首页 > 其他分享> > |NO.Z.00168|——————————|CloudNative|——|KuberNetes&服务发布.V19|------------------------------------------
|NO.Z.00168|——————————|CloudNative|——|KuberNetes&服务发布.V19|------------------------------------------
作者：互联网
[CloudNative：KuberNetes&服务发布.V19] [Applications.KuberNetes][|DevOps|k8s|服务发布|什么是HPA|自动扩缩容HPA实践|]
一、自动扩缩容HPA实践
### --- 创建deployment，添加内存参数
~~~ 首先需要创建一个deployment
[root@k8s-master01 ~]# kubectl get deploy
NAME READY UP-TO-DATE AVAILABLE AGE
demo-nginx 2/2 2 2 85s### --- 添加触发条件，触发扩容
~~~ 添加它的内存参数
[root@k8s-master01 ~]# kubectl set resources deploy demo-nginx -c demo-nginx --limits=cpu=200m,memory=128Mi --requests=cpu=10m,memory=16Mi
deployment.apps/nginx resource requirements updated
[root@k8s-master01 ~]# kubectl get deploy demo-nginx -oyaml~~~ # 更改CPU参数，把它的Requests更改的小一些，
[root@k8s-master01 ~]# kubectl edit deploy demo-nginx
resources:
limits:
cpu: 1000m
memory: 170Mi
requests:
cpu: 10m
memory: 70Mi
[root@k8s-master01 ~]# kubectl get deploy
NAME READY UP-TO-DATE AVAILABLE AGE
demo-nginx 2/2 2 2 4m14s~~~ # 到100%的时候给他扩容，现在它的状态是0m
[root@k8s-master01 ~]# kubectl top po
NAME CPU(cores) MEMORY(bytes)
busybox 0m 0Mi
demo-nginx-59569f79-2bmnd 2m 2Mi
demo-nginx-59569f79-44z9c 1m 2Mi ### --- 添加触发条件
~~~ --cpu-percent=20:cpu达到CPU的20就扩容 --min=2：最小1个 --max=5：最大5个
[root@k8s-master01 ~]# kubectl autoscale deploy demo-nginx --cpu-percent=20 --min=2 --max=5
horizontalpodautoscaler.autoscaling/nginx autoscaled ### --- 查看它的监控状态，
[root@k8s-master01 ~]# kubectl get hpa
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
demo-nginx Deployment/demo-nginx 0%/20% 2 5 2 82s### --- 当前CPU的占比是0%，当达到20%的时候触发；20%是扩容的指标
~~~ 测试，不下载页面，只是做请求
[root@k8s-master01 ~]# kubectl describe hpa
Name: demo-nginx
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Fri, 23 Apr 2021 15:20:33 +0800
Reference: Deployment/demo-nginx
Metrics: ( current / target )
resource cpu on pods (as a percentage of request): 0% (0) / 20%
Min replicas: 2
Max replicas: 5
Deployment pods: 2 current / 2 desired### --- 找到这个service的地址
[root@k8s-master01 ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
my-service ClusterIP 10.97.126.187 <none> 80/TCP,443/TCP 99s
nginx-svc NodePort 10.101.145.83 <none> 80:31000/TCP,443:32765/TCP 2d20h### --- 触发扩容，当它的CPU达到100%的时候，就会触发扩容，扩容最大的pod数量为5个
~~~ 直接请求去下载它的主页（在不同的节点都启动，增加它的触发效果）
~~~ 请求它的Nginx，但是不下载
[root@k8s-master01 ~]# while true; do wget -q -o- http://10.97.126.187 > /dev/null ;done### --- 查看pod的访问日志，看它是否被请求
[root@k8s-master01 ~]# kubectl get po -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
busybox 1/1 Running 52 3d18h 172.25.244.211 k8s-master01 <none> <none>
demo-nginx-59569f79-2bmnd 1/1 Running 0 123m 172.25.244.218 k8s-master01 <none> <none>
demo-nginx-59569f79-7wzrp 1/1 Running 0 31s 172.18.195.19 k8s-master03 <none> <none>
[root@k8s-master01 ~]# kubectl logs -f demo-nginx-59569f79-2bmnd ### --- 查看po的CPU参数的变化
