Node Affinity & Anti-Affinity with Elastic/Kibana

Affinity is exposed through values options for this package. If you want to schedule your pods to deploy on specific nodes you can do that through the nodeSelector value and as needed the affinity value. Additional info is provided below as well to help in configuring this.

It is good to have a basic knowledge of node affinity and available options to you before customizing in this way - the upstream kubernetes documentation has a good walkthrough of this.

Values for NodeSelector

The nodeSelector value at multiple levels can be set to do basic node selection for deployments. See the below example for an example to schedule pods to only nodes with the label node-type equal to the “pod type”:

kibana:
  nodeSelector:
    node-type: kibana
elasticsearch:
  master:
    nodeSelector:
      node-type: elastic-master
  data:
    nodeSelector:
      node-type: elastic-data

Values for Affinity

The affinity value at multiple levels should be used to specify affinity. The format to include follows what you’d specify at a pod/deployment level. See the example below for scheduling the operator pods only to nodes with the label node-type equal to the “pod type”:

kibana:
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
        - matchExpressions:
          - key: node-type
            operator: In
            values:
            - "kibana"
elasticsearch:
  master:
    affinity:
      nodeAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
          nodeSelectorTerms:
          - matchExpressions:
            - key: node-type
              operator: In
              values:
              - "elastic-master"
  data:
    affinity:
      nodeAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
          nodeSelectorTerms:
          - matchExpressions:
            - key: node-type
              operator: In
              values:
              - "elastic-data"

Values for Anti-Affinity

The affinity value at multiple levels can be set in the same way to schedule pods based on anti-affinity. See the below example to schedule pods to not be present on the nodes that already have pods with the dont-schedule-with: “pod type” label:

kibana:
  affinity:
    podAntiAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        - topologyKey: "kubernetes.io/hostname"
          labelSelector:
            matchLabels:
              dont-schedule-with: kibana
elasticsearch:
  master:
    affinity:
      podAntiAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
          - topologyKey: "kubernetes.io/hostname"
            labelSelector:
              matchLabels:
                dont-schedule-with: elastic-master
  data:
    affinity:
      podAntiAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
          - topologyKey: "kubernetes.io/hostname"
            labelSelector:
              matchLabels:
                dont-schedule-with: elastic-data