Skip to content

Restart K8S Node

1. Restart master node

Just restart each unit one by one

2. Restart worker node

For example: if worker 1 is faulty or slow, you need to restart worker node. (You can restart worker node one by one.)

  • 2.1 Before the server is restarted, you need to set the node state of the worker to Drained

Click ☰ -> wzs-sat-poc-01 -> * Nodes

image-20250502160055345

  • 2.2 Select node and click ┇ -> Cordon

image-20250502160453063

  • 2.3 Login master node and check worker node status
kubectl get nodes

image-20250502160821794

  • 2.4 Dispelling Worker node Pods Can be dispelled through the Web or system naming line, as shown in the following command
kubectl drain zstmesqw1 --ignore-daemonsets --delete-emptydir-data --force
  • 2.5 Restart worker node system
reboot
  • 2.6 After reboot, must to cancel despel status
kubectl uncordon zstmesqw1
kubectl get node

image-20250502161321807