K8S部署Metrics-Server服务

阅读原文：https://www.cnblogs.com/wuchangblog/p/14163960.html

1.下载并解压Metrics-Server

1 2	wget https://github.com/kubernetes-sigs/metrics-server/archive/v0.3.6.tar.gz tar -zxvf v0.3.6.tar.gz

2.修改Metrics-Server配置文件

1 2	# 文件路径 metrics-server-0.3.6/deploy/1.8+/metrics-server-deployment.yaml

编辑 metrics-server-deployment.yaml 以下内容

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 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

--- apiVersion: v1 kind: ServiceAccount metadata:   name: metrics-server   namespace: kube-system --- apiVersion: apps/v1 kind: Deployment metadata:   name: metrics-server   namespace: kube-system   labels:     k8s-app: metrics-server spec:   selector:     matchLabels:       k8s-app: metrics-server   template:     metadata:       name: metrics-server       labels:         k8s-app: metrics-server     spec:       serviceAccountName: metrics-server       volumes:       # mount in tmp so we can safely use from-scratch images and/or read-only containers       - name: tmp-dir         emptyDir: {}       containers:       - name: metrics-server         # 修改image 和 imagePullPolicy         image: mirrorgooglecontainers/metrics-server-amd64:v0.3.6         imagePullPolicy: IfNotPresent         # 新增command配置         command:         - /metrics-server         - --kubelet-insecure-tls         - --kubelet-preferred-address-types=InternalDNS,InternalIP,ExternalDNS,ExternalIP,Hostname         volumeMounts:         - name: tmp-dir           mountPath: /tmp         # 新增resources配置         resources:           limits:             cpu: 300m             memory: 200Mi           requests:             cpu: 200m             memory: 100Mi

3.安装Metrics-Server

1	kubectl apply -f metrics-server-0.3.6/deploy/1.8+/

4.查看node信息

1 2 3 4 5

[root@binghe101 ~]# kubectl top node NAME        CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%   binghe101   141m         7%     1113Mi          65%       binghe102   62m          3%     549Mi           32% binghe103   100m         5%     832Mi           48%

 5、测试HPA

1）、创建部署deployment

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

apiVersion: apps/v1 kind: Deployment metadata:   name: nginx-deployment spec:   replicas: 2   selector:      matchLabels:        app: nginx   template:     metadata:       labels:         app: nginx     spec:       containers:       - name: nginx         image: docker.io/nginx         ports:         - containerPort: 80         resources:           limits:              cpu: 500m           requests:              cpu: 500m

2)、设定自动扩容的条件

1	kubectl autoscale deployment hpa-demo --cpu-percent=10 --min=1 --max=10

 --cpu-percent=10      cpu使用率超过10%

3）、使用ab工具进行压测，随者cpu负载升高，pod数量会增长

1	ab -c 5000 -n 2000000 http://10.244.1.10:80/

10.244.1.10  ---为pod的ip