kubernetes之CronHPA的部署和实战

CronHPA是针对定时水平扩容的一个kubernetes的组件，类似linux的crontab的功能。CronHPA会依据HPA的特点和自身的特点，在使用过程中可以单独使用也可以同时使用，下面先看看cronHPA的安装方法吧：

下载cronHPA

[wolf@wulaoer.org🔥🔥🔥🔥 ~]# git clone https://github.com/AliyunContainerService/kubernetes-cronhpa-controller.git
Cloning into 'kubernetes-cronhpa-controller'...
remote: Enumerating objects: 34191, done.
remote: Counting objects: 100% (442/442), done.
remote: Compressing objects: 100% (350/350), done.
Receiving objects:  39% (13335/34191), 23.14 MiB | 1007.00 KiB/Receiving objects:  39% (13645/34191), 23.14 MiB | 1007.00 KiB/Receiving objects:  40% (13677/34191), 23.14 MiB | 1007.00 KiB/Receiving objects:  41% (14019/34191), 23.61 MiB | 1008.00 KiB/Receiving objects:  42% (14361/34191), 23.61 MiB | 1008.00 KiB/Receiving objects:  43% (14703/34191), 24.05 MiB | 986.00 Kiremote: Total 34191 (delta 137), reused 279 (delta 78), pack-reused 33749
Receiving objects: 100% (34191/34191), 54.10 MiB | 1.04 MiB/s, done.
Resolving deltas: 100% (13799/13799), done.

可能会因为翻墙的问题，不过没关系，可以先下载到本地，然后在传到节点上,在进行解压，后面就都一样了。

安装 RBAC 设置

[wolf@wulaoer.org🔥🔥🔥🔥 hpa]# cd kubernetes-cronhpa-controller-master/
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl apply -f config/crds/autoscaling_v1beta1_cronhorizontalpodautoscaler.yaml
Warning: apiextensions.k8s.io/v1beta1 CustomResourceDefinition is deprecated in v1.16+, unavailable in v1.22+; use apiextensions.k8s.io/v1 CustomResourceDefinition
customresourcedefinition.apiextensions.k8s.io/cronhorizontalpodautoscalers.autoscaling.alibabacloud.com created
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl apply -f config/rbac/rbac_role.yaml
clusterrole.rbac.authorization.k8s.io/kubernetes-cronhpa-controller-role created
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl apply -f config/rbac/rbac_role_binding.yaml
serviceaccount/kubernetes-cronhpa-controller created
clusterrolebinding.rbac.authorization.k8s.io/kubernetes-cronhpa-controller-rolebinding created
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl api-resources  |grep cronhpa
cronhorizontalpodautoscalers      cronhpa      autoscaling.alibabacloud.com/v1beta1   true         CronHorizontalPodAutoscaler

部署kubernetes-cronhpa-controller

[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl apply -f config/deploy/deploy.yaml
deployment.apps/kubernetes-cronhpa-controller created
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl get deploy kubernetes-cronhpa-controller -n kube-system
NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
kubernetes-cronhpa-controller   1/1     1            1           26s

验证kubernetes-cronhpa-controller，并根据文件中的例子说明。

[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl apply -f examples/deployment_cronhpa.yaml
deployment.apps/nginx-deployment-basic created
cronhorizontalpodautoscaler.autoscaling.alibabacloud.com/cronhpa-sample created

[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# cat examples/deployment_cronhpa.yaml
---
apiVersion: apps/v1 # for versions before 1.8.0 use apps/v1beta1
kind: Deployment
metadata:
  name: nginx-deployment-basic
  labels:
    app: nginx
spec:
  replicas: 2
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.7.9 # replace it with your exactly <image_name:tags>
        ports:
        - containerPort: 80
---
apiVersion: autoscaling.alibabacloud.com/v1beta1
kind: CronHorizontalPodAutoscaler
metadata:
  labels:
    controller-tools.k8s.io: "1.0"
  name: cronhpa-sample
spec:
   scaleTargetRef:
      apiVersion: apps/v1beta2
      kind: Deployment
      name: nginx-deployment-basic
   jobs:
   - name: "scale-down"
     schedule: "30 */1 * * * *"
     targetSize: 1
   - name: "scale-up"
     schedule: "01 */1 * * * *"
     targetSize: 3

[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl get deploy nginx-deployment-basic
NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment-basic   0/2     2            0           21s
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl get deploy nginx-deployment-basic
NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment-basic   0/3     3            0           25s
[wolf@wulaoer.org🔥🔥🔥🔥 kubernetes-cronhpa-controller-master]# kubectl get deploy nginx-deployment-basic
NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment-basic   1/1     1            1           57s

以上是针对只有cronhpa的环境，这时会出现一个问题，比如cronhpa定义启动4个副本，但是请求量比较大，就会出现服务重启，如果这个时候加了HPA，那么就会出现定时启动后，如果资源的使用率低于HPA的限制，pod数就会缩回去，而且cronhpa的最小数不能低于HPA的最小副本数。更多可以参考：https://github.com/AliyunContainerService/kubernetes-cronhpa-controller。下一章说一下hpa和cronhpa如何同时使用。