Scale Che for high availability

Scale Che for high availability by defining a Kubernetes HorizontalPodAutoscaler (HPA) resource for Che operands. The HPA dynamically adjusts the number of replicas based on specified metrics.

Procedure

  1. Create an HPA resource for a deployment, specifying the target metrics and desired replica count.

    apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: scaler
  namespace: eclipse-che
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: <deployment_name>
  ...

    where:

    <deployment_name>

    One of the following deployments:

    • che

    • che-gateway

    • che-dashboard

    • plugin-registry

    • devfile-registry

      For example:

      apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: che-scaler
  namespace: eclipse-che
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: che
  minReplicas: 2
  maxReplicas: 5
  metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 75

      In this example, the HPA targets the che deployment with a minimum of 2 replicas, a maximum of 5 replicas, and scales based on CPU utilization.

Verification

  • Verify that the HPA resource is created and targeting the correct deployment:

    kubectl get hpa -n eclipse-che
Additional resources