prometheus-adapter的配置解析

adapter的配置以configmap的形式,被挂载到adapter的pod中:

# kubectl get cm adapter-config -n monitoring
NAME             DATA   AGE
adapter-config   1      2d

一.配置文件的结构

整个配置包含2个重要的部分:

  • rules部分:用于custom.metrics.k8s.io,用以自定义指标的HPA伸缩;
  • resourceRule部分:用于metrics.k8s.io,用以cpu、mem指标的HPA伸缩;
rules:
  - seriesQuery: '{__name__=~"^container_.*",container_name!="POD",namespace!="",pod!=""}'
    seriesFilters: []
    resources:
      overrides:
        namespace:
          resource: namespace
        pod: 
          resource: pods
    name:
      matches: ^container_(.*)_seconds_total$
      as: ""
    metricsQuery: sum(rate(<<.Series>>{<<.LabelMatchers>>,container_name!="POD"}[1m])) by (<<.GroupBy>>)
---
resourceRules:
  cpu:
    containerQuery: sum(rate(container_cpu_usage_seconds_total{<<.LabelMatchers>>}[1m])) by (<<.GroupBy>>)
    nodeQuery: sum(rate(container_cpu_usage_seconds_total{<<.LabelMatchers>>, id='/'}[1m])) by (<<.GroupBy>>)
    resources:
      overrides:
        instance:
          resource: nodes
        namespace:
          resource: namespace
        pod:
          resource: pods
    containerLabel: container_name  

二. rules配置的结构

rules:
- seriesQuery: '{__name__=~"^container_.*",container!="POD",namespace!="",pod!="”}'
  seriesFilters: []
  resources:
    overrides:
      namespace: {resource: "namespace"},
      pod: {resource: "pod"},
  name:
    matches: "^container_(.*)_seconds_total$"
  metricsQuery: "sum(rate(<<.Series>>{<<.LabelMatchers>>,container!="POD"}[2m])) by (<<.GroupBy>>)"

每个rule通常由4部分构成:

  • Discovery: 筛选series,包括:

    • seriesQuery
    • seriesFilter
  • Association: 将prom的label与resource类型关联起来

    • resources
  • Naming: 自定义输出的指标名称

    • name
  • Querying: Go template用于定义PromQL表达式

    • metricsQuery

三.rules配置详解

rules定义了指标的转换规则,用作custom.metrics.k8s.io HPA使用。

rules:
  - seriesQuery: '{__name__=~"^container_.*",container_name!="POD",namespace!="",pod!=""}'
    seriesFilters: []
    resources:
      overrides:
        namespace:
          resource: namespace
        pod:
          resource: pods
    name:
      matches: ^container_(.*)_seconds_total$
      as: ""
    metricsQuery: sum(rate(<<.Series>>{<<.LabelMatchers>>,container_name!="POD"}[1m])) by (<<.GroupBy>>)
  • seriesQuery: 用于筛选series;
  • seriesFilters: 用于过滤掉不关心的series;

    • is : 只获取regex匹配到的指标名称;
    • isNot :
  • resource: 将series中的标签与目标资源类型关联起来

    • namespace: 将series中的namespace字段,与资源中的namespace字段关联;
    • pod: 将series中的pod字段,与资源中的pods字段关联;
  • name: 用于给指标重命名,比如以_total结尾的指标,去掉后缀

    • matches: 匹配源指标名;
    • as: 目标指标名称,默认=$1,即第一个分组;
  • metricsQuery: 查询语句

    • <<.Series>>=seriesQuery的结果;
    • LabelMatches,通常是pod/namespace;
    • GroupBy,通常是pod名称;

参考:

1.https://juejin.cn/post/6844903967218991117
2.config介绍:https://github.com/kubernetes-sigs/prometheus-adapter/blob/master/docs/config.md

你可能感兴趣的:(prometheus-adapter的配置解析)