Question 560A CRD upgrade removes a served version while stored objects still depend on the old conversion path. Which Kubernetes administration action most directly addresses the observed failure? Install the CRD and wait for its Established condition before applying dependent custom resources. Plan served and storage-version conversion carefully and test existing object reads before changing the CRD. Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. Grant the controller the minimum required update or patch permission for the status subresource. ExplanationCorrect Answer B:Plan served and storage-version conversion carefully and test existing object reads before changing the CRD. This corrects the specific failure because cRD versioning separates served API versions from the storage version and may require conversion webhooks. Verify it by checking: Existing custom objects are readable and writable through the supported new version after migration.Incorrect Answers A:The API server must register an extension type before it can accept objects of that kind. That is relevant to a different intervention. For the stated situation, the required action is: Plan served and storage-version conversion carefully and test existing object reads before changing the CRD. The governing reason is: CRD versioning separates served API versions from the storage version and may require conversion webhooks. C:A CRD defines the API schema while a running controller implements the desired behavior. That is relevant to a different intervention. For the stated situation, the required action is: Plan served and storage-version conversion carefully and test existing object reads before changing the CRD. The governing reason is: CRD versioning separates served API versions from the storage version and may require conversion webhooks. D:Custom resource status updates are separately authorized from spec updates when status is enabled. That is relevant to a different intervention. For the stated situation, the required action is: Plan served and storage-version conversion carefully and test existing object reads before changing the CRD. The governing reason is: CRD versioning separates served API versions from the storage version and may require conversion webhooks.
Question 561A storage class uses Immediate binding in a multi-zone cluster and creates a disk before the Pod's node is selected. Following the proposed change—Use WaitForFirstConsumer binding for topology-constrained storage where supported—which observation best demonstrates that the particular fault is resolved? A new PVC binds in the same topology as its consuming Pod and the Pod starts. A test unqualified PVC picks the expected class and critical manifests remain explicit. A test PVC becomes Bound and the expected CSI volume is created. PVC capacity grows to the requested value and the filesystem reports additional usable space. ExplanationCorrect Answer A:A new PVC binds in the same topology as its consuming Pod and the Pod starts. This validates the stated intervention—use WaitForFirstConsumer binding for topology-constrained storage where supported.—rather than an unrelated cluster change. The relevant mechanism is: Delayed volume binding allows scheduling constraints and storage topology to be considered together.Incorrect Answers B:PVCs without a specified class can inherit the cluster's default storage class. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A new PVC binds in the same topology as its consuming Pod and the Pod starts. C:Dynamic provisioning depends on a controller that recognizes the StorageClass provisioner identifier. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A new PVC binds in the same topology as its consuming Pod and the Pod starts. D:PVC size expansion depends on both StorageClass permission and storage-driver capabilities. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A new PVC binds in the same topology as its consuming Pod and the Pod starts.
Question 562A team wants its viewer ClusterRole to cover a new namespaced custom resource but the permissions do not propagate. Which Kubernetes administration action most directly addresses the observed failure? Replace the cluster-wide binding with a namespaced RoleBinding granting only needed resource verbs. Bind Secret get permissions only to that service account for the required namespace and named resources when feasible. Update the intended ClusterRole rules or its controlled aggregation labels and verify the bound roles. Replace wildcard rules with explicit get list watch permissions on the exact resources the collector reads. ExplanationCorrect Answer C:Update the intended ClusterRole rules or its controlled aggregation labels and verify the bound roles. This corrects the specific failure because aggregated ClusterRoles rely on matching labels on source ClusterRoles; ordinary ClusterRoles do not aggregate automatically. Verify it by checking: The viewer can list only the approved custom resources in the bound namespace after reconciliation.Incorrect Answers A:A RoleBinding restricts a Role or referenced ClusterRole to its own namespace; a ClusterRoleBinding grants across namespaces. That is relevant to a different intervention. For the stated situation, the required action is: Update the intended ClusterRole rules or its controlled aggregation labels and verify the bound roles. The governing reason is: Aggregated ClusterRoles rely on matching labels on source ClusterRoles; ordinary ClusterRoles do not aggregate automatically. B:Kubernetes service accounts do not automatically inherit administrator permissions from the Pod owner. That is relevant to a different intervention. For the stated situation, the required action is: Update the intended ClusterRole rules or its controlled aggregation labels and verify the bound roles. The governing reason is: Aggregated ClusterRoles rely on matching labels on source ClusterRoles; ordinary ClusterRoles do not aggregate automatically. D:Least-privilege RBAC narrows blast radius by limiting the allowed verbs, resources, and API groups. That is relevant to a different intervention. For the stated situation, the required action is: Update the intended ClusterRole rules or its controlled aggregation labels and verify the bound roles. The governing reason is: Aggregated ClusterRoles rely on matching labels on source ClusterRoles; ordinary ClusterRoles do not aggregate automatically.
