Deploying and Managing Kubernetes Pods Using Node Selectors

This document explains how to deploy a pod in a Kubernetes cluster and assign it to a specific node using a nodeSelector. Follow the step-by-step instructions to understand the workflow.

Step 1: Verify Initial Cluster Setup

1. Check for existing pods:

    controlplane $ k get pods
   

   Output:

    No resources found in default namespace.
   

   This confirms that no pods are currently running in the default namespace.

2. Check node details:

    controlplane $ k get nodes -o wide
   

   Example output:

    NAME           STATUS   ROLES           AGE    VERSION   INTERNAL-IP   EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
    controlplane   Ready    control-plane   7d3h   v1.31.0   172.30.1.2    <none>        Ubuntu 20.04.5 LTS   5.4.0-131-generic   containerd://1.7.13
    node01         Ready    <none>          7d2h   v1.31.0   172.30.2.2    <none>        Ubuntu 20.04.5 LTS   5.4.0-131-generic   containerd://1.7.13
   

Step 2: Deploy a Pod

1. Create a Pod definition file (pod.yaml):

    controlplane $ cat pod.yaml
   

   Content of pod.yaml:

    apiVersion: v1
    kind: Pod
    metadata:
      name: web-pod
    spec:
      containers:
      - image: nginx
        name: web-pod
        ports:
         - containerPort: 80
   

2. Apply the Pod configuration:

    controlplane $ kubectl apply -f pod.yaml
   

   Output:

    pod/web-pod created
   

3. Verify Pod status:

    controlplane $ k get pods
   

   Example output:

    NAME      READY   STATUS    RESTARTS   AGE
    web-pod   1/1     Running   0          20s
   

Step 3: Label the Node

1. Add a label to the controlplane node:

    controlplane $ kubectl label node controlplane node=master
   

   Output:

    node/controlplane labeled
   

2. Describe the node to confirm the label:

    controlplane $ k describe node controlplane
   

   Example output:

    Name:               controlplane
    Roles:              control-plane
    Labels:             beta.kubernetes.io/arch=amd64
                        beta.kubernetes.io/os=linux
                        kubernetes.io/arch=amd64
                        kubernetes.io/hostname=controlplane
                        kubernetes.io/os=linux
                        node=master
   

Step 4: Assign Pod to Specific Node

1. Update the Pod definition to include nodeSelector:

    controlplane $ cat pod.yaml
   

   Updated content of pod.yaml:

    apiVersion: v1
    kind: Pod
    metadata:
      name: web-pod
    spec:
      containers:
      - image: nginx
        name: web-pod
        ports:
         - containerPort: 80
      nodeSelector:
        node: "master"
   

2. Delete the existing Pod:

    controlplane $ k delete pod web-pod
   

   Output:

    pod "web-pod" deleted
   

3. Reapply the updated Pod configuration:

    controlplane $ k apply -f pod.yaml
   

   Output:

    pod/web-pod created
   

4. Verify Pod details with node assignment:

    controlplane $ k get pod -o wide
   

   Example output:

    NAME      READY   STATUS    RESTARTS   AGE   IP            NODE           NOMINATED NODE   READINESS GATES
    web-pod   1/1     Running   0          78s   192.168.0.4   controlplane   <none>           <none>
   

   This confirms that the Pod is running on the controlplane node.

Summary

By using a nodeSelector in the Pod configuration, you can assign a Pod to a specific node in the Kubernetes cluster. This workflow ensures that workloads are deployed to the desired node based on labels.