Grafana Agent启动失败：apache_http配置字段不存在错误

错误原因 #

你的问题源于两个配置错误：

1. 集成字段名称错误：Grafana Agent中Apache HTTP Server集成的标准配置字段为apache，而非你使用的apache_http。
2. 配置层级错误：所有集成相关的配置（包括Apache、Node Exporter）必须嵌套在integrations顶层配置块内，不能直接放在配置文件的顶层。

修正后的配置 #

将/etc/grafana-agent.yaml修改为以下内容：

integrations:  # 新增顶层integrations块，所有集成配置移入此处
  agent:
    enabled: true
    metric_relabel_configs:
    - source_labels:
      - exported_job
      target_label: job
    - source_labels:
      - exported_instance
      target_label: instance
    - action: labeldrop
      regex: (exported_instance|exported_job)
  apache:  # 将apache_http重命名为apache
    enabled: true
    instance: <yourhostname>
    scrape_uri: http://localhost:9090/server-status?auto
  node_exporter:
    enabled: true
    relabel_configs:
    - replacement: hostname
      target_label: instance
  prometheus_remote_write:
  - basic_auth:
      password: test
      username: test
    url: test
logs:
  configs:
  - clients:
    - basic_auth:
        password: test
        username: test
      url: test
    name: integrations
    positions:
      filename: /tmp/positions.yaml
    scrape_configs:
    - job_name: integrations/node_exporter_journal_scrape
      journal:
        labels:
          instance: hostname
          job: integrations/node_exporter
        max_age: 24h
      relabel_configs:
      - source_labels:
        - __journal__systemd_unit
        target_label: unit
      - source_labels:
        - __journal__boot_id
        target_label: boot_id
      - source_labels:
        - __journal__transport
        target_label: transport
      - source_labels:
        - __journal_priority_keyword
        target_label: level
    - job_name: integrations/apache_error
      pipeline_stages:
      - regex:
          expression: '^\[[^ ]* (?P<timestamp>[^\]]*)\] \[(?:(?P<module>[^:\]]+):)?(?P<level>[^\]]+)\](?: \[pid (?P<pid>[^\]]*)\])?(?: \[client (?P<client>[^\]]*)\])? (?P<message>.*)$'
      - labels:
          level: null
          module: null
      - static_labels:
          logtype: error
      static_configs:
      - labels:
          __path__: /var/log/httpd/error.log
          instance: <yourhostname>
          job: integrations/apache  # 同步修改日志配置中的job字段
        targets:
        - localhost
    - job_name: integrations/apache_access
      pipeline_stages:
      - regex:
          expression: '^(?P<ip>[^ ]*) [^ ]* (?P<user>[^ ]*) \[(?P<timestamp>[^\]]*)\] "(?P<method>\S+)(?: +(?P<path>[^ ]*) +\S*)?" (?P<code>[^ ]*) (?P<size>[^ ]*)(?: "(?P<referer>[^\"]*)" "(?P<agent>.*)")?$'
      - metrics:
          response_http_codes:
            config:
              buckets:
              - 199
              - 299
              - 399
              - 499
              - 599
            description: Apache responses by HTTP codes
            prefix: apache_
            source: code
            type: Histogram
      - labels:
          method: null
      - static_labels:
          logtype: access
      static_configs:
      - labels:
          __path__: /var/log/httpd/access.log
          instance: <yourhostname>
          job: integrations/apache  # 同步修改日志配置中的job字段
        targets:
        - localhost
    target_config:
      sync_period: 10s
metrics:
  configs:
  - name: integrations
    remote_write:
    - basic_auth:
        password: test
        username: test
      url: test
  global:
    scrape_interval: 60s
  wal_directory: /tmp/grafana-agent-wal

关键修改点 #

• 新增顶层integrations配置块，将原有的agent、apache_http、node_exporter、prometheus_remote_write全部移入该块内，匹配Grafana Agent的配置结构规范。
• 将apache_http字段重命名为apache，对齐官方集成的配置字段定义。
• 同步修改日志配置中的job字段值为integrations/apache，确保日志与指标的集成标识统一。

修改完成后，重新启动Grafana Agent即可正常加载配置。

内容的提问来源于stack exchange，提问作者DASTAN ZHOLDAS