "$$(uname -m)" = "x86_64" ]; then
        ARCH="amd64"
      elif [ "$$(uname -m)" = "aarch64" ]; then
        ARCH="arm64"
      else
        echo "Unsupported architecture"
        sleep 30
        exit 1
      fi &&
      /go/bin/all-in-one-linux
      --grpc-storage-plugin.binary=/etc/jaeger/bin/jaeger-clickhouse-linux-$$ARCH
      --grpc-storage-plugin.configuration-file=/etc/jaeger/plugin-config.yml
      --grpc-storage-plugin.log-level=info'
    container_name: jaeger
    hostname: jaeger
    expose:
      - 16686/tcp
      - 14250/tcp
      - 14268/tcp
      - 5778/tcp
      - 5775/udp
      - 6831/udp
      - 6832/udp
    ulimits:
      nofile:
        hard: 65000
        soft: 65000
      nproc: 65535
    volumes:
      - ./observability/jaeger:/etc/jaeger:rw
    environment:
      SPAN_STORAGE_TYPE: grpc-plugin
    depends_on:
      clickstore:
        condition: service_healthy
    logging:
      options:
        max-size: 50m
        max-file: "7"
    networks:
      - observability-network