四、Kubernetes系列之Kubernetes部署metrics-server
#一、metrics-server簡介
自kubernetes 1.8開始，資源使用指標（如容器 CPU 和內存使用率）通過 Metrics API 在 Kubernetes 中獲取，metrics-server 替代了heapster。Metrics Server 實現了Resource Metrics API，Metrics Server 是集羣範圍資源使用數據的聚合器。
Metrics Server 從每個節點上的 Kubelet 公開的 Summary API 中採集指標信息。
Kubernetes中有些組件依賴資源指標API(metric API)的功能 ，如kubectl top 、hpa。如果沒有資源指標API接口，這些組件無法運行。在之前使用的是Heapster，Heapster廢棄後改用metrics-server。
- 通過Metrics API可以獲取指定node或者pod的當前資源使用情況（而無法獲取歷史數據）
- Metrics API的api路徑：/apis/metrics.k8s.io/
- Metrics API的使用需要在K8S集羣中成功部署metrics server
kubernetes metrics server 參考文檔 https://github.com/kubernetes-incubator/metrics-server
#二、安裝metrics-server
1、下載所需文件
# mkdir ./metrics-server
# cd metrics-server/
# for file in aggregated-metrics-reader.yaml auth-delegator.yaml auth-reader.yaml metrics-apiservice.yaml metrics-server-deployment.yaml metrics-server-service.yaml resource-reader.yaml; do wget https://raw.githubusercontent.com/kubernetes-incubator/metrics-server/master/deploy/1.8%2B/$file;done
2、這裏需要改2個地方，一個是鏡像的問題，一個是服務啓動的問題。
# 在所有Node節點上執行。先把鏡像國通阿里雲的鏡像源下載下來。再改個名字。
docker pull registry.cn-hangzhou.aliyuncs.com/google_containers/metrics-server-amd64:v0.3.3
docker tag registry.cn-hangzhou.aliyuncs.com/google_containers/metrics-server-amd64:v0.3.3 k8s.gcr.io/metrics-server-amd64:v0.3.3
# 修改metrics-server-deployment.yaml，增加一個imagePullPolicy，增加command內容，否則會報錯no metrics known for node。相關問題原因自行百度。
# vim metrics-server-deployment.yaml
30 containers:
31 - name: metrics-server
32 image: k8s.gcr.io/metrics-server-amd64:v0.3.3
33 imagePullPolicy: IfNotPresent
34 command:
35 - /metrics-server
36 - --kubelet-preferred-address-types=InternalIP,Hostname,InternalDNS,ExternalDNS,ExternalIP
37 - --kubelet-insecure-tls
38 volumeMounts:
39 - name: tmp-dir
40 mountPath: /tmp
# kubectl apply -f ./
clusterrole.rbac.authorization.k8s.io/system:aggregated-metrics-reader created
clusterrolebinding.rbac.authorization.k8s.io/metrics-server:system:auth-delegator created
rolebinding.rbac.authorization.k8s.io/metrics-server-auth-reader created
apiservice.apiregistration.k8s.io/v1beta1.metrics.k8s.io created
serviceaccount/metrics-server created
deployment.extensions/metrics-server created
service/metrics-server created
clusterrole.rbac.authorization.k8s.io/system:metrics-server created
clusterrolebinding.rbac.authorization.k8s.io/system:metrics-server created
3、查看pod狀態
metrics-server-55898485b6-pdhnz 1/1 Running 0 93m 10.244.4.2 huoban-k8s-node01
4、查看node資源使用情況（ 一定要等幾分鐘，採集數據需要時間）
kubectl top node
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
huoban-k8s-master01 72m 3% 612Mi 16%
huoban-k8s-master02 93m 4% 713Mi 19%
huoban-k8s-master03 108m 5% 674Mi 18%
huoban-k8s-node01 26m 1% 334Mi 9%
huoban-k8s-node02 26m 1% 339Mi 9%
5、查看pod資源使用情況
# kubectl top pods -n kube-system
NAME CPU(cores) MEMORY(bytes)
coredns-6967fb4995-4qlfb 1m 13Mi
coredns-6967fb4995-frv2p 1m 12Mi
etcd-huoban-k8s-master01 22m 358Mi
etcd-huoban-k8s-master02 25m 364Mi
etcd-huoban-k8s-master03 23m 381Mi
kube-apiserver-huoban-k8s-master01 15m 267Mi
kube-apiserver-huoban-k8s-master02 13m 255Mi
kube-apiserver-huoban-k8s-master03 17m 249Mi
kube-controller-manager-huoban-k8s-master01 7m 69Mi
kube-controller-manager-huoban-k8s-master02 0m 14Mi
kube-controller-manager-huoban-k8s-master03 0m 14Mi
kube-flannel-ds-amd64-6bp76 1m 21Mi
kube-flannel-ds-amd64-nrvvz 1m 15Mi
kube-flannel-ds-amd64-shv4n 1m 16Mi
kube-flannel-ds-amd64-t77n4 1m 15Mi
kube-proxy-8d522 1m 19Mi
kube-proxy-9ng4j 1m 18Mi
kube-proxy-htw7p 1m 20Mi
kube-proxy-n9r48 1m 17Mi
kube-proxy-nsqgh 1m 17Mi
kube-scheduler-huoban-k8s-master01 1m 27Mi
kube-scheduler-huoban-k8s-master02 0m 16Mi
kube-scheduler-huoban-k8s-master03 0m 13Mi
kubernetes-dashboard-86844cc55f-sz4gn 0m 13Mi
metrics-server-d9d75756b-l75wj 1m 17Mi