kubernetes(k8s)安装metrics-server实现资源使用情况监控方式详解

1. Metrics Server 与 kubenetes版本

Metrics Server    Metrics API group/version    Supported Kubernetes version

0.6x    metrics.k8s.io/v1beta1    *1.19+

0.5x    metrics.k8s.io/v1beta1    *1.8+

0.4x    metrics.k8s.io/v1beta1    *1.8+

0.3x    metrics.k8s.io/v1beta1    1.8-1.21


2. Metrics Server 下载方式

github：https://github.com/kubernetes-sigs/metrics-server


3. k8s集群安装部署metrics

本次安装的是metrics0.5.0版本

下载地址：https://github.com/kubernetes-sigs/metrics-server/releases

docker镜像地址：docker pull cnskylee/metrics-server:v0.5.0

4. 创建components-v0.5.0.yaml文件，并将下面的脚本copy到文件中

1. apiVersion: v1
2. kind: ServiceAccount
3. metadata:
4.   labels:
5.     k8s-APP: metrics-server
6.   name: metrics-server
7.   namespace: kube-system
8. —
9. apiVersion: rbac.authorization.k8s.io/v1
10. kind: ClusterRole
11. metadata:
12.   labels:
13.     k8s-app: metrics-server
14.     rbac.authorization.k8s.io/aggregate-to-admin: “true”
15.     rbac.authorization.k8s.io/aggregate-to-edit: “true”
16.     rbac.authorization.k8s.io/aggregate-to-view: “true”
17.   name: system:aggregated-metrics-reader
18. rules:
19. – apiGroups:
20.   – metrics.k8s.io
21.   resources:
22.   – pods
23.   – nodes
24.   verbs:
25.   – get
26.   – list
27.   – watch
28. —
29. apiVersion: rbac.authorization.k8s.io/v1
30. kind: ClusterRole
31. metadata:
32.   labels:
33.     k8s-app: metrics-server
34.   name: system:metrics-server
35. rules:
36. – apiGroups:
37.   – “”
38.   resources:
39.   – pods
40.   – nodes
41.   – nodes/stats
42.   – namespaces
43.   – configmaps
44.   verbs:
45.   – get
46.   – list
47.   – watch
48. —
49. apiVersion: rbac.authorization.k8s.io/v1
50. kind: RoleBinding
51. metadata:
52.   labels:
53.     k8s-app: metrics-server
54.   name: metrics-server-auth-reader
55.   namespace: kube-system
56. roleRef:
57.   apiGroup: rbac.authorization.k8s.io
58.   kind: Role
59.   name: extension-apiserver-authentication-reader
60. subjects:
61. – kind: ServiceAccount
62.   name: metrics-server
63.   namespace: kube-system
64. —
65. apiVersion: rbac.authorization.k8s.io/v1
66. kind: ClusterRoleBinding
67. metadata:
68.   labels:
69.     k8s-app: metrics-server
70.   name: metrics-server:system:auth-delegator
71. roleRef:
72.   apiGroup: rbac.authorization.k8s.io
73.   kind: ClusterRole
74.   name: system:auth-delegator
75. subjects:
76. – kind: ServiceAccount
77.   name: metrics-server
78.   namespace: kube-system
79. —
80. apiVersion: rbac.authorization.k8s.io/v1
81. kind: ClusterRoleBinding
82. metadata:
83.   labels:
84.     k8s-app: metrics-server
85.   name: system:metrics-server
86. roleRef:
87.   apiGroup: rbac.authorization.k8s.io
88.   kind: ClusterRole
89.   name: system:metrics-server
90. subjects:
91. – kind: ServiceAccount
92.   name: metrics-server
93.   namespace: kube-system
94. —
95. apiVersion: v1
96. kind: Service
97. metadata:
98.   labels:
99.     k8s-app: metrics-server
100.   name: metrics-server
101.   namespace: kube-system
102. spec:
103.   ports:
104.   – name: https
105.     port: 443
106.     protocol: TCP
107.     targetPort: https
108.   selector:
109.     k8s-app: metrics-server
110. —
111. apiVersion: apps/v1
112. kind: Deployment
113. metadata:
114.   labels:
115.     k8s-app: metrics-server
116.   name: metrics-server
117.   namespace: kube-system
118. spec:
119.   selector:
120.     matchLabels:
121.       k8s-app: metrics-server
122.   strategy:
123.     rollingUpdate:
124.       maxUnavailable: 0
125.   template:
126.     metadata:
127.       labels:
128.         k8s-app: metrics-server
129.     spec:
130.       containers:
131.       – args:
132.         – –cert-dir=/tmp
133.         – –secure-port=4443
134.         – –kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
135.         – –kubelet-use-node-status-port
136.         – –metric-resolution=15s
137.         – –kubelet-insecure-tls
138.         image: registry.cn-shenzhen.aliyuncs.com/zengfengjin/metrics-server:v0.5.0
139.         imagePullPolicy: IfNotPresent
140.         livenessProbe:
141.           failureThreshold: 3
142.           httpGet:
143.             path: /livez
144.             port: https
145.             scheme: HTTPS
146.           periodSeconds: 10
147.         name: metrics-server
148.         ports:
149.         – containerPort: 4443
150.           name: https
151.           protocol: TCP
152.         readinessProbe:
153.           failureThreshold: 3
154.           httpGet:
155.             path: /readyz
156.             port: https
157.             scheme: HTTPS
158.           initialDelaySeconds: 20
159.           periodSeconds: 10
160.         resources:
161.           requests:
162.             cpu: 100m
163.             memory: 200Mi
164.         securityContext:
165.           readOnlyRootFilesystem: true
166.           runAsNonRoot: true
167.           runAsUser: 1000
168.         volumeMounts:
169.         – mountPath: /tmp
170.           name: tmp-dir
171.       nodeSelector:
172.         kubernetes.io/os: linux
173.       priorityClassName: system-cluster-critical
174.       serviceAccountName: metrics-server
175.       volumes:
176.       – emptyDir: {}
177.         name: tmp-dir
178. —
179. apiVersion: apiregistration.k8s.io/v1
180. kind: APIService
181. metadata:
182.   labels:
183.     k8s-app: metrics-server
184.   name: v1beta1.metrics.k8s.io
185. spec:
186.   group: metrics.k8s.io
187.   groupPriorityMinimum: 100
188.   insecureSkipTLSVerify: true
189.   service:
190.     name: metrics-server
191.     namespace: kube-system
192.   version: v1beta1
193.   versionPriority: 100

