Kubernetes 部署metrics-server

Metrics Server是Kubernetes内置自动缩放管道的可扩展，高效的容器资源指标来源。

Metrics Server从Kubelet收集资源指标,并通过Metrics API在Kubernetes apiserver中公开它们,以供Horizo​​ntal Pod Autoscaler使用，也可以通过访问Metrics API kubectl top，用于监测node,pod等的CPU，内存使用情况。

metrics-server  github地址：

https://github.com/kubernetes-sigs/metrics-server
https://github.com/kubernetes-sigs/metrics-server/archive/v0.3.6.tar.gz

软件版本信息：

Kubernetes 　　    v1.18.0
metrics-server 　　0.3.6

1、开启聚合层,查看证书是否存在：

[root@k8s-master ~]# ll /etc/kubernetes/pki/
total 56
-rw-r--r-- 1 root root 1224 Jan 25 15:18 apiserver.crt
-rw-r--r-- 1 root root 1090 Jan 25 15:18 apiserver-etcd-client.crt
-rw------- 1 root root 1675 Jan 25 15:18 apiserver-etcd-client.key
-rw------- 1 root root 1675 Jan 25 15:18 apiserver.key
-rw-r--r-- 1 root root 1099 Jan 25 15:18 apiserver-kubelet-client.crt
-rw------- 1 root root 1679 Jan 25 15:18 apiserver-kubelet-client.key
-rw-r--r-- 1 root root 1025 Jan 25 15:18 ca.crt
-rw------- 1 root root 1675 Jan 25 15:18 ca.key
drwxr-xr-x 2 root root  162 Jan 25 15:18 etcd
-rw------- 1 root root 1675 Jan 25 15:18 sa.key
-rw------- 1 root root  451 Jan 25 15:18 sa.pub
## 以下几个文件
-rw-r--r-- 1 root root 1038 Jan 25 15:18 front-proxy-ca.crt
-rw------- 1 root root 1675 Jan 25 15:18 front-proxy-ca.key
-rw-r--r-- 1 root root 1058 Jan 25 15:18 front-proxy-client.crt
-rw------- 1 root root 1675 Jan 25 15:18 front-proxy-client.key
[root@k8s-master ~]# 

2、修改apiserver文件,新增配置

[root@k8s-master ~]# vim /etc/kubernetes/manifests/kube-apiserver.yaml 
...
    - --requestheader-allowed-names=aggregator
    - --requestheader-extra-headers-prefix=X-Remote-Extra-
    - --requestheader-group-headers=X-Remote-Group
    - --requestheader-username-headers=X-Remote-User
    - --runtime-config=api/all=true
    - --enable-aggregator-routing=true
...

3、重启apiservide服务：

[root@k8s-master ~]#  systemctl restart kubelet.service

4、下载并部署metrics-server

下载地址：
https://github.com/kubernetes-sigs/metrics-server/archive/v0.3.6.tar.gz

5、修改对应文件，添加和修改

[root@k8s-master ~]# cd k8s-addn/metrics-server-0.3.6/deploy/1.8+/
　　　　　　　　　　　 # vim metrics-server-deployment.yaml

    ......
      - name: metrics-server
        ## 修改国内镜像
        image: registry.cn-hangzhou.aliyuncs.com/ljck8s/metrics-server-amd64:v0.3.6
        imagePullPolicy: IfNotPresent
        # 添加command
        command:
        - /metrics-server
        - --metric-resolution=30s
        - --kubelet-insecure-tls
        - --kubelet-preferred-address-types=InternalIP,Hostname,InternalDNS,ExternalDNS,ExternalIP
    ......

6、部署metrics-server

#部署服务
[root@k8s-master 1.8+]# kubectl apply -f .

#查看 metrics-server服务是否正常
#查看apiservice
[root@k8s-master deploy]# kubectl get apiservice |grep metrics-server
v1beta1.metrics.k8s.io                  kube-system/metrics-server   True        31m

#查看 metrics-server pod 
[root@k8s-master deploy]# kubectl get pod -A
NAMESPACE              NAME                                         READY   STATUS    RESTARTS   AGE
kube-system            coredns-7ff77c879f-4d2wt                     1/1     Running   0          14d
kube-system            coredns-7ff77c879f-kjvtg                     1/1     Running   0          14d
kube-system            etcd-k8s-master                              1/1     Running   0          14d
kube-system            kube-apiserver-k8s-master                    1/1     Running   0          35m
kube-system            kube-controller-manager-k8s-master           1/1     Running   1          14d
kube-system            kube-flannel-ds-amd64-49dr9                  1/1     Running   0          14d
kube-system            kube-flannel-ds-amd64-fhd55                  1/1     Running   0          14d
kube-system            kube-flannel-ds-amd64-tx5gg                  1/1     Running   0          14d
kube-system            kube-proxy-92v9b                             1/1     Running   0          14d
kube-system            kube-proxy-dj6gz                             1/1     Running   0          14d
kube-system            kube-proxy-dr7nc                             1/1     Running   0          14d
kube-system            kube-scheduler-k8s-master                    1/1     Running   1          14d
kube-system            metrics-server-6495d87ccb-4zp2l              1/1     Running   0          33m
kubernetes-dashboard   dashboard-metrics-scraper-694557449d-nhxc9   1/1     Running   0          14d
kubernetes-dashboard   kubernetes-dashboard-9774cc786-r2qnx         1/1     Running   0          14d
#查看 metrics-server的svc服务
[root@k8s-master deploy]# kubectl get svc -A
NAMESPACE              NAME                        TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)                  AGE
default                kubernetes                  ClusterIP   10.1.0.1       <none>        443/TCP                  14d
kube-system            kube-dns                    ClusterIP   10.1.0.10      <none>        53/UDP,53/TCP,9153/TCP   14d
kube-system            metrics-server              ClusterIP   10.1.131.73    <none>        443/TCP                  130m
kubernetes-dashboard   dashboard-metrics-scraper   ClusterIP   10.1.150.226   <none>        8000/TCP                 14d
kubernetes-dashboard   kubernetes-dashboard        NodePort    10.1.134.131   <none>        443:30001/TCP            14d

