Kubernetes部署metrics-server监控集群性能

发表于 | 更新于: | 分类于 | 阅读次数:
字数统计: 693

说明

  • Metrics Server
    是实现了 Metrics API 的元件,其目标是取代 Heapster 作位 Pod 与 Node 提供资源的 Usage metrics,该元件会从每个 Kubernetes 节点上的 Kubelet 所公开的 Summary API 中收集 Metrics
  • Horizontal Pod Autoscaler（HPA）控制器用于实现基于CPU使用率进行自动Pod伸缩的功能。
  • HPA控制器基于Master的kube-controller-manager服务启动参数–horizontal-pod-autoscaler-sync-period定义是时长（默认30秒）,周期性监控目标Pod的CPU使用率,并在满足条件时对ReplicationController或deployment中的Pod副本数进行调整,以符合用户定义的平均Pod CPU使用率。
  • 在新版本的kubernetes中 Pod CPU使用率不在来源于heapster,而是来自于metrics-server
  • 官网原话是 The –horizontal-pod-autoscaler-use-rest-clients is true or unset. Setting this to false switches to Heapster-based autoscaling, which is deprecated.

准备工作

kube-apiserver运行参数

metric-server是扩展的apiserver，依赖于kube-aggregator，因此需要在apiserver中开启相关参数

1
2
3
4
5
6
7
--requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.pem
--proxy-client-cert-file=/etc/kubernetes/pki/front-proxy-client.pem
--proxy-client-key-file=/etc/kubernetes/pki/front-proxy-client-key.pem
--requestheader-allowed-names=aggregator
--requestheader-group-headers=X-Remote-Group
--requestheader-extra-headers-prefix=X-Remote-Extra-
--requestheader-username-headers=X-Remote-User

metrics-server版本兼容

Metrics ServerMetrics API group/versionSupported Kubernetes version
0.3.xmetrics.k8s.io/v1beta11.8+
0.2.xmetrics.k8s.io/v1beta11.8+
0.1.xmetrics/v1alpha11.7

metrics-server部署

创建工作目录

1
mkdir -p /home/k8s/

下载项目代码

1
wget -O - https://github.com/kubernetes-incubator/metrics-server/archive/v0.3.1.tar.gz | tar xz --directory=/home/k8s/

修改deployment

文件路径/home/k8s/metrics-server-v0.3.1/deploy/1.8+/metrics-server-deployment.yaml

1
2
3
4
5
6
7
8
9
10
11
containers:
- name: metrics-server
  image: gcrxio/metrics-server-amd64:v0.3.1
  command:
  - /metrics-server
  - --kubelet-preferred-address-types=InternalIP,Hostname,InternalDNS
  - --kubelet-insecure-tls
  - --requestheader-extra-headers-prefix=x-remote-extra-
  - --requestheader-group-headers=x-remote-group
  - --requestheader-username-headers=x-remote-user
  - -v=2

部署metrics-server

1
kubectl apply -f /home/k8s/metrics-server-v0.3.1/deploy/1.8+/

查看pod状态

1
2
3
kubectl -n kube-system get pod -l k8s-app=metrics-server
NAME                                  READY     STATUS    RESTARTS   AGE
pod/metrics-server-86bd9d7667-5hbn6   1/1       Running   0          1m

验证metrics

完成后,等待一段时间(约 30s - 1m)收集 Metrics

  • 请求metrics api的结果
1
kubectl get --raw /apis/metrics.k8s.io/v1beta1

示例输出

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
{
  "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"
      ]
    }
  ]
}
  • 获取节点性能数据
1
kubectl top node

输出示例

1
2
3
4
5
6
NAME     CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%   
k8s-m1   695m         17%    914Mi           11%       
k8s-m2   360m         9%     553Mi           7%        
k8s-m3   492m         12%    533Mi           6%        
k8s-n1   144m         3%     311Mi           3%        
k8s-n2   149m         3%     321Mi           4%
  • 获取Pod性能数据
1
kubectl -n kube-system top pod

输出示例

1
2
3
4
5
6
7
8
9
10
11
12
13
14
NAME                                    CPU(cores)   MEMORY(bytes)   
coredns-56c964478c-g7q8n                14m          14Mi            
coredns-56c964478c-zmt2j                13m          14Mi            
kube-flannel-ds-amd64-8g6f9             6m           15Mi            
kube-flannel-ds-amd64-hslkh             6m           21Mi            
kube-flannel-ds-amd64-ppqkb             8m           14Mi            
kube-flannel-ds-amd64-swj8m             7m           14Mi            
kube-flannel-ds-amd64-zs2sd             8m           17Mi            
kube-proxy-2dq5x                        18m          23Mi            
kube-proxy-9lzmj                        28m          23Mi            
kube-proxy-9xtjc                        18m          23Mi            
kube-proxy-w7mtg                        2m           23Mi            
kube-proxy-zvp8d                        2m           21Mi            
metrics-server-v0.3.1-98bdfb766-s6jf6   8m           22Mi