[root@k8s-master01 ~]# kubectl top po
NAME CPU(cores) MEMORY(bytes)
busybox 0m 0Mi
demo-nginx-59569f79-2bmnd 9m 3Mi
demo-nginx-59569f79-7wzrp 25m 4Mi ### --- 查看它的hpa状态，是否达到100%，若是达到100%就会自动扩容
~~~ CPU的占比已经达到170%了，开始扩容副本，最大扩容数量为5个
[root@k8s-master01 ~]# kubectl get hpa
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
demo-nginx Deployment/demo-nginx 170%/20% 2 5 5 116m### --- 最多扩容pod节点数为5个，我们当时定义的
~~~ 这时候它会把流量分摊到5个节点上面
[root@k8s-master01 ~]# kubectl get po
[root@k8s-master01 ~]# kubectl get po -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
busybox 1/1 Running 53 3d18h 172.25.244.211 k8s-master01 <none> <none>
demo-nginx-59569f79-2bmnd 1/1 Running 0 136m 172.25.244.218 k8s-master01 <none> <none>
demo-nginx-59569f79-4b2nc 1/1 Running 0 11m 172.25.92.77 k8s-master02 <none> <none>
demo-nginx-59569f79-7wzrp 1/1 Running 0 13m 172.18.195.19 k8s-master03 <none> <none>
demo-nginx-59569f79-vtct4 1/1 Running 0 3m32s 172.17.125.17 k8s-node01 <none> <none>
demo-nginx-59569f79-vwf8l 1/1 Running 0 3m32s 172.27.14.212 k8s-node02 <none> <none>### --- 新扩的节点，建议是分摊到不同的宿主机上。
[root@k8s-master01 ~]# kubectl top po### --- 关闭触发条件，等到他的CPU占比恢复到20%的时候，pod节点的数量最少恢复到2个
[root@k8s-master01 ~]# kubectl get po -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
busybox 1/1 Running 53 3d18h 172.25.244.211 k8s-master01 <none> <none>
demo-nginx-59569f79-2bmnd 1/1 Running 0 138m 172.25.244.218 k8s-master01 <none> <none>
demo-nginx-59569f79-4b2nc 1/1 Running 0 13m 172.25.92.77 k8s-master02 <none> <none>
demo-nginx-59569f79-7wzrp 1/1 Running 0 15m 172.18.195.19 k8s-master03 <none> <none>
demo-nginx-59569f79-vtct4 1/1 Running 0 5m33s 172.17.125.17 k8s-node01 <none> <none>
demo-nginx-59569f79-vwf8l 1/1 Running 0 5m33s 172.27.14.212 k8s-node02 <none> <none>### --- 退出访问
~~~ 访问量下降后，pod节点数量变为最少个数
[root@k8s-master01 ~]# while true; do wget -q -o- http://10.97.126.187 > /dev/null ;done
[1]+ Stopped wget -q -o- http://10.97.126.187 > /dev/null
[root@k8s-master01 ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
busybox 1/1 Running 53 3d18h
demo-nginx-59569f79-2bmnd 1/1 Running 0 139m
demo-nginx-59569f79-4b2nc 0/1 Terminating 0 14m
demo-nginx-59569f79-7wzrp 1/1 Running 0 15m
demo-nginx-59569f79-vtct4 0/1 Terminating 0 6m4s
demo-nginx-59569f79-vwf8l 0/1 Terminating 0 6m4s ### --- 查看pod缩容节点数量
[root@k8s-master01 ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
busybox 1/1 Running 53 3d18h
demo-nginx-59569f79-2bmnd 1/1 Running 0 140m
demo-nginx-59569f79-7wzrp 1/1 Running 0 16m ===============================END===============================
Walter Savage Landor:strove with none,for none was worth my strife.Nature I loved and, next to Nature, Art:I warm'd both hands before the fire of life.It sinks, and I am ready to depart ——W.S.Landor
来自为知笔记(Wiz)
标签：缩容,demo,master01,59569f79,------------------------------------------------------ 来源： https://www.cnblogs.com/yanqivip/p/16076561.html