k8s安装使用metrics-server

fzzjoy

已于 2022-07-06 23:39:36 修改

阅读量3k

点赞数

分类专栏：云原生文章标签： kubernetes docker 容器

于 2022-01-30 21:43:37 首次发布

本文链接：https://blog.csdn.net/u010275850/article/details/122746780

版权

云原生专栏收录该内容

22 篇文章 1 订阅

订阅专栏

目前最新版V1.24.0版本的k8s会导致metircs-server无法正常工作
issue
如果没有正常安装metrics-server的话，kubectl top的命令会报错

$ kubectl top pod 
error: Metrics API not available
或者
$ kubectl top pod
Error from server (ServiceUnavailable): the server is currently unable to handle the request (get pods.metrics.k8s.io)

install metrics-server

kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
如果这时候查看deployment提示出错：

$ kubectl describe pod metrics-server-65b979d49c-bm4vz -n kube-system
······
Events:
  Type     Reason     Age                  From               Message
  ----     ------     ----                 ----               -------
  Normal   Scheduled  5m14s                default-scheduler  Successfully assigned kube-system/metrics-server-65b979d49c-bm4vz to docker-desktop
  Normal   Pulling    5m13s                kubelet            Pulling image "k8s.gcr.io/metrics-server/metrics-server:v0.6.0"
  Normal   Pulled     5m                   kubelet            Successfully pulled image "k8s.gcr.io/metrics-server/metrics-server:v0.6.0" in 12.7280389s
  Normal   Created    5m                   kubelet            Created container metrics-server
  Normal   Started    5m                   kubelet            Started container metrics-server
  Warning  Unhealthy  4s (x28 over 4m34s)  kubelet            Readiness probe failed: HTTP probe failed with statuscode: 500

可以修改yaml文件添加–kubelet-insecure-tls，更改后如下所示：

      containers:
      - args:
        - --cert-dir=/tmp
        - --secure-port=4443
        - --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
        - --kubelet-use-node-status-port
        - --metric-resolution=15s
        - --kubelet-insecure-tls

并保存，修改更新指令如下

$ kubectl edit deploy metrics-server --save-config -n kube-system

安装完之后，便可以通过api访问：

package main

import (
    "context"
    "fmt"
    metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
    "k8s.io/client-go/tools/clientcmd"
    metrics "k8s.io/metrics/pkg/client/clientset/versioned"
    "os"
    "os/user"
    "strings"
)

func main() {
    u, err := user.Current()
    if err != nil {
       fmt.Println(err)
    }
    fmt.Println("Home dir:", u.HomeDir)

    cmdapi, err := clientcmd.LoadFromFile(u.HomeDir + "/.kube/config")
    if err != nil {
       fmt.Println("LoadFromFile err:", err)
    }
    // 当前集群名称
    fmt.Printf("current cluster: [%s]\n", cmdapi.CurrentContext)

    // uses the current context in kubeconfig
    // path-to-kubeconfig -- for example, /root/.kube/config
    // 需要传入url，否则有可能会访问失败
    masterUrl := strings.Replace(os.Getenv("KUBERNETES_PORT"), "tcp", "https", -1)
    config, err := clientcmd.BuildConfigFromFlags(masterUrl, u.HomeDir + "/.kube/config")
    if err != nil {
       fmt.Printf("build config fail: %s", err)
       return
    }

    // build metrics
    mc, err := metrics.NewForConfig(config)
    if err != nil {
       fmt.Println("NewForConfig metrics err is ", err)
    }

    nodeMC, err := mc.MetricsV1beta1().NodeMetricses().Get(context.TODO(), cmdapi.CurrentContext, metav1.GetOptions{})
    if err != nil {
        fmt.Println("node get metrics error: ", err)
    }
    fmt.Println("node[", nodeMC.Name, "] cpu usage: ", nodeMC.Usage.Cpu().String())
    //mc.MetricsV1beta1().NodeMetricses().List(metav1.ListOptions{})
    //mc.MetricsV1beta1().PodMetricses(metav1.NamespaceAll).List(metav1.ListOptions{})
    //mc.MetricsV1beta1().PodMetricses(metav1.NamespaceAll).Get("your pod name", metav1.GetOptions{})
}