发生了什么？

HPA目标状态未知。

❯ k get hpa
NAME             REFERENCE                   TARGETS         MINPODS   MAXPODS   REPLICAS   AGE
{{name}}   Deployment/{{deployment}}   <unknown>/50%   1         10        1          105m

但是，部署和pod资源设置是正常的。

kubectl get -o json deployment {{deployment}} | jq '.spec.template.spec.containers[].resources'
{
  "limits": {
    "cpu": "1",
    "memory": "1500Mi"
  },
  "requests": {
    "cpu": "500m",
    "memory": "1000Mi"
  }
}

metric-server日志找到了。
hpa设置正常。
hpa.yaml

apiVersion: v1
items:
- apiVersion: autoscaling/v2
  kind: HorizontalPodAutoscaler
  metadata:
    labels:
      hpa.test: hpa
    name: {{name}}
    namespace: {{namespace}}
  spec:
    maxReplicas: 10
    metrics:
    - resource:
        name: cpu
        target:
          averageUtilization: 50
          type: Utilization
      type: Resource
    minReplicas: 1
    scaleTargetRef:
      apiVersion: apps/v1
      kind: Deployment
      name: {{deployment name}}

(cronjobs没有hpa.test: hpa标签，但deployment有hpa.test: hpa标签)
HPA生成一些日志

k describe hpa
...
  Type     Reason                   Age                     From                       Message
  ----     ------                   ----                    ----                       -------
  Warning  FailedGetResourceMetric  49m (x13 over 101m)     horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}} of Pod {{cronjob pod name}} 
  Warning  FailedGetResourceMetric  39m (x72 over 110m)     horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}} of Pod {{cronjob pod name}} 
  Warning  FailedGetResourceMetric  34m (x7 over 99m)       horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}}of Pod {{cronjob pod name}} 
  Warning  FailedGetResourceMetric  29m (x2 over 82m)       horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}}of Pod {{cronjob pod name}} 
  Warning  FailedGetResourceMetric  19m (x4 over 63m)       horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}} of Pod {{cronjob pod name}}  
  Warning  FailedGetResourceMetric  14m (x5 over 85m)       horizontal-pod-autoscaler  failed to get cpu utilization: missing request for cpu in container {{cronjob container}} of Pod {{cronjob pod name}}  
  Warning  FailedGetResourceMetric  4m50s (x167 over 106m)  horizontal-pod-autoscaler  (combined from similar

有很多pod的状态是"Completed"

❯ k get po
NAME                                                              READY   STATUS      RESTARTS      AGE
{{deployment name}}                                1/1     Running     0             19m
{{cronjob pod }}   0/1     Completed   4             7h43m
{{cronjob pod }} 0/1     Completed   3             5h43m
{{cronjob pod }}  0/1     Completed   3             163m
{{cronjob pod }}          0/1     Completed   4             6h53m

所以，我认为hpa不仅读取deployment的指标，还读取cronjob的指标。

你期望发生什么？

我期望hpa能表达一个精确的数字。

我们如何尽可能最小精确地重现它？

一起部署deployment和cronjob,并将cronjob置于已完成状态。

我们需要了解其他信息吗？

• 无响应*

Kubernetes版本

v1.28.5

云提供商

AWS EKS

OS版本

# On Linux:
$ cat /etc/os-release
# paste output here
$ uname -a
# paste output here
# On Windows:
C:\> wmic os get Caption, Version, BuildNumber, OSArchitecture
# paste output here

安装工具

容器运行时(CRI)和版本(如适用)

相关插件(CNI,CSI,...)和版本(如适用)