Understanding Kubernetes Node Selectors

NodeSelector is the most basic recommended node selection restriction. You can use the nodeSelector field in your Pod specification to define the node labels you want the target node to have. Kubernetes only schedules the Pod on nodes with each label you give. NodeSelector is the simplest approach to limit Pods to nodes with certain labels. Affinity and anti-affinity allow you to establish more types of constraints.

What is Kubernetes Node Selector?

A node selector is used to place specific pods on specific nodes, cluster-wide node selectors and new pods on certain nodes throughout the cluster, and project node selectors to set new pods on particular nodes within a project. You can use node selectors and labels to control where a pod is scheduled. With node selectors, OpenShift Container Platform schedules pods on nodes with matching labels.

How Does Kubernetes Node Selector Work?

• Adding the node selector field to your pod specification as a key-value pair allows you to specify the labels you want the target node to have.
• Kubernetes will only schedule pods to nodes that match the labels you specify.
• The node selector is adequate in small clusters but is typically inappropriate for complicated instances.
• For example, you could have an app that has to run in different availability zones. You may also want to keep the API and database distinct, especially if you don't have many replicas.

Implementation of Kubernetes Node Selectors

Step 1: Label the Nodes

Provide each node in your Kubernetes cluster a name first. Assume you have nodes node1 and node2. Node1 should be labeled with disktype=ssd, and Node2 with disktype=hdd.

kubectl label nodes node1 disktype=ssd
kubectl label nodes node2 disktype=hdd

OUTPUT

Step 2: Create a Pod with Node Selector

For the pod, create a YAML file called pod-with-node-selector. yaml and add the following text to it:

apiVersion: v1
kind: Pod
metadata:
  name: with-node-affinity
spec:
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
        - matchExpressions:
          - key: topology.kubernetes.io/zone
            operator: In
            values:
            - antarctica-east1
            - antarctica-west1
      preferredDuringSchedulingIgnoredDuringExecution:
      - weight: 1
        preference:
          matchExpressions:
          - key: another-node-label-key
            operator: In
            values:
            - another-node-label-value
  containers:
  - name: with-node-affinity
    image: registry.k8s.io/pause:2.0

Step 3: Apply the Pod

Next, you have to apply the pod configuration using the kubectl apply command then ensure that the pods are maintained properly:

kubectl apply -f pod-with-node-selector.yaml

OUTPUT

Step 4: Verify the Pod Placement

At last, verify the pod is scheduled on the correct node, and use the below command:

kubectl get pods -o wide

OUTPUT

Best practices of using Node Selector

• Avoid using too many labels: Selectors can be useful, but they should be used sparingly because they can make it difficult to maintain and interpret your selector. Use only the labels you need for your applications, and consider consolidating labels if possible.
• UseNode selector consistent values: Similarly, using them on your selector is critical to ensure that they are simple to grasp and apply.
• Avoid using extremely broad or generic labels: They can be difficult to work with and give little value. Consider using a more particular selector, such as frontend or backend, to describe the sort of application rather than a generic selector such as app.

Difference between Nodeselector and NodeAffinity

Conclusion

In this article, we have learned about Kubernetes Node Selectors. Node selectors allow you to specify the label requirements that a node must fulfill to run a pod. This provides a simple way to direct pods to specified nodes based on their labels.