Creating a Node Scaling Policy

Prerequisites

Before using the node scaling function, you must install the autoscaler add-on of v1.13.8 or later in the cluster.

Constraints

• Auto scaling policies apply to node pools. When the number of nodes in a node pool is 0 and the scaling policy is based on CPU or memory usage, node scaling is not triggered.
• Node scale-in will cause PVC/PV data loss for the local PVs associated with the node. These PVCs and PVs cannot be restored or used again. In a node scale-in, the pod that uses the local PV is evicted from the node. A new pod is created and stays in the pending state. This is because the PVC used by the pod has a node label, due to which the pod cannot be scheduled.
• When autoscaler is used, some taints or annotations may affect auto scaling. Therefore, do not use the following taints or annotations in clusters:
  • ignore-taint.cluster-autoscaler.kubernetes.io: The taint works on nodes. Kubernetes-native autoscaler supports protection against abnormal scale outs and periodically evaluates the proportion of available nodes in the cluster. When the proportion of non-ready nodes exceeds 45%, protection will be triggered. In this case, all nodes with the ignore-taint.cluster-autoscaler.kubernetes.io taint in the cluster are filtered out from the autoscaler template and recorded as non-ready nodes, which affects cluster scaling.
  • cluster-autoscaler.kubernetes.io/enable-ds-eviction: The annotation works on pods, which determines whether DaemonSet pods can be evicted by autoscaler. For details, see Well-Known Labels, Annotations and Taints.

Procedure

1. Log in to the CCE console and click the cluster name to access the cluster console.
2. Choose Node Scaling in the navigation pane.
   • If Uninstalled is displayed next to the add-on name, click Install, set add-on parameters as required, and click Install to install the add-on.
   • If Installed is displayed next to the add-on name, the add-on has been installed.

3. Click Create Node Scaling Policy in the upper right corner and set the parameters as follows:
   • Policy Name: name of the policy to be created, which can be customized.
   • Associated Node Pools: Select the node pool to be associated. You can associate multiple node pools to use the same scaling policy.
   • Rules: Click Add Rule. In the dialog box displayed, set the following parameters:

     Rule Name: Enter a rule name.

     Rule Type: You can select Metric-based or Periodic. The differences between the two types are as follows:

     • Metric-based:
       Condition: Select CPU allocation rate or Memory allocation rate and enter a value. The value must be greater than the scale-in percentage configured in the autoscaler add-on.

       

       • Resource allocation (%) = Resources requested by pods in the node pool/Resources allocatable to pods in the node pool
       • If multiple rules meet the conditions, the rules are executed in either of the following modes:

         If rules based on the CPU allocation rate and memory allocation rate are configured and two or more rules meet the scale-out conditions, the rule that will add the most nodes will be executed.

         If a rule based on the CPU allocation rate and a periodic rule are configured and they both meet the scale-out conditions, one of them will be executed randomly. The rule executed first (rule A) changes the node pool to the scaling state. As a result, the other rule (rule B) cannot be executed. After rule A is executed and the node pool status becomes normal, rule B will not be executed.

       • If rules based on the CPU allocation rate and memory allocation rate are configured, the policy detection period varies with the processing logic of each loop of the autoscaler add-on. Scale-out is triggered once the conditions are met, but it is constrained by other factors such as the cool-down interval and node pool status.
     • Periodic:

       Trigger Time: You can select a specific time point every day, every week, every month, or every year.

     Action: Set an action to be performed when the trigger condition is met.

     You can click Add Rule again to add more node scaling policies. You can add a maximum of one CPU usage-based rule and one memory usage-based rule. The total number of rules cannot exceed 10.

4. Click OK.

Constraints on Scale-in

You can set node scale-in policies only when installing the autoscaler add-on.

Node scale-in can be triggered only by the resource allocation rate. When CPU and memory allocation rates in a cluster are lower than the specified thresholds (set when the autoscaler add-on is installed or modified), scale-in is triggered for nodes in the node pool (this function can be disabled).

Example YAML

The following is a YAML example of a node scaling policy:

apiVersion: autoscaling.cce.io/v1alpha1
kind: HorizontalNodeAutoscaler
metadata:
  creationTimestamp: "2020-02-13T12:47:49Z"
  generation: 1
  name: xxxx
  namespace: kube-system
  resourceVersion: "11433270"
  selfLink: /apis/autoscaling.cce.io/v1alpha1/namespaces/kube-system/horizontalnodeautoscalers/xxxx
  uid: c2bd1e1d-60aa-47b5-938c-6bf3fadbe91f
spec:
  disable: false
  rules:
  - action:
      type: ScaleUp
      unit: Node
      value: 1
    cronTrigger:
      schedule: 47 20 * * *
    disable: false
    ruleName: cronrule
    type: Cron
  - action:
      type: ScaleUp
      unit: Node
      value: 2
    disable: false
    metricTrigger:
      metricName: Cpu
      metricOperation: '>'
      metricValue: "40"
      unit: Percent
    ruleName: metricrule
    type: Metric
  targetNodepoolIds:
  - 7d48eca7-3419-11ea-bc29-0255ac1001a8