No New Node Is Displayed After a Node Is Added to the Kubernetes Cluster

Symptom

After the kubeadm command is executed on a secondary Kubernetes node, a message is displayed indicating that the node is successfully added to the cluster. However, the node cannot be queried on the primary node.

Key Process and Cause Analysis

The versions of the Kubernetes component on the primary and secondary nodes are different.

Conclusion and Solution

1. Before creating a cluster, ensure that the Kubernetes component versions on the primary and secondary nodes are the same. If the versions of the Kubernetes components on the primary and secondary nodes are different, replace them with the same version. Run the following command to check the version:

   docker images | grep k8s.gcr.io

2. Deploy Kubernetes again when the versions of the Kubernetes components on the primary and secondary nodes are the same. For details, see Kubernetes 1.18.20 Deployment Guide (CentOS & openEuler).