Question 563A cluster permits dual-stack networking but a workload assumes every endpoint address is IPv4. Following the proposed change—Validate the Pod and Service IP family policy and ensure clients handle supported IPv4/IPv6 endpoint families—which observation best demonstrates that the particular fault is resolved? A test client connects over each supported IP family to healthy backends. A test client connects directly to a Pod IP on another node and network counters show the expected path. Clients continue connecting after the target Pod is deleted and replaced. DNS resolution returns the intended namespace's Service IP and the target responds. ExplanationCorrect Answer A:A test client connects over each supported IP family to healthy backends. This validates the stated intervention—validate the Pod and Service IP family policy and ensure clients handle supported IPv4/IPv6 endpoint families.—rather than an unrelated cluster change. The relevant mechanism is: Dual-stack Services and Pods may expose more than one IP family depending on cluster configuration.Incorrect Answers B:Kubernetes expects Pod-to-Pod reachability without application-managed NAT across nodes. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A test client connects over each supported IP family to healthy backends. C:Pods are replaceable and their IP addresses can change even when the application identity remains. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A test client connects over each supported IP family to healthy backends. D:Unqualified DNS lookups are affected by search domains in the caller's namespace. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A test client connects over each supported IP family to healthy backends.
Question 564Pods can reach an API server but a replaced node is marked NotReady due to a container runtime socket path change. Which Kubernetes administration action most directly addresses the observed failure? Inspect kubelet service status, node conditions and runtime endpoint, then repair the failing component. Update the kubelet CRI endpoint configuration to the installed runtime's actual socket. Restore the CNI node plugin and inspect CNI configuration files for incompatible changes. Renew or rotate kubelet client credentials using the supported cluster certificate process. ExplanationCorrect Answer B:Update the kubelet CRI endpoint configuration to the installed runtime's actual socket. This corrects the specific failure because the kubelet must connect to the correct container runtime interface to manage Pod sandboxes. Verify it by checking: Kubelet logs stop showing runtime endpoint errors and the node becomes Ready.Incorrect Answers A:A node's Ready condition depends on kubelet heartbeats and necessary local runtime services. That is relevant to a different intervention. For the stated situation, the required action is: Update the kubelet CRI endpoint configuration to the installed runtime's actual socket. The governing reason is: The kubelet must connect to the correct container runtime interface to manage Pod sandboxes. C:Node readiness can depend on successful Pod network initialization. That is relevant to a different intervention. For the stated situation, the required action is: Update the kubelet CRI endpoint configuration to the installed runtime's actual socket. The governing reason is: The kubelet must connect to the correct container runtime interface to manage Pod sandboxes. D:Node status updates require authenticated kubelet API requests using valid credentials. That is relevant to a different intervention. For the stated situation, the required action is: Update the kubelet CRI endpoint configuration to the installed runtime's actual socket. The governing reason is: The kubelet must connect to the correct container runtime interface to manage Pod sandboxes.
