Container Service for Kubernetes:Implement horizontal auto scaling based on Alibaba Cloud metrics

The following example shows how to configure HPA by creating a Deployment and a Service that are both named Nginx.

1. Log on to the ACK console. In the left-side navigation pane, click Clusters.

2. On the Clusters page, find the cluster that you want to manage and click its name. In the left-side pane, choose Workloads > Deployments.

3. On the Deployments page, click Create from YAML in the upper-right corner.

4. Set Sample Template to Custom. Use the following YAML template to create a Deployment and a ClusterIP Service. Then, click Create.

   apiVersion: apps/v1 
   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 
           ports:
           - containerPort: 80
   ---
   apiVersion: v1
   kind: Service
   metadata:
     name: nginx
     namespace: default
   spec:
     ports:
       - port: 80
         protocol: TCP
         targetPort: 80
     selector:
       app: nginx
     type: ClusterIP

5. In the left-side navigation pane, choose Network > Ingresses. On the Ingresses page, click Create Ingress in the upper-right corner.

6. In the Create Ingress dialog box, set the parameters and click OK. After the Ingress is created, the Ingresses page appears.

7. In the Name column, find and click the name of the newly created Ingress to view information about the Ingress.

8. Configure HPA.

   1. Click the icon in the upper-left corner to return to the Clusters page.

   2. In the left-side navigation pane of the ACK console, choose Marketplace > Orchestration Templates.

   3. On the Templates page, click Create in the upper-right corner.

   4. In the Create dialog box, configure the template and click Save.

      • Name: Enter a name for the template. In this example, HPA is used.

      • Description: Enter a description for the template. This parameter is optional.

      • Template: Configure an orchestration template that complies with the YAML syntax of Kubernetes. For more information, see the following template.

      apiVersion: autoscaling/v2
      kind: HorizontalPodAutoscaler
      metadata:
        name: ingress-hpa
      spec:
        scaleTargetRef:
          apiVersion: apps/v1
          kind: Deployment
          name: nginx-deployment-basic
        minReplicas: 2
        maxReplicas: 10
        metrics:
          - type: External
            external:
              metric:
                name: sls_ingress_qps
                selector:
                  matchLabels:
                    sls.project: "***" # Specify the Simple Log Service project that you want to use. 
                    sls.logstore: "nginx-ingress"
                    sls.ingress.route: "default-nginx-80"
              target:
                type: AverageValue
                averageValue: 10
          - type: External
            external:
              metric:
                name: sls_ingress_latency_p9999
                selector:
                  matchLabels:
                    # default ingress log project is k8s-log-clusterId
                    sls.project: "***" 
                    # default ingress logstre is nginx-ingress
                    sls.logstore: "nginx-ingress"
                    # namespace-svc-port
                    sls.ingress.route: "default-nginx-80"
                    # sls vpc endpoint, default true
                    # sls.internal.endpoint:true
              target:
                type: Value
                # sls_ingress_latency_p9999 > 10ms
                value: 10

      The following table describes the parameters that are used to configure HPA.

      Parameter	Description
      sls.ingress.route	Set the value in the <namespace>-<svc>-<port> format. Example: default-nginx-80. This parameter is required. Note <namespace> specifies the namespace to which the Ingress belongs. <svc> specifies the name of the Service that you selected when you created the Ingress. <port> specifies the port of the Service.
      sls.logstore	The name of the Logstore in Simple Log Service. This parameter is required. The default value of sls.logstore is nginx-ingress.
      sls.project	The name of the Simple Log Service project. This parameter is required. The default value of sls.project is k8s-log-cluster ID.
      sls.internal.endpoint	Specifies whether to access Simple Log Service over the internal network. Default value: true. If you set the value to true, you can access Simple Log Service over the internal network. If you set the value to false, you can access Simple Log Service over the Internet.

      Note

      In this example, HPA triggers scaling activities based on the sls_ingress_qps and sls_ingress_latency_p9999 metrics. In the target sections, each metric has a different type value:

      • The type value of the sls_ingress_qps metric is set to AverageValue. This indicates that the metric value is the result of dividing the total QPS by the number of pods.

      • The type value of the sls_ingress_latency_p9999 metric is set to Value. This indicates that the latency is not divided by the number of pods.

      The two type values are commonly used in HPA configurations.

   5. On the Templates page, select HPA and click Create Application.

   6. On the HPA page, click Create.

9. After HPA is configured, run the following script to perform a stress test:

   #!/bin/bash
   ##Use Apache Benchmark to send requests to the Service exposed by the Ingress. The test lasts 300 seconds and 10 concurrent requests are sent per second. 
   ab -t 300 -c 10 <The domain name of the Ingress>

10. Check whether HPA works as expected.

    1. In the left-side navigation pane of the ACK console, click Clusters. On the Clusters page, find the cluster that you want to manage and choose More > Open Cloud Shell in the Actions column.

    2. Run the following command to check the status of HPA:

       kubectl get hpa ingress-hpa

       Expected results:

       NAME            REFERENCE                              TARGETS           MINPODS    MAXPODS    REPLICAPS   AGE
       ingress-hpa     Depolyment/nginx-deployment-basic      21/10 (avg)       2          10         10          7m49s

       If the value of REPLICAS is the same as the value of MAXPODS, it indicates that HPA scaled out the application as expected.

Run the following command to query data. In this example, the sls_ingress_qps metric is queried.

kubectl get --raw  /apis/external.metrics.k8s.io/v1beta1/namespaces/*/sls_ingress_qps?labelSelector=sls.project={{SLS_Project}},sls.logstore=nginx-ingress

{{SLS_Project}} is the name of the Simple Log Service project used by the ACK cluster. The default name of the Simple Log Service project used by an ACK cluster is k8s-log-{{ClusterId}}. {{ClusterId}} is the ID of the cluster.

Expected output:

Error from server: {
    "httpCode": 400,
    "errorCode": "ParameterInvalid",
    "errorMessage": "key (slb_pool_name) is not config as key value config,if symbol : is  in your log,please wrap : with quotation mark \"",
    "requestID": "xxxxxxx"
}

If the preceding output is returned, the metric is invalid. No data is returned for the sls_alb_ingress_qps metric because no ALB Ingress is created.