01. Application owners report intermittent slowness on VMs in a VCF workload domain cluster. In VCF Operations, the engineer sees that several hosts in that cluster show sustained memory ballooning and compression, and a colleague proposes ordering two more hosts.
Following the performance-management approach documented for VCF Operations, what should the engineer do FIRST?
a) Check the Cluster Utilization dashboard for the ballooning hosts
b) Check the Cluster Contention dashboard for KPI breaches
c) Add the two hosts and let DRS rebalance the VMs across the cluster
d) Raise memory reservations on the ballooning VMs to their configured size
02. An engineer edits a VKS cluster to a newer compatible VKr, but the rolling update never starts. The cluster shows:
SystemChecksSucceeded = False, reason NotSucceeded
Message: MisconfiguredSoftwareChecks failed
A third-party admission webhook was installed in the cluster last month.
What should the engineer do FIRST?
a) Fix or remove the blocking webhook
b) Upgrade the Supervisor to trigger a fresh rolling update of the cluster
c) Annotate the cluster with kubernetes.vmware.com/dangerous-skip-misconfigured-software-check-for-update and retry the edit
d) Revert the cluster to its previous VKr and reapply the new one with kubectl apply
03. After a maintenance window, the Summary tab of a VCF workload domain cluster shows Cluster Services health as Unhealthy, and DRS has stopped generating recommendations. An engineer browsing the Hosts and Clusters view reports that the cluster contains no vCLS VMs at all.
What should the engineer do NEXT?
a) Change the DRS automation level to fully automated and run DRS again for the cluster
b) Deactivate vSphere HA admission control and re-enable DRS on the cluster
c) Restart the vpxd service with vmon-cli on the vCenter appliance
d) Open the VMs and Templates view and check the vCLS folder for the vCLS VMs and their power state
04. A Tier-1 gateway providing NAT runs active-standby on edge-01 (preferred) and edge-02. Edge-01 was rebooted for maintenance, and the Tier-1 SR on edge-02 became active.
Edge-01 has returned with node status Up and all BFD sessions up.
The Tier-1 SR on edge-01 is still standby an hour later.
The Tier-1 failover mode was never changed from its default.
Why has the active role not returned to edge-01?
a) Standby relocation moved the standby SR to a third edge node
b) The failover mode is non-preemptive
c) The Tier-1 must be switched to active-active before edge-01 can resume
d) Edge-01 still has BFD sessions down and has taken itself out of service
05. After a top-of-rack switch upgrade, vSAN Skyline Health for a VCF Workload Domain cluster shows several Unhealthy findings across the network, data and performance categories. The network team reports that the uplink misconfiguration introduced by the upgrade has now been corrected.
Which two actions should the engineer take in Skyline Health NEXT?
(Choose two.)
a) Retest the health findings instead of waiting for the next periodic run
b) Silence the Unhealthy findings that were raised during the switch upgrade
c) Sort the findings by root cause and resolve the remaining primary findings first
d) Run the proactive VM creation test on each production host
e) Turn off the Historical Health Service to clear the health data recorded during the upgrade
06. A developer deploys a workload to a VKS cluster whose VKr runs Kubernetes 1.30. The pods do not start and report Error: container has runAsNonRoot and image will run as root. The images pull from the registry successfully.
Which change resolves the error?
a) Add the registry CA to the cluster's trusted CAs
b) Create a PodSecurityPolicy and bind it to the authenticated users group in the cluster
c) Configure Pod Security Admission for authenticated users on the cluster
d) Recreate the image pull secret for the registry
07. VMs on a VMFS datastore backed by a Fibre Channel LUN report slow disk response. On one ESX host the engineer runs esxtop and opens the storage device panel for the LUN:
DAVG/cmd: 38 ms
KAVG/cmd: 0.1 ms
QAVG/cmd: 0 ms
GAVG/cmd: 38 ms
Where should the engineer focus the investigation NEXT?
a) On the device queue depth and adapter queue settings of the host
b) On the storage array and SAN fabric that serve the LUN
c) On VMkernel CPU contention on the ESX host
d) On the guest operating system inside the affected VMs
08. An engineer reviews the Datastore Performance dashboard in VCF Operations for two shared VMFS datastores on a Fibre Channel array over the last seven days:
DS-SQL-01: worst VM disk latency 41 ms, 95th percentile 38 ms
DS-WEB-02: worst VM disk latency 55 ms, 95th percentile 4 ms
Which conclusion do these values support?
a) DS-SQL-01 has a sustained latency problem to investigate
b) Both datastores show a sustained latency problem at the same severity
c) Neither datastore has a latency problem
d) DS-WEB-02 has the more serious problem, with the highest worst-case latency
09. A VCF Operations administrator created a custom alert definition for ESX hosts in a management domain. The definition has two symptom sets: one for sustained CPU contention and one for a host property value. The alert-definition wait and cancel cycles are 1, and the definition is active in the default policy.
The CPU contention symptom shows as triggered on esx-mgmt-03.
The property symptom is not triggered on any host.
No alert for this definition appears in the Alerts list for esx-mgmt-03.
What is the most likely reason no alert was generated?
a) The definition has no notification rule or running outbound plug-in for esx-mgmt-03 and its parent domain
b) The alert impact is set to Efficiency instead of Health
c) The definition requires all symptom sets to be true, and only one set is true
d) The alert-definition wait cycle must be raised above 1 first
10. In an All Apps organization, a project administrator opens Kubernetes Management > Policies > Insights and sees a single Threshold insight against the payments-dev namespace. No Violation, Sync or Health insights are listed. A developer asks which Gatekeeper constraint their pods broke.
What does the insight indicate?
a) The quota policy failed to synchronize from VKS cluster management to the cluster
b) Resources in the namespace are non-compliant with a Gatekeeper-based policy
c) Gatekeeper on the VKS cluster is not managed by VKS cluster management
d) Namespace usage exceeds 80% of a quota policy