Question 565An internal app needs to reach a Pod using its own namespace-local DNS alias, but the client uses a stale Pod hostname. Following the proposed change—Use the intended Service name with namespace-aware DNS resolution and test reachability—which observation best demonstrates that the particular fault is resolved? DNS resolution returns the intended namespace's Service IP and the target responds. Clients continue connecting after the target Pod is deleted and replaced. The client resolves the current Service record and reaches a ready backend. A test client connects directly to a Pod IP on another node and network counters show the expected path. ExplanationCorrect Answer C:The client resolves the current Service record and reaches a ready backend. This validates the stated intervention—use the intended Service name with namespace-aware DNS resolution and test reachability.—rather than an unrelated cluster change. The relevant mechanism is: Kubernetes Service DNS gives workloads a stable logical endpoint independent of Pod replacement.Incorrect Answers A:Unqualified DNS lookups are affected by search domains in the caller's namespace. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The client resolves the current Service record and reaches a ready backend. B:Pods are replaceable and their IP addresses can change even when the application identity remains. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The client resolves the current Service record and reaches a ready backend. D:Kubernetes expects Pod-to-Pod reachability without application-managed NAT across nodes. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The client resolves the current Service record and reaches a ready backend.
Question 566Only Pods on one worker cannot resolve names and their resolv.conf points at an obsolete cluster DNS IP. Which Kubernetes administration action most directly addresses the observed failure? Restore a valid Corefile and validate the CoreDNS configuration before rollout. Inspect CoreDNS Kubernetes plugin configuration and its access to the cluster API and service domain. Correct the kubelet clusterDNS configuration or node DNS distribution and restart affected Pods as needed. Use a Service DNS name or other stable discovery mechanism instead of a Pod IP. ExplanationCorrect Answer C:Correct the kubelet clusterDNS configuration or node DNS distribution and restart affected Pods as needed. This corrects the specific failure because pod DNS settings may be generated from kubelet node configuration and dnsPolicy. Verify it by checking: New Pods on the worker receive the current cluster DNS IP and resolve known Services.Incorrect Answers A:CoreDNS must parse its plugin chain before it can serve cluster DNS queries. That is relevant to a different intervention. For the stated situation, the required action is: Correct the kubelet clusterDNS configuration or node DNS distribution and restart affected Pods as needed. The governing reason is: Pod DNS settings may be generated from kubelet node configuration and dnsPolicy. B:Cluster-local DNS records depend on CoreDNS observing Service and EndpointSlice resources. That is relevant to a different intervention. For the stated situation, the required action is: Correct the kubelet clusterDNS configuration or node DNS distribution and restart affected Pods as needed. The governing reason is: Pod DNS settings may be generated from kubelet node configuration and dnsPolicy. D:Pods are replaceable and their IP addresses can change even when the application identity remains. That is relevant to a different intervention. For the stated situation, the required action is: Correct the kubelet clusterDNS configuration or node DNS distribution and restart affected Pods as needed. The governing reason is: Pod DNS settings may be generated from kubelet node configuration and dnsPolicy.
Question 567A node's kubelet reports NetworkPluginNotReady after a CNI upgrade although the runtime is healthy. Following the proposed change—Restore the CNI node plugin and inspect CNI configuration files for incompatible changes—which observation best demonstrates that the particular fault is resolved? Kubelet logs stop showing runtime endpoint errors and the node becomes Ready. Node conditions return Ready and a scheduled test Pod starts after kubelet recovery. Node DiskPressure clears and new Pods are admitted without deleting active volume data. The CNI plugin reports initialized and regular Pod sandboxes start with usable addresses. ExplanationCorrect Answer D:The CNI plugin reports initialized and regular Pod sandboxes start with usable addresses. This validates the stated intervention—restore the CNI node plugin and inspect CNI configuration files for incompatible changes.—rather than an unrelated cluster change. The relevant mechanism is: Node readiness can depend on successful Pod network initialization.Incorrect Answers A:The kubelet must connect to the correct container runtime interface to manage Pod sandboxes. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The CNI plugin reports initialized and regular Pod sandboxes start with usable addresses. B:A node's Ready condition depends on kubelet heartbeats and necessary local runtime services. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The CNI plugin reports initialized and regular Pod sandboxes start with usable addresses. C:Kubelet disk-pressure signals can cause eviction and scheduling taints independent of CPU utilization. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The CNI plugin reports initialized and regular Pod sandboxes start with usable addresses.