复制代码 需要注意的是端口和镜像地址。


5. 执行部署

1. kubectl apply -f ./components-v0.5.0.yaml

复制代码
4.查看metrics-server的pod运行状态

1. kubectl get pods -n kube-system| egrep ‘NAME|metrics-server’

复制代码
5. 查看运行日志

1. # kubectl logs metrics-server-56874cd58b-b2gj9 -n kube-system                              
2. I0418 09:49:44.461828       1 serving.go:341] Generated self-signed cert (/tmp/apiserver.crt, /tmp/apiserver.key)
3. I0418 09:49:45.252957       1 requestheader_controller.go:169] Starting RequestHeaderAuthRequestController
4. I0418 09:49:45.252972       1 shared_informer.go:240] Waiting for caches to sync for RequestHeaderAuthRequestController
5. I0418 09:49:45.252994       1 configmap_cafile_content.go:202] Starting client-ca::kube-system::extension-apiserver-authentication::client-ca-file
6. I0418 09:49:45.253002       1 shared_informer.go:240] Waiting for caches to sync for client-ca::kube-system::extension-apiserver-authentication::client-ca-file
7. I0418 09:49:45.253142       1 configmap_cafile_content.go:202] Starting client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file
8. I0418 09:49:45.253153       1 shared_informer.go:240] Waiting for caches to sync for client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file
9. I0418 09:49:45.254091       1 secure_serving.go:197] Serving securely on [::]:4443
10. I0418 09:49:45.254139       1 tlsconfig.go:240] Starting DynamicServingCertificateController
11. I0418 09:49:45.254342       1 dynamic_serving_content.go:130] Starting serving-cert::/tmp/apiserver.crt::/tmp/apiserver.key
12. I0418 09:49:45.353087       1 shared_informer.go:247] Caches are synced for RequestHeaderAuthRequestController
13. I0418 09:49:45.353087       1 shared_informer.go:247] Caches are synced for client-ca::kube-system::extension-apiserver-authentication::client-ca-file
14. I0418 09:49:45.354631       1 shared_informer.go:247] Caches are synced for client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file

复制代码
6.测试kubectl top命令的使用

1. # kubectl top pods -n kube-system
2. NAME                                 CPU(cores)   MEMORY(bytes)   
3. coredns-58cc8c89f4-jdfc7             4m           12Mi            
4. coredns-58cc8c89f4-z8t26             5m           13Mi            
5. etcd-k8s-master                      9m           281Mi           
6. kube-apiserver-k8s-master            22m          296Mi

复制代码 查看节点资源：

1. # kubectl top nodes
2. NAME         CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%   
3. k8s-master   96m          2%     9330Mi          59%      
4. k8s-node83   66m          1%     5829Mi          37%      
5. k8s-node85   116m         2%     8555Mi          54%

复制代码 查看默认空间pod资源：

1. # kubectl top pods
2. NAME                     CPU(cores)   MEMORY(bytes)   
3. nginx-6867cdf567-rprv9   0m           1Mi

复制代码 到此这篇关于kubernetes(k8s)安装metrics-server实现资源使用情况监控的文章就介绍到这了,更多相关kubernetes安装metrics-server资源监控内容请搜索共生网络以前的文章或继续浏览下面的相关文章希望大家以后多多支持共生网络！