kube_state_metrics

• node

• #节点 数量
  sum(kube_node_info)
  #不可调度的节点数量
  sum(kube_node_spec_unschedulable)
  # 集群cpu 数量
  sum(kube_node_status_capacity{resource="cpu"})
  # 集群内存 数量
  sum(kube_node_status_capacity{resource="memory"})
   
  # 存在 磁盘压力的节点
  kube_node_status_condition{condition="DiskPressure"}
  # 存在 磁盘压力的节点
  kube_node_status_condition{condition="MemoryPressure"}
  # 存在 pid压力的节点
  kube_node_status_condition{condition="PIDPressure"}
  # 存在 网络不可达节点
  kube_node_status_condition{condition="NetworkUnavailable"}

• deployment

• 各个deployment的副本数量
  kube_deployment_status_replicas
   
  各个deployment不可用的副本数量
  kube_deployment_status_replicas_unavailable

• ds
• sts
• job
• cronjob
• pod

• kube_pod_status_phase{phase="Failed"}
  kube_pod_status_phase{phase="Pending"}
  kube_pod_status_phase{phase="Running"}
  kube_pod_status_phase{phase="Succeeded"}
  kube_pod_status_phase{phase="Unknown"}
   
   

• container

• #
  kube_pod_container_status_running
  kube_pod_container_status_waiting
  kube_pod_container_status_ready
  kube_pod_container_status_terminated
  kube_pod_container_status_terminated_reason
  # 30分钟内重启过的pod
  changes(kube_pod_container_status_restarts_total[30m])

• kube_pod_container_resource_requests{resource="cpu"}
  kube_pod_container_resource_requests{resource="memory"}

• pvc

• kube_persistentvolumeclaim_status_phase{phase="Bound"}
  kube_persistentvolumeclaim_status_phase{phase="Lost"}
  kube_persistentvolumeclaim_status_phase{phase="Pending"}

• kube_persistentvolume_status_phase{phase="Available"}
  kube_persistentvolume_status_phase{phase="Bound"}
  kube_persistentvolume_status_phase{phase="Failed"}
  kube_persistentvolume_status_phase{phase="Pending"}
  kube_persistentvolume_status_phase{phase="Released"}

• svc
• configmap
• secert
• poddisruptionbudget