McAfee Secure

CNCF CKA Bundle

Exam Code: CKA

Exam Name Certified Kubernetes Administrator

Certification Provider: CNCF

Corresponding Certification: CKA

CNCF CKA Bundle $44.99
CKA Sample 1 CKA Sample 2 CKA Sample 3 CKA Sample 4 CKA Sample 5 CKA Sample 6 CKA Sample 7 CKA Sample 8 CKA Sample 9 CKA Sample 10

100% CNCF CKA Guaranteed Success With Testking Training Materials

Get Real CKA Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

  • Questions & Answers

    CKA Questions & Answers

    58 Questions & Answers

    The ultimate exam preparation tool, Questions & Answers mirrors the actual CKA exam while allowing you to hone and measure every aspect of your test-taking ability.

  • CKA Video Course

    CKA Training Course

    138 Video Lectures

    CKA Video Course is developed by CNCF Professionals to help you pass the CKA exam.

    Description

    This course will improve your knowledge and skills required to pass Certified Kubernetes Administrator exam.
  • Study Guide

    CKA Study Guide

    268 PDF Pages

    Developed by industry experts, this 268-page guide spells out in painstaking detail all of the information you need to ace CKA exam.

Certified Kubernetes Administrator (CKA) Practice Tests

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Total Cost: $194.97
Bundle Price: $149.98

Purchase Individually

  • Questions & Answers

    Questions & Answers

    58 Questions

    $124.99
  • CKA Video Course

    Training Course

    138 Video Lectures

    $39.99
  • Study Guide

    Study Guide

    268 PDF Pages

    $29.99

CNCF CKA Practice Test 112

Question 1110

At Horizon observability, in a disposable simulated environment during a monitored release window: Node worker-b reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which Kubernetes behavior best explains the required change or diagnosis?

  1. The container runtime can terminate a container when it exceeds its memory cgroup limit; increasing CPU requests does not repair a memory limit failure.
  2. HPA CPU utilization is computed from reported metrics relative to requests; missing metrics or requests can prevent scaling decisions.
  3. Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.
  4. CrashLoopBackOff is a restart backoff symptom, not a root cause; the preceding container logs often preserve the terminating process error.

Explanation

Correct Answer

C:

Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. This explains the stated condition. The corresponding action is: Run kubectl describe node worker-b and kubectl top node worker-b where metrics are available before altering Pod requests. The direct acceptance evidence is: Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning.

Incorrect Answers

A:

The container runtime can terminate a container when it exceeds its memory cgroup limit; increasing CPU requests does not repair a memory limit failure. That explanation applies to a different technical failure. Here the controlling principle is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

B:

The offered statement is true for this other mechanism: HPA CPU utilization is computed from reported metrics relative to requests; missing metrics or requests can prevent scaling decisions. It does not explain the facts described here, which follow from: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

D:

CrashLoopBackOff is a restart backoff symptom, not a root cause; the preceding container logs often preserve the terminating process error. That explanation applies to a different technical failure. Here the controlling principle is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

Question 1111

At Horizon observability, in a disposable simulated environment during a monitored release window: Node worker-b reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which observation best verifies the requested outcome rather than an unrelated state?

  1. Verify HPA reports a numeric utilization and responds to a controlled load change within configured bounds.
  2. Run kubectl get node worker-a and confirm Ready after restoring the actual failed component.
  3. Confirm the scheduler events clear and the Pod is assigned after capacity or request adjustments.
  4. Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning.

Explanation

Correct Answer

D:

Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning. This validates the affected object or control path, rather than only showing a command completed. The underlying reason is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The intended action is: Run kubectl describe node worker-b and kubectl top node worker-b where metrics are available before altering Pod requests.

Incorrect Answers

A:

That test belongs to another control with this basis: HPA CPU utilization is computed from reported metrics relative to requests; missing metrics or requests can prevent scaling decisions. In the current case, the governing reason is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The direct verification is “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

B:

This observation would test a different technical dependency: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. The current issue is governed by: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Its decisive check is “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

C:

The alternative verifies a different condition: The scheduler places Pods by requested resources and constraints, not the current shell's observed free CPU alone. The current case instead depends on: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Confirm that “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

Question 1112

At Horizon observability, in a disposable simulated environment during a monitored release window: Node worker-b reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which operational control most directly prevents this failure or configuration drift?

