如何排查Telegraf无法将Openweathermap数据写入InfluxDB故障

云原生阿岚

2026-8-28

TIG栈OpenWeatherMap数据写入失败排查

问题现象

按如下链路搭建TIG栈时，启动出现listen udp :8125: bind: address already in use报错，且inputs.openweathermap采集的数据无法写入InfluxDB：

Openweathermap -> Telegraf -> InfluxDB -> Grafana

已确认基础环境状态：

InfluxDB对应存储桶已创建
Telegraf容器内可正常ping通InfluxDB服务，网络连通性正常

复现操作

进入Telegraf容器手动触发单次采集的命令：

docker exec -it telegraf /bin/sh
telegraf --config /etc/telegraf/telegraf.conf --once

执行后输出日志：

2022-06-22T13:09:27Z W! DeprecationWarning: Option "parse_data_dog_tags" of plugin "inputs.statsd" deprecated since version 1.10.0 and will be removed in 2.0.0: use 'datadog_extensions' instead
2022-06-22T13:09:27Z I! Starting Telegraf 1.23.0
2022-06-22T13:09:27Z I! Loaded inputs: cpu disk diskio kernel mem openweathermap processes statsd swap system
2022-06-22T13:09:27Z I! Loaded aggregators: 
2022-06-22T13:09:27Z I! Loaded processors: 
2022-06-22T13:09:27Z I! Loaded outputs: influxdb_v2 (2x)
2022-06-22T13:09:27Z I! Tags enabled: host=telegraf
2022-06-22T13:09:27Z W! Deprecated inputs: 0 and 1 options
2022-06-22T13:09:27Z E! [agent] Starting input inputs.statsd: listen udp :8125: bind: address already in use
2022-06-22T13:09:29Z I! [inputs.statsd] Stopping the statsd service
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x2d3f999]

当前配置

telegraf.conf

[global_tags]
[agent]
  interval = "10s"
  round_interval = true
  metric_batch_size = 1000
  metric_buffer_limit = 10000
  collection_jitter = "0s"
  flush_interval = "10s"
  flush_jitter = "0s"
  precision = ""
  hostname = ""
  omit_hostname = false

[[outputs.influxdb_v2]]
  urls = ["http://influxdb:8086"]
  token = "$INFLUXDB_INIT_ADMIN_TOKEN"
  organization = "$INFLUXDB_INIT_ORG"
  bucket = "$INFLUXDB_INIT_BUCKET"
  namedrop = ["weather*"]
  
[[outputs.influxdb_v2]]
  urls = ["http://influxdb:8086"]
  token = "$INFLUXDB_OPENWEATHER_TOKEN"
  organization = "$INFLUXDB_INIT_ORG"
  bucket = "$INFLUXDB_OPENWEATHER_BUCKET"
  namepass = ["weather*"]
  
[[inputs.cpu]]
  percpu = true
  totalcpu = true
  collect_cpu_time = false
  report_active = false
[[inputs.disk]]
  ignore_fs = ["tmpfs", "devtmpfs", "devfs", "iso9660", "overlay", "aufs", "squashfs"]
[[inputs.diskio]]
[[inputs.kernel]]
[[inputs.mem]]
[[inputs.processes]]
[[inputs.swap]]
[[inputs.system]]

[[inputs.openweathermap]]
  app_id = "09b8953b9f8f891e15af01a369aeXXX"
  lang = "es"
  fetch = ["weather","forecast"]
  units = "metric"
  interval = "10m"
  
[[inputs.statsd]]
  protocol = "udp"
  max_tcp_connections = 250
  tcp_keep_alive = false
  service_address = ":8125"
  delete_gauges = true
  delete_counters = true
  delete_sets = true
  delete_timings = true
  percentiles = [50.0, 90.0, 99.0, 99.9, 99.95, 100.0]
  metric_separator = "_"
  parse_data_dog_tags = true
  datadog_extensions = false
  datadog_distributions = false
  allowed_pending_messages = 10000
  percentile_limit = 1000

docker-compose.yml

version: "3.9"

