Scheduler KCSA Practice Questions (Page 4)

Questions 16–19

1. 16application · medium

   A DevOps team runs a Kubernetes cluster with a mix of CPU-intensive batch jobs and latency-sensitive web services. They need to ensure that batch jobs do not consume resources needed by web services during peak hours. Which scheduling configuration should they implement?

   • ADefine PriorityClasses for the web services and batch jobs, and set the web service pods' priorityClassName to a higher priority.
   • BUse taints and tolerations to prevent batch jobs from being scheduled on nodes that run web services.
   • CSet resource requests and limits on the web service pods and use a ResourceQuota for the batch job namespace.
   • DAdd node affinity rules to the batch job pods to prefer nodes with a specific label.
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A cluster has nodes with different hardware: some have GPUs, others have SSDs. The team wants to ensure that GPU-intensive pods are scheduled only on GPU nodes, and SSD-intensive pods only on SSD nodes. What is the most appropriate scheduling constraint?

   • AUse taints on non-GPU nodes and tolerations on GPU pods.
   • BSet resource requests for GPU and SSD in the pod spec.
   • CUse node affinity with requiredDuringSchedulingIgnoredDuringExecution rules for each pod type.
   • DUse pod affinity to co-locate pods with similar hardware requirements.
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A cluster runs a custom scheduler alongside the default scheduler. The team wants to ensure that only pods with a specific annotation are handled by the custom scheduler, while all others use the default. How should they configure this?

   • ASet the schedulerName field in the pod spec to the custom scheduler's name for the desired pods.
   • BAdd a taint to the custom scheduler and toleration to the pods.
   • CConfigure the custom scheduler to watch all pods and ignore the annotation.
   • DUse a nodeSelector to route pods to nodes managed by the custom scheduler.
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   A company runs a Kubernetes cluster with a custom scheduler that has a vulnerability. They want to minimize the impact if the scheduler is compromised. Which of the following is the most effective mitigation?

   • AGrant the scheduler's ServiceAccount permissions to create pods and nodes.
   • BUse a ServiceAccount with the same permissions as the default ServiceAccount.
   • CRun the scheduler with a read-only root filesystem and drop all capabilities.
   • DDisable the scheduler's API access entirely.
   Reveal Solution Discussion Select an answer first