  1. Baseline memory requests and track node allocatable resources and eviction signals.
  2. Govern scheduler labels and include eligible-node checks before promoting hard node selectors.
  3. Maintain image garbage-collection, log rotation and filesystem capacity alerts.
  4. Maintain an inventory of dedicated-node taints and explicitly reviewed workload tolerations.

Explanation

Correct Answer

A:

Baseline memory requests and track node allocatable resources and eviction signals. This reduces recurrence of the specific operational failure. It reinforces the action: Run kubectl describe node worker-b and kubectl top node worker-b where metrics are available before altering Pod requests. The relevant mechanism is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

Incorrect Answers

B:

The suggested review tracks a different technical risk: A hard nodeSelector must match a Node's actual labels; tolerations or Service routing cannot bypass the unmatched selector. Here the sustained control must address: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Choose “Baseline memory requests and track node allocatable resources and eviction signals”.

C:

That safeguard is relevant to: DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads. The current failure instead follows from: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Its prevention measure is “Baseline memory requests and track node allocatable resources and eviction signals”.

D:

This routine guards against another mechanism: A NoSchedule taint excludes Pods lacking a matching toleration even if the Node has free resources. The recurring exposure in the current case is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The matching preventive check is “Baseline memory requests and track node allocatable resources and eviction signals”.

Question 1113

At Pinecrest logistics, in a sandbox Kubernetes cluster during a recovery tabletop exercise: Node worker-c reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which command or initial safe diagnostic step best addresses the stated lab task?

  1. On worker-b, run sudo systemctl status kubelet and sudo journalctl -u kubelet -n 80 --no-pager along with kubectl describe node worker-b from an admin context.
  2. Run kubectl -n observability describe pod metrics-api and compare CPU requests with kubectl describe node worker-b allocatable resources.
  3. Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests.
  4. Run kubectl -n observability describe pod metrics-api and compare Pod tolerations with kubectl describe node worker-b taints.

Explanation

Correct Answer

C:

Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests. This is the appropriate command-line action or first safe diagnostic. The technical reason is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Validate the result with: Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning. Execute changes only in an authorized disposable lab or after normal production change controls.

Incorrect Answers

A:

The proposed action belongs to the mechanism: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. Here the stated fault follows a different mechanism: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The corrective choice is “Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests”.

B:

The offered action targets: The scheduler places Pods by requested resources and constraints, not the current shell's observed free CPU alone. The case facts point instead to: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Implement “Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests” to address that specific condition.

D:

The proposed action belongs to the mechanism: A NoSchedule taint excludes Pods lacking a matching toleration even if the Node has free resources. Here the stated fault follows a different mechanism: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The corrective choice is “Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests”.

Question 1114

At Pinecrest logistics, in a sandbox Kubernetes cluster during a recovery tabletop exercise: Node worker-c reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which Kubernetes behavior best explains the required change or diagnosis?

  1. The scheduler places Pods by requested resources and constraints, not the current shell's observed free CPU alone.
  2. Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service.
  3. Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime.
  4. Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

Explanation

Correct Answer

D:

Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. This explains the stated condition. The corresponding action is: Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests. The direct acceptance evidence is: Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning.

Incorrect Answers

A:

The offered statement is true for this other mechanism: The scheduler places Pods by requested resources and constraints, not the current shell's observed free CPU alone. It does not explain the facts described here, which follow from: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

B:

This principle describes a separate cause: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. The reported condition is governed by: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The two mechanisms should not be conflated.

C:

Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. That explanation applies to a different technical failure. Here the controlling principle is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

Question 1115

At Pinecrest logistics, in a sandbox Kubernetes cluster during a recovery tabletop exercise: Node worker-c reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which observation best verifies the requested outcome rather than an unrelated state?

  1. Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning.
  2. Confirm the expected CNI Pod is Ready on worker-c and test Pod-to-Pod connectivity from that node.
  3. Run kubectl -n observability get daemonset metrics-api and inspect desired/ready counts after the specific node issue is fixed.
  4. Run kubectl -n fulfillment get daemonset dispatch-api and inspect desired/ready counts after the specific node issue is fixed.

