背景

最近在着手学习HPA的过程，HPA需要metric-service进行采集数据。在实现过程中，遇到了一些问题，进行解决。

问题描述

查看当前HPA的情况。

(base) root@gzk8s-master1:/home/cyye# kubectl get hpa -A
NAMESPACE      NAME                   REFERENCE                         TARGETS         MINPODS   MAXPODS   REPLICAS   AGE
istio-system   istio-ingressgateway   Deployment/istio-ingressgateway   17%/80%         1         5         1          96d
istio-system   istiod                 Deployment/istiod                 1%/80%          1         5         1          96d
ztb-test       test-hpa               Deployment/test-hpa               <unknown>/80%   1         5         2          18h

观察到TARGETS为< unknown >状态，但存在目标80%的阈值信息。初步可以判断是没有获取到当前的数值。
进一步获取详细信息。

事件描述为：

Conditions:
  Type           Status  Reason                   Message
  ----           ------  ------                   -------
  AbleToScale    True    SucceededGetScale        the HPA controller was able to get the target's current scale
  ScalingActive  False   FailedGetResourceMetric  the HPA was unable to compute the replica count: unable to get metrics for resource cpu: no metrics returned from resource metrics API
Events:
  Type     Reason                   Age                     From                       Message
  ----     ------                   ----                    ----                       -------
  Warning  FailedGetResourceMetric  2m57s (x4340 over 18h)  horizontal-pod-autoscaler  missing request for cpu
  

问题分析

检查metric-service服务是否正常运行。

(base) root@gzk8s-master1:/home/cyye#  kubectl get pods -A | grep metrics
kube-system     metrics-server-56b49c5f5b-ntzjk                1/1     Running            93         133d

根据HPA的event里面日志信息，查阅相关资料，是由于HPA的target的Deployment管理的Pod没有设置CPU的request值。

添加的信息为：

          resources:
            requests:
              cpu: 50m
              memory: 50Mi

更新前后示意图如下所示：

解决问题

解决后的效果为：

(base) root@gzk8s-master1:/home/cyye# kubectl get hpa -A
NAMESPACE      NAME                   REFERENCE                         TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
istio-system   istio-ingressgateway   Deployment/istio-ingressgateway   16%/80%   1         5         1          96d
istio-system   istiod                 Deployment/istiod                 1%/80%    1         5         1          96d
ztb-test       test-hpa               Deployment/test-hpa               4%/80%    1         5         1          18h

后记

安装了metrics-server组件后，可以执行下述命令：

(base) root@gzk8s-master1:# kubectl top nodes
NAME            CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%     
gzk8s-master1   13759m       57%    33135Mi         51%         
            
(base) root@gzk8s-master1:# kubectl top pods
NAME                              CPU(cores)   MEMORY(bytes)   
details-v1-5f449bdbb9-rdl2j       5m           90Mi            
productpage-v1-6f9df695b7-xbbsj   24m          135Mi           
ratings-v1-857bb87c57-d7927       5m           81Mi            
reviews-v1-68f9c47f69-vrm56       6m           190Mi           
reviews-v2-5d56c488f5-lhh6r       13m          183Mi           
reviews-v3-869ff44845-pwp4l       6m           168Mi