Prerequisites

Check whether alicloud-monitor-controller has been deployed in the kube-system namespace. If not, upgrade the version of the cluster.

Procedure

  1. Log on to the Container Service console.
  2. In the left-side navigation pane under Container Service-Kubernetes, choose Applications > Deployments.
  3. Select the target deployment, click Monitor on the right. You can also click Monitor on the Deployment page of the built-in kubernetes dashboard.

    In this case, you jump to the corresponding Application group details page of CloudMonitor.

  4. Application group supports monitoring in two dimensions: group and instance.
  5. For alarm settings, the index of group level starts with group, and the instance level index starts with pod.

Upgrade cluster version

  1. Log on to the Container Service console.
  2. In the left-side navigation pane under Container Service-Kubernetes, choose Applications > Deployments. Then, in the upper-right corner, click Create by Template.
  3. Select the target cluster, kube-system namespace, and use the following sample template. Then click Create.
    Note Replace REGION and CLUSTER_ID with your actual cluster information, and redeploy heapster yaml template.

    An example of heapster template is as follows. If you have an earlier version of the heapster in the cluster, you can log on to the Kubernetes cluster and run the kubectl apply -f xxx.yaml command to upgrade it. 

    apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: heapster 
  namespace: kube-system 
spec:
  replicas: 1 
  template:
    metadata:
      labels:
        task: monitoring
        k8s-app: heapster
      annotations:
        scheduler.alpha.kubernetes.io/critical-pod: ''
    spec:
      serviceAccount: admin
      containers:
      - name: heapster
        image: registry. ##REGION##.aliyuncs.com/acs/heapster-amd64:v1.5.1.1
        imagePullPolicy: IfNotPresent
        command:
        - /heapster
        - --source=kubernetes:https://kubernetes.default
        - --historical-source=influxdb:http://monitoring-influxdb:8086
        - --sink=influxdb:http://monitoring-influxdb:8086
        - --sink=socket:tcp://monitor.csk. ##REGION##.aliyuncs.com:8093? clusterId=##CLUSTER_ID##&public=true
    The example layout of alicloud-monitor-controller is as follows. Run the kubectl create -f xxx.yaml command to deploy alicloud-monitor-controller. 
    apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: alicloud-monitor-controller
  namespace: kube-system
spec: 
  replicas: 1 
  template:
    metadata:
      labels:
        task: monitoring
        k8s-app: alicloud-monitor-controller
      annotations:
        scheduler.alpha.kubernetes.io/critical-pod: ''
    spec:
      hostNetwork: true
      tolerations:
      - effect: NoSchedule
        operator: Exists
        key: node-role.kubernetes.io/master
      - effect: NoSchedule
        operator: Exists
        key: node.cloudprovider.kubernetes.io/uninitialized
      serviceAccount: admin
      containers:
      - name: alicloud-monitor-controller
        image: registry. ##REGION##.aliyuncs.com/acs/alicloud-monitor-controller:v1.0.0
        imagePullPolicy: IfNotPresent
        command:
        - /alicloud-monitor-controller
        - agent
        - --regionId=##REGION##
        - --clusterId=##CLUSTER_ID##
        - --logtostderr
        - --v=4
  4. Go to the Kubernetes console. In the kube-system namespace, you can see that the two deployments are running, and the upgrade is complete.

If you do not know the REGION information, you can go to the ECS console and select the region where your cluster resides. The last segment of the page URL address is REGION.