volumes:
  influxdb_data: {}
  postgres_data: {}
  grafana_data: {}
  

networks:
  influx:
    name: influx-frontend
  postgres:
    external: false

services:
  influxdb:
    container_name: influxdb
    image: influxdb:${INFLUXDB_VERSION}
    networks:
      - influx
    ports:
      - 8086:8086
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8086/ping"]
      interval: 10s
      timeout: 10s
      retries: 5
      start_period: 40s
    restart: always
    environment:
        - TZ=America/Argentina/Buenos_Aires
    env_file:
      - ./influxdb/influxdb.env
    volumes:
      - influxdb_data:/var/lib/influxdb2
      - ./influxdb/config.yml:/etc/influxdb2/config.yml
  
  telegraf:
    container_name: telegraf
    image: telegraf:${TELEGRAF_VERSION}
    hostname: telegraf
    profiles: ["telegraf"]
    environment:
        - TZ=America/Argentina/Buenos_Aires
    networks:
      - influx
    ports:
      - 8125:8125/udp
    restart: always
    depends_on:
      - influxdb
    env_file:
     - ./telegraf/telegraf.env
    volumes:
      - ./telegraf/telegraf.conf:/etc/telegraf/telegraf.conf:ro

  postgres:
    container_name: postgres
    image: bitnami/postgresql:${POSTGRES_VERSION}
    profiles: ["grafana"]
    networks:
      - postgres
    ports:
      - 5432:5432
    restart: always
    env_file:
      - ./postgres/postgres.env
    volumes:
      - postgres_data:/bitnami/postgresql

  grafana:
    build:
      context: ./grafana
      dockerfile: Dockerfile
      args:
        - GRAFANA_VERSION=${GRAFANA_VERSION:-latest}
    container_name: grafana
    image: grafana/grafana-enterprise:${GRAFANA_VERSION}
    profiles: ["grafana"]
    networks:
      - postgres
      - influx
    ports:
      - 3000:3000
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:3000/api/health"]
      interval: 10s
      timeout: 10s
      retries: 5
      start_period: 40s
    restart: always
    depends_on:
      - postgres
      - influxdb
    env_file:
      - ./grafana/grafana.env
    volumes:
      - grafana_data:/var/lib/grafana
      - ./grafana/provisioning:/etc/grafana/provisioning
      - ./grafana/dashboards:/etc/dashboards

根因定位

两个独立问题共同导致数据写入失败：

端口冲突触发进程崩溃：手动执行单次采集命令时，容器内已经存在entrypoint启动的常驻Telegraf进程，该进程已将UDP 8125端口绑定给statsd输入插件使用。手动启动的第二个Telegraf进程无法绑定同一端口，触发statsd插件启动失败，而Telegraf 1.23版本存在已知bug，statsd启动失败后会触发空指针panic，整个进程直接退出，完全不会执行后续的采集、写入流程。
采集周期不匹配单次执行逻辑：openweathermap插件单独配置了interval = "10m"的采集间隔，--once模式默认仅等待全局flush_interval（即10s）就会退出，不会等待长周期插件完成采集，即便进程不崩溃，单次执行也无法获取到openweathermap的采集数据。

修复步骤

修复statsd配置废弃项：删除parse_data_dog_tags = true配置，改为datadog_extensions = true，消除版本兼容警告。如果不需要使用statsd采集，直接注释掉整个[[inputs.statsd]]配置段，从根源避免端口占用问题。
手动测试采集时，先终止容器内常驻Telegraf进程，避免端口冲突：

# 进入容器后先杀掉常驻进程
pkill telegraf
# 过滤仅加载openweathermap输入和influxdb输出，加debug日志查看采集详情，等待15s给接口留足响应时间
telegraf --config /etc/telegraf/telegraf.conf --once --debug --input-filter openweathermap --output-filter influxdb_v2 --wait 15s

测试阶段可临时将openweathermap的采集间隔改为和全局一致的10s，确认数据正常写入后再改回10m。
检查./telegraf/telegraf.env文件中INFLUXDB_OPENWEATHER_TOKEN的配置值，确认该token对目标存储桶有写入权限。
修复完成后重启Telegraf容器，等待10分钟后即可在InfluxDB中查询到weather开头的气象指标数据。

内容的提问来源于stack exchange，提问作者Adrián Freisinger

如何排查Telegraf无法将Openweathermap数据写入InfluxDB故障

问题现象 #

复现操作 #

当前配置 #

telegraf.conf #

docker-compose.yml #

根因定位 #

修复步骤 #