Microsoft AZ-104 Exam Dumps, Practice Test Questions

Moving resources between resource groups changes their management container but not the VM or managed disk deployment region. The workload still runs in the original region; a supported regional migration and dependent-resource cutover are necessary.

A VM's deployed region is not a mutable setting that can be reassigned in place by an incremental template update. Regional relocation must account for actual compute, disk and network resources and application cutover.

Resource Mover can help orchestrate supported resource moves between regions, but a metadata-only move cannot relocate the VM. The plan must migrate the supported VM and associated dependencies, then validate and cut over the application.

Deallocation preserves attached persistent storage by default, allowing restart with existing data. Automatic disk deletion is not implied.

A correctly deallocated VM can continue generating disk storage charges. The presence of those charges does not contradict the reported compute state.

Managed disks are separately metered storage resources. DNS use is not the necessary explanation for ongoing disk charges.

Deleting disks would destroy the persistent state that must be retained; the requirement is host replacement, not a new system image.

A guest restart could fix an operating-system problem, but it does not explicitly request Azure to allocate a different underlying host after host trouble is suspected.

Changing ARM resource group metadata does not itself relocate the VM compute instance to a healthy physical host.

Dynamic address allocation is not pinned to the VM resource name. After deallocation, the public address may be reassigned even when the VM name remains the same.

Private IP allocation and the associated public IP allocation are different mechanisms. The relevant allowlist risk in this case is the dynamic public IPv4 address.

Resize operations do not guarantee that a dynamic public IP becomes statically reserved. The administrator must explicitly manage address allocation requirements.

Persistent storage preserves data across compute interruption; it does not guarantee uninterrupted Spot VM capacity.

Spot capacity is explicitly interruptible and should not be treated as more reliable than standard allocated capacity.

A generous price cap may avoid price-based eviction, but Azure can still reclaim Spot capacity when it is needed elsewhere.

Reimage reconstructs the OS state and can discard local changes; relying on it to preserve unique application output is unsafe.

The delete option on an attached persistent data disk affects that disk's lifecycle, not files stored on the separate ephemeral OS disk. The report was never placed on the managed data disk, so it remains vulnerable to reimage loss.

Encryption protects confidentiality of stored bytes but does not make host-local OS changes durable across reimage or VM lifecycle events. Irreplaceable reports must be saved to persistent external storage or a supported managed data disk.

Guest shutdown can stop the operating system while the VM remains Azure-allocated. Only the Azure stop/deallocate operation releases the provisioned compute allocation for billing purposes.

Azure VM PowerOff can leave a VM in Stopped (allocated), a state that is still billed for compute. The distinct stop-and-deallocate operation releases the host allocation. An Azure-initiated power-off is not the same as deallocation.

A future automatic shutdown policy is not evidence that the VM has already entered Stopped (deallocated). For the current stopped-but-allocated instance, explicitly issue a stop/deallocate operation and verify the resulting state; separately billed disks may continue to incur charges.

Delete eviction can remove the VM resource and may affect attached resource recovery. It does not satisfy the requirement to keep the VM ready for restart.

Snapshots can preserve data for later recreation, but they do not prevent the Spot platform's Delete eviction policy from removing the VM resource. The scenario requires retaining a restartable VM with its disks, for which the supported Deallocate policy is the closer fit.

Spot eviction is driven by Azure capacity and pricing policies, not the principal's RBAC role. Owner cannot guarantee the VM remains allocated.

The VM’s zonal placement is not a mutable in-place property merely because the instance is deallocated. The move requires a supported migration or creation of a VM in the destination zone.

Availability sets are not an in-place zonal migration mechanism. VM placement into a new zone requires the supported redeployment or migration procedure with compatible resources.

Disk redundancy does not assign the running VM to the requested compute availability zone. A supported VM move or new deployment is still needed.

Network privileges are not the issue if the size is unavailable in the region. Additional RBAC rights cannot override the Compute catalog.

Quota limits restrict how much of a supported compute family a subscription can allocate. A quota increase does not add a VM series absent from the region’s offered SKU catalog. Choose a supported regional size or plan a migration.

Zonal capacity differences matter when a VM SKU is offered in the region, but the stem states that the target series is not offered anywhere in that region. A zone switch cannot create a missing regional SKU. Deallocation and zone selection do not substitute for checking availability.

A VM NIC's DNS server configuration affects the guest's name-resolution behavior, not publication of an Azure-managed cloudapp.azure.com FQDN. The requested regional DNS label belongs on the public IP resource.

Azure resource-group names are management identifiers inside a subscription, not automatic public hostnames. Renaming the group is not supported in place and would not register an A record in cloudapp.azure.com. The public IP's domain-name-label property is the appropriate DNS configuration point.

An NSG rule filters traffic by addresses, ports, protocol and priority. Its descriptive rule name doesn't publish an Azure-managed DNS record or modify an IP resource's hostname. Even after setting the correct label, the workload still needs suitable network permissions and a listening service.