Explanation

Correct Answer

A:

Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning. This validates the affected object or control path, rather than only showing a command completed. The underlying reason is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The intended action is: Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests.

Incorrect Answers

B:

The alternative verifies a different condition: Kubernetes network connectivity requires a functioning CNI implementation on nodes; core API objects alone do not create functional Pod networking. The current case instead depends on: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Confirm that “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

C:

That test belongs to another control with this basis: DaemonSets schedule according to node selection and toleration rules; unavailable node agents can break networking or logging on selected workers. In the current case, the governing reason is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The direct verification is “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

D:

This observation would test a different technical dependency: DaemonSets schedule according to node selection and toleration rules; unavailable node agents can break networking or logging on selected workers. The current issue is governed by: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Its decisive check is “Verify the node returns to Ready without MemoryPressure and Pod eviction rates stabilize after capacity planning”.

Question 1116

At Pinecrest logistics, in a sandbox Kubernetes cluster during a recovery tabletop exercise: Node worker-c reports MemoryPressure during a workload burst while several Pods have no realistic memory requests. Work only within the stated target and verify before declaring success. Which operational control most directly prevents this failure or configuration drift?

  1. Monitor kubelet health and coordinate OS upgrades with supported runtime versions.
  2. Baseline memory requests and track node allocatable resources and eviction signals.
  3. Gate worker acceptance on CNI readiness and a cross-node connectivity check.
  4. Monitor metrics-server availability and required resource requests when enabling autoscaling.

Explanation

Correct Answer

B:

Baseline memory requests and track node allocatable resources and eviction signals. This reduces recurrence of the specific operational failure. It reinforces the action: Run kubectl describe node worker-c and kubectl top node worker-c where metrics are available before altering Pod requests. The relevant mechanism is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.

Incorrect Answers

A:

The suggested review tracks a different technical risk: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. Here the sustained control must address: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Choose “Baseline memory requests and track node allocatable resources and eviction signals”.

C:

This routine guards against another mechanism: Kubernetes network connectivity requires a functioning CNI implementation on nodes; core API objects alone do not create functional Pod networking. The recurring exposure in the current case is: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The matching preventive check is “Baseline memory requests and track node allocatable resources and eviction signals”.

D:

The suggested review tracks a different technical risk: HPA CPU utilization is computed from reported metrics relative to requests; missing metrics or requests can prevent scaling decisions. Here the sustained control must address: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. Choose “Baseline memory requests and track node allocatable resources and eviction signals”.

Question 1117

At Northstar finance, in an isolated training cluster during a scheduled maintenance rehearsal: Node worker-a becomes NotReady and kubelet logs contain container runtime unavailable errors after a host patch. Work only within the stated target and verify before declaring success. Which command or initial safe diagnostic step best addresses the stated lab task?

  1. Run kubectl describe node worker-c and on the host inspect df -h and kubelet eviction messages.
  2. Run kubectl describe node worker-a and on the host inspect df -h and kubelet eviction messages.
  3. On the control-plane host, run sudo crictl ps -a and sudo journalctl -u kubelet -n 100 --no-pager to inspect kube-apiserver container failures.
  4. On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration.

Explanation

Correct Answer

D:

On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration. This is the appropriate command-line action or first safe diagnostic. The technical reason is: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Validate the result with: Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery. Execute changes only in an authorized disposable lab or after normal production change controls.

Incorrect Answers

A:

The offered action targets: DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads. The case facts point instead to: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Implement “On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration” to address that specific condition.

B:

This alternative is a response to another technical cause: DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads. The reported case instead depends on: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. The action that directly addresses it is “On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration”.

C:

The proposed action belongs to the mechanism: Kubeadm control-plane components are commonly static Pods managed by kubelet; kubectl cannot diagnose them if the API server itself is unavailable. Here the stated fault follows a different mechanism: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. The corrective choice is “On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration”.

