Manage the Lifecycle of Kubernetes Clusters CKA Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   You need to reboot a worker node for a kernel update. The node runs a ReplicaSet with 3 replicas spread across 3 nodes. What is the safest way to reboot the node without causing downtime?

   • ARun 'kubectl drain <node> --ignore-daemonsets --delete-emptydir-data', reboot the node, then run 'kubectl uncordon <node>'
   • BRun 'kubectl cordon <node>', reboot the node, then run 'kubectl uncordon <node>'
   • CReboot the node directly without any preparation
   • DScale down the ReplicaSet to 0, reboot the node, then scale back up
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   You need to scale down your cluster by removing a worker node. The node is running a critical application with a single replica. What is the correct way to ensure the application is rescheduled on another node?

   • AScale down the application to 0, then delete the node
   • BDelete the pod manually, then delete the node
   • CCordon the node, then delete the node
   • DDrain the node, which will evict the pod and reschedule it on another node
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   Which command is used to remove a worker node from a Kubernetes cluster?

   • Akubeadm reset
   • Bkubectl delete node <node-name>
   • Ckubectl drain <node-name>
   • Dkubectl cordon <node-name>
   Reveal Solution Discussion Select an answer first
5. 24application · medium

   You need to remove a worker node from the cluster permanently. The node is currently running workloads. What is the correct sequence of steps?

   • ADelete the node object, then drain the node
   • BJust run 'kubectl delete node <node>'
   • CRun 'kubeadm reset' on the node, then delete the node object
   • DDrain the node, delete the node object, then run 'kubeadm reset' on the node
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   You need to back up the cluster state, including all Kubernetes objects and etcd data. What is the most comprehensive approach?

   • ACopy the /etc/kubernetes directory
   • BOnly export all Kubernetes objects using 'kubectl get all --export'
   • CTake an etcd snapshot and also export all Kubernetes objects using 'kubectl get all --export'
   • DOnly take an etcd snapshot
   Reveal Solution Discussion Select an answer first