KubeCost¶

配置¶

前置条件¶

安装 K8S 环境
安装 KubeCost
安装 DataKit
安装 Prometheus Operator

CRD 配置¶

KubeCost 已暴露了指标，只需要让 DataKit 能够发现指标并上报。

新增 kubecost-serverMonitor.yaml

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: kubecost-metrics
  labels:
    app.kubernetes.io/name: cost-analyzer
  namespace: kubecost
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: cost-analyzer
  endpoints:
  - port: metrics

执行

kubectl apply -f kubecost-serverMonitor.yaml

DataKit 配置¶

如已开启，请忽略

开启 DataKit Service Monitor 自动发现

添加 env : ENV_INPUT_CONTAINER_ENABLE_AUTO_DISCOVERY_OF_PROMETHEUS_SERVICE_MONITORS

apiVersion: apps/v1
kind: DaemonSet
metadata:
  labels:
    app: daemonset-datakit
  name: datakit
  namespace: datakit
spec:
  ...
  template:
    ...
    spec:
      ...
      containers:
      - env:
        ...
        - name: ENV_INPUT_CONTAINER_ENABLE_AUTO_DISCOVERY_OF_PROMETHEUS_SERVICE_MONITORS
          value: "true"
        ...

重启 DataKit

指标¶

`kubecost-cost-analyzer`¶

Metric	描述
container_cpu_allocation	container cpu 分配
container_gpu_allocation	container gpu 分配
container_memory_allocation_bytes	container 内存分配
pv_hourly_cost	PersistentVolume 每小时成本
node_total_hourly_cost	节点每小时总成本
node_cpu_hourly_cost	节点 cpu 每小时成本
node_ram_hourly_cost	节点 ram 每小时成本
node_gpu_hourly_cost	节点 gpu 每小时成本