是否可以在kubernetes中应用HPA for Statefulset？

Question

我正在尝试在kubernetes环境中为我的statefulset（针对elasticsearch）设置HPA。我打算使用cpu利用率来扩展statefulset。我已经从https://github.com/stefanprodan/k8s-prom-hpa/tree/master/metrics-server创建了度量服务器。

和我的HPA yaml for statefulset如下：

apiVersion: autoscaling/v2beta1
kind: HorizontalPodAutoscaler
metadata:
  name: dz-es-cluster
spec:
  scaleTargetRef:
    apiVersion: extensions/v1beta1
    kind: StatefulSet
    name: dz-es-cluster
  minReplicas: 2
  maxReplicas: 3
  metrics:
  - type: Resource
    resource:
      name: cpu
      targetAverageUtilization: 80

但是以hpa获得输出如下：

Conditions:
  Type         Status  Reason          Message
  ----         ------  ------          -------
  AbleToScale  False   FailedGetScale  the HPA controller was unable to get the target's current scale: the server could not find the requested resource
Events:
  Type     Reason          Age                From                       Message
  ----     ------          ----               ----                       -------
  Warning  FailedGetScale  1m (x71 over 36m)  horizontal-pod-autoscaler  the server could not find the requested resource

有人请帮帮我..

Answer 1

在kubernetes 1.9中添加了使用HPA自动调度状态集的支持，因此您的版本不支持它。在kubernetes 1.9之后，您可以使用以下命令自动调整状态集：

apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
  name: YOUR_HPA_NAME
spec:
  maxReplicas: 3
  minReplicas: 1
  scaleTargetRef:
    apiVersion: apps/v1
    kind: StatefulSet
    name: YOUR_STATEFUL_SET_NAME
  targetCPUUtilizationPercentage: 80

有关更多信息，请参阅以下链接：

https://github.com/kubernetes/kubernetes/issues/44033