7、服务都已经正常，执行kubectl top命令查看资源情况.

[root@k8s-master deploy]# kubectl top nodes
NAME         CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%   
k8s-master   335m         8%     1105Mi          64%       
k8s-node01   105m         2%     924Mi           25%       
k8s-node02   103m         2%     767Mi           20%       
[root@k8s-master deploy]# 
[root@k8s-master deploy]# kubectl top pods -A
NAMESPACE              NAME                                         CPU(cores)   MEMORY(bytes)   
kube-system            coredns-7ff77c879f-4d2wt                     6m           15Mi            
kube-system            coredns-7ff77c879f-kjvtg                     8m           14Mi            
kube-system            etcd-k8s-master                              40m          83Mi            
kube-system            kube-apiserver-k8s-master                    137m         372Mi           
kube-system            kube-controller-manager-k8s-master           48m          62Mi            
kube-system            kube-flannel-ds-amd64-49dr9                  4m           18Mi            
kube-system            kube-flannel-ds-amd64-fhd55                  3m           16Mi            
kube-system            kube-flannel-ds-amd64-tx5gg                  3m           19Mi            
kube-system            kube-proxy-92v9b                             2m           20Mi            
kube-system            kube-proxy-dj6gz                             2m           32Mi            
kube-system            kube-proxy-dr7nc                             2m           17Mi            
kube-system            kube-scheduler-k8s-master                    8m           26Mi            
kube-system            metrics-server-6495d87ccb-4zp2l              2m           14Mi            

可以看到每个服务的CPU和内存使用信息，metrics-server部署成功.

8、试试metrics-server的API服务，执行命令kubectl proxy --port=8080，用来开代理端口；

再开打一个同样的ssh连接，执行命令curl localhost:8080/apis/metrics.k8s.io/v1beta1/，如下图，可以看到API返回的内容：

[root@k8s-master deploy]# curl localhost:8080/apis/metrics.k8s.io/v1beta1/
{
  "kind": "APIResourceList",
  "apiVersion": "v1",
  "groupVersion": "metrics.k8s.io/v1beta1",
  "resources": [
    {
      "name": "nodes",
      "singularName": "",
      "namespaced": false,
      "kind": "NodeMetrics",
      "verbs": [
        "get",
        "list"
      ]
    },
    {
      "name": "pods",
      "singularName": "",
      "namespaced": true,
      "kind": "PodMetrics",
      "verbs": [
        "get",
        "list"
      ]
    }
  ]
}
[root@k8s-master deploy]# 

查看名为 metrics-server-6495d87ccb-4zp2l 的pod的基本信息，命令是:

[root@k8s-master deploy]# curl localhost:8080/apis/metrics.k8s.io/v1beta1/namespaces/kube-system/pods/metrics-server-6495d87ccb-4zp2l
{
  "kind": "PodMetrics",
  "apiVersion": "metrics.k8s.io/v1beta1",
  "metadata": {
    "name": "metrics-server-6495d87ccb-4zp2l",
    "namespace": "kube-system",
    "selfLink": "/apis/metrics.k8s.io/v1beta1/namespaces/kube-system/pods/metrics-server-6495d87ccb-4zp2l",
    "creationTimestamp": "2021-02-09T05:18:36Z"
  },
  "timestamp": "2021-02-09T05:18:02Z",
  "window": "30s",
  "containers": [
    {
      "name": "metrics-server",
      "usage": {
        "cpu": "724562n",
        "memory": "14584Ki"
      }
    }
  ]
}
[root@k8s-master deploy]# 

部署metrics-server时遇到的问题:

[root@k8s-master 1.8+]# kubectl top nodes
Error from server (ServiceUnavailable): the server is currently unable to handle the request (get nodes.metrics.k8s.io)

解决方法：
修改apiserver文件（在第一步，开启聚合层）

参考文档：

　　https://github.com/kubernetes-sigs/metrics-server
　　https://www.jianshu.com/p/e990aaf3ae28
　　https://blog.csdn.net/boling_cavalry/article/details/105006295
　　https://blog.csdn.net/passnetY/article/details/104659618