Kubernetes27--弹性伸缩--HPA实践

参考链接https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/

Horizontal Pod Autoscaler automatically scales the number of pods in a replication controller, deployment or replica set based on observed CPU utilization (or, with beta support, on some other, application-provided metrics).

首先构建一个基础镜像用来进行开方计算，会消耗大量的cpu资源

准备index.php

[root@Ac-private-1 hpa]# cat index.php 
<?php
  $x = 0.0001;
  for ($i = 0; $i <= 1000000; $i++) {
    $x += sqrt($x);
  }
  echo "OK!";
?>

准备apache基础镜像dockerfile文件

[root@Ac-private-1 hpa]# cat Dockerfile 
FROM php:5-apache
ADD index.php /var/www/html/index.php
RUN chmod a+rx index.php

构建镜像

启动容器限制cpu=200m 开放80端口

kubectl run php-apache --image=chenwenkai123456/hpa-example --requests=cpu=200m --expose --port=80

创建hpa策略 cpu利用率阈值50% 副本数量在1--10之间

kubectl autoscale deployment php-apache --cpu-percent=50 --min=1 --max=10

kubectl get hpa

kubectl get deployment php-apache

查看当前hpa情况

[root@Ac-private-1 hpa]# kubectl get hpa
NAME         REFERENCE               TARGETS         MINPODS   MAXPODS   REPLICAS   AGE
php-apache   Deployment/php-apache   <unknown>/50%   1         10        0          7s
[root@Ac-private-1 hpa]# kubectl get deployment php-apache
NAME         DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
php-apache   1         1         1            1           5m40s

使用v1 CPU利用率来自动控制

可以直接使用docker hub镜像

docker pull pilchard/hpa-example

使用yaml方式来部署php服务

[root@Ac-private-1 hpa]# cat php-svc.yaml 
apiVersion: extensions/v1beta1  
kind: Deployment  
metadata:  
  name: hpa-ds
spec:  
  replicas: 1
  template:
    metadata:
      labels:
        app: hpa-ds
    spec:
      containers:
      - name: hps-ds
        image: pilchard/hpa-example
        ports:
        - containerPort: 80
        resources:  
          limits:  
            cpu: 0.2  
            memory: 64Mi
---
apiVersion: v1  
kind: Service  
metadata:  
  name: hpa-svc
  labels:
    app: hpa-ds
spec:  
  ports:
  - port: 80
    targetPort: 80
    nodePort: 30964
  type: NodePort
  selector:
    app: hpa-ds

定义HPA对象

[root@Ac-private-1 hpa]# cat hpa.yaml 
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
  name: my-app-hpa
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: extensions/v1beta1
    kind: Deployment
    name: hpa-ds
  minReplicas: 1
  maxReplicas: 10
  targetCPUUtilizationPercentage: 50

访问php service暴露的NodePort

while true; do wget -q -O- http://192.168.1.16:30964; done

观察HPA的策略变化

初始时cpu利用率集群数量1

[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   0%/50%    1         10        1          25s

并发访问cpu以及集群数量的变化

[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   55%/50%   1         10        2          5m49s

[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   32%/50%   1         10        3          6m43s

[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   59%/50%   1         10        4          7m39s

停止访问时

[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   0%/50%    1         10        4          12m
[root@Ac-private-1 ~]# kubectl get hpa
NAME         REFERENCE           TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
my-app-hpa   Deployment/hpa-ds   0%/50%    1         10        1          15m

可知当并发访问增加时，HPA通过不断调整集群数量使得集群cpu利用率维持在50%左右。

使用v2 更多指标以及自定义指标来控制

获取v2版本的yaml文件

kubectl get hpa.v2beta2.autoscaling -o yaml > /tmp/hpa-v2.yaml

v1版本cpu利用率

apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
  name: my-app-hpa
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: extensions/v1beta1
    kind: Deployment
    name: hpa-ds
  minReplicas: 1
  maxReplicas: 10
  targetCPUUtilizationPercentage: 50

v2版本自定义指标

apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
  name: php-apache
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: php-apache
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50

添加metrics标签 type=Resource name=cpu type=Utilization

type可以选择Pods以及Object对象

apiVersion: autoscaling/v2beta1
kind: HorizontalPodAutoscaler
metadata:
  name: php-apache
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: php-apache
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        kind: AverageUtilization
        averageUtilization: 50
  - type: Pods
    pods:
      metric:
        name: packets-per-second
      targetAverageValue: 1k
  - type: Object
    object:
      metric:
        name: requests-per-second
      describedObject:
        apiVersion: extensions/v1beta1
        kind: Ingress
        name: main-route
      target:
        kind: Value
        value: 10k

Then, your HorizontalPodAutoscaler would attempt to ensure that each pod was consuming roughly 50% of its requested CPU, serving 1000 packets per second, and that all pods behind the main-route Ingress were serving a total of 10000 requests per second

查看日志

[root@Ac-private-1 ~]# kubectl describe hpa php-apache
Name:                                                  php-apache
Namespace:                                             default
Labels:                                                <none>
Annotations:                                           <none>
CreationTimestamp:                                     Tue, 18 Dec 2018 15:48:43 +0800
Reference:                                             Deployment/hpa-ds
Metrics:                                               ( current / target )
  resource cpu on pods  (as a percentage of request):  0% (1m) / 50%
Min replicas:                                          1
Max replicas:                                          10
Deployment pods:                                       1 current / 1 desired
Conditions:
  Type            Status  Reason            Message
  ----            ------  ------            -------
  AbleToScale     True    ReadyForNewScale  recommended size matches current size
  ScalingActive   True    ValidMetricFound  the HPA was able to successfully calculate a replica count from cpu resource utilization (percentage of request)
  ScalingLimited  True    TooFewReplicas    the desired replica count is increasing faster than the maximum scale rate
Events:           <none>