Question 1118

At Northstar finance, in an isolated training cluster during a scheduled maintenance rehearsal: Node worker-a becomes NotReady and kubelet logs contain container runtime unavailable errors after a host patch. Work only within the stated target and verify before declaring success. Which Kubernetes behavior best explains the required change or diagnosis?

  1. Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime.
  2. Kubeadm control-plane components are commonly static Pods managed by kubelet; kubectl cannot diagnose them if the API server itself is unavailable.
  3. Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status.
  4. DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads.

Explanation

Correct Answer

A:

Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. This explains the stated condition. The corresponding action is: On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration. The direct acceptance evidence is: Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery.

Incorrect Answers

B:

Kubeadm control-plane components are commonly static Pods managed by kubelet; kubectl cannot diagnose them if the API server itself is unavailable. That explanation applies to a different technical failure. Here the controlling principle is: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime.

C:

The offered statement is true for this other mechanism: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. It does not explain the facts described here, which follow from: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime.

D:

This principle describes a separate cause: DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads. The reported condition is governed by: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. The two mechanisms should not be conflated.

Question 1119

At Northstar finance, in an isolated training cluster during a scheduled maintenance rehearsal: Node worker-a becomes NotReady and kubelet logs contain container runtime unavailable errors after a host patch. Work only within the stated target and verify before declaring success. Which observation best verifies the requested outcome rather than an unrelated state?

  1. Run kubectl get node worker-c and confirm Ready after restoring the actual failed component.
  2. Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery.
  3. Run kubectl get node worker-a and confirm Ready after restoring the actual failed component.
  4. Confirm local API server container health and successful kubectl get --raw=/readyz after restoring the valid manifest.

Explanation

Correct Answer

B:

Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery. This validates the affected object or control path, rather than only showing a command completed. The underlying reason is: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. The intended action is: On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration.

Incorrect Answers

A:

The alternative verifies a different condition: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. The current case instead depends on: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Confirm that “Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery”.

C:

This observation would test a different technical dependency: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. The current issue is governed by: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Its decisive check is “Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery”.

D:

The alternative verifies a different condition: Kubeadm control-plane components are commonly static Pods managed by kubelet; kubectl cannot diagnose them if the API server itself is unavailable. The current case instead depends on: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Confirm that “Confirm the CRI service is healthy and kubectl get node worker-a returns Ready after supported recovery”.

Question 1120

At Northstar finance, in an isolated training cluster during a scheduled maintenance rehearsal: Node worker-a becomes NotReady and kubelet logs contain container runtime unavailable errors after a host patch. Work only within the stated target and verify before declaring success. Which operational control most directly prevents this failure or configuration drift?

  1. Monitor kubelet health and coordinate OS upgrades with supported runtime versions.
  2. Baseline memory requests and track node allocatable resources and eviction signals.
  3. Test kubelet/CRI version compatibility and runtime services in node patch runbooks.
  4. Maintain image garbage-collection, log rotation and filesystem capacity alerts.

Explanation

Correct Answer

C:

Test kubelet/CRI version compatibility and runtime services in node patch runbooks. This reduces recurrence of the specific operational failure. It reinforces the action: On worker-a, run sudo systemctl status containerd and sudo journalctl -u containerd -n 80 --no-pager before modifying runtime configuration. The relevant mechanism is: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime.

Incorrect Answers

A:

That safeguard is relevant to: Node readiness depends on the kubelet and node conditions; application restarts cannot fix a stopped kubelet service. The current failure instead follows from: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Its prevention measure is “Test kubelet/CRI version compatibility and runtime services in node patch runbooks”.

B:

This routine guards against another mechanism: Node memory pressure can cause evictions and scheduling instability independent of one container's OOMKilled status. The recurring exposure in the current case is: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. The matching preventive check is “Test kubelet/CRI version compatibility and runtime services in node patch runbooks”.

D:

That safeguard is relevant to: DiskPressure is a node condition triggered by storage thresholds; deleting live runtime data blindly can corrupt workloads. The current failure instead follows from: Kubelet relies on a CRI-compatible runtime for container creation; restarting application Pods cannot repair an unavailable runtime. Its prevention measure is “Test kubelet/CRI version compatibility and runtime services in node patch runbooks”.