Question 568A production cluster relies on one control-plane VM and its failure would remove API access. Following the proposed change—Design and test multiple control-plane nodes behind a durable API endpoint with compatible etcd quorum—which observation best demonstrates that the particular fault is resolved? An entire host outage leaves a majority of etcd voters and API access available. A TLS connection to the load-balanced hostname succeeds without insecure verification overrides. The endpoint resolves to an active listener and kubeadm discovery succeeds from a worker. A simulated single-control-plane outage retains API availability and healthy etcd quorum. ExplanationCorrect Answer D:A simulated single-control-plane outage retains API availability and healthy etcd quorum. This validates the stated intervention—design and test multiple control-plane nodes behind a durable API endpoint with compatible etcd quorum.—rather than an unrelated cluster change. The relevant mechanism is: Highly available control planes depend on load-balanced API access and durable quorum-backed state.Incorrect Answers A:Quorum and replication do not protect against a common power or hypervisor failure domain. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A simulated single-control-plane outage retains API availability and healthy etcd quorum. B:TLS clients verify the endpoint hostname against the presented certificate SANs. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A simulated single-control-plane outage retains API availability and healthy etcd quorum. C:Clients and join operations need a durable API endpoint independent of one control-plane host. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A simulated single-control-plane outage retains API availability and healthy etcd quorum.
Question 569A replacement worker joins with the same hostname as an existing stale Node object and reconciliation becomes confusing. An engineer recommends: Drain and decommission the old identity cleanly before joining the replacement with a unique node identity. Which Kubernetes behavior best explains why this correction fits? DaemonSet Pods are managed for node coverage and are not treated like ordinary replicated workloads during drain. Node names are identity keys in the API and stale entries can confuse lifecycle operations. Node status updates require authenticated kubelet API requests using valid credentials. Kubelet disk-pressure signals can cause eviction and scheduling taints independent of CPU utilization. ExplanationCorrect Answer B:Node names are identity keys in the API and stale entries can confuse lifecycle operations. This explains why the proposed action—drain and decommission the old identity cleanly before joining the replacement with a unique node identity.—is appropriate. Direct confirmation is: The cluster lists one active intended Node identity and its Lease and Pods reference the replacement correctly.Incorrect Answers A:DaemonSet Pods are managed for node coverage and are not treated like ordinary replicated workloads during drain. This does not establish the proposed intervention. The governing principle here is: Node names are identity keys in the API and stale entries can confuse lifecycle operations. Apply it by: Drain and decommission the old identity cleanly before joining the replacement with a unique node identity. C:Node status updates require authenticated kubelet API requests using valid credentials. This does not establish the proposed intervention. The governing principle here is: Node names are identity keys in the API and stale entries can confuse lifecycle operations. Apply it by: Drain and decommission the old identity cleanly before joining the replacement with a unique node identity. D:Kubelet disk-pressure signals can cause eviction and scheduling taints independent of CPU utilization. This does not establish the proposed intervention. The governing principle here is: Node names are identity keys in the API and stale entries can confuse lifecycle operations. Apply it by: Drain and decommission the old identity cleanly before joining the replacement with a unique node identity.
Question 570A cluster admin enables NodeLocal DNSCache but leaves pods pointing to an inaccessible DNS address. Following the proposed change—Align the node-local cache listen IP and Pod resolver configuration with the deployed cache topology—which observation best demonstrates that the particular fault is resolved? CoreDNS is Ready with clean logs and a test Pod resolves internal and external names. New Pods on the worker receive the current cluster DNS IP and resolve known Services. Queries on affected nodes reach the cache and fall back to upstream CoreDNS appropriately. Lookup returns the intended namespace's Service address and an HTTP probe succeeds. ExplanationCorrect Answer C:Queries on affected nodes reach the cache and fall back to upstream CoreDNS appropriately. This validates the stated intervention—align the node-local cache listen IP and Pod resolver configuration with the deployed cache topology.—rather than an unrelated cluster change. The relevant mechanism is: Node-local DNS caching still requires Pod resolvers to reach the chosen local listener.Incorrect Answers A:CoreDNS must parse its plugin chain before it can serve cluster DNS queries. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: Queries on affected nodes reach the cache and fall back to upstream CoreDNS appropriately. B:Pod DNS settings may be generated from kubelet node configuration and dnsPolicy. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: Queries on affected nodes reach the cache and fall back to upstream CoreDNS appropriately. D:Pod DNS search domains are namespace-relative and short names are not globally unique. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: Queries on affected nodes reach the cache and fall back to upstream CoreDNS appropriately.