Microsoft AZ-104 Exam Dumps, Practice Test Questions

A snapshot is a read-only recovery artifact rather than a standard writable managed disk. It cannot be used as a normal writable VM data disk without creating a disk.

A snapshot represents a read-only point-in-time disk state. A writable VM disk must be created from the snapshot to be attached for normal modification.

Changing the original disk would risk modifying production data and does not exercise the isolated pre-upgrade snapshot. A separate disk derived from the snapshot is safer.

Caching affects IO behavior; it does not initialize partition tables, file systems or mount points inside Linux.

Host caching affects the disk's IO caching, not guest filesystem creation. The attached disk can be visible through lsblk but still needs an appropriate partition/filesystem and mount in the Linux guest before an application writes files.

More capacity changes the block device specification, but the guest still needs the filesystem prepared and mounted before it can hold application files.

Stopping the VM contradicts the requirement that it remain running, and a VM's boot OS disk cannot be detached as a normal data disk while preserving that running system.

OS and data disk lifecycles differ. Assuming every attached disk can be removed live would make the original VM unbootable.

Disk caching affects I/O behavior but does not change the disk's role as the VM boot volume. Changing cache mode does not make the OS disk hot-detachable as a data disk.

VM size support for an advanced data disk cannot override Premium SSD v2’s separate OS-disk prohibition. Select a supported boot disk type and, if compatible, attach Premium SSD v2 as a data disk.

Gen2 boot and managed disk SKU eligibility are distinct constraints. Premium SSD v2 is not supported as an operating-system disk even on a Gen2 VM; changing firmware generation cannot make that disk type bootable.

Managed disk offerings have different supported scenarios. Treating an advanced data-disk SKU as universally bootable causes the validation failure.

Caching capabilities differ by disk offering. Treating all disk tiers as equivalent creates invalid configuration attempts.

Premium SSD v2 cannot be an OS disk and does not gain host caching by changing its attachment role.

Premium SSD v2 does not support Azure VM host caching in either ReadOnly or ReadWrite mode. Selecting another caching mode cannot overcome an unsupported disk-SKU capability; the database architecture must use supported Premium SSD v2 performance characteristics or a different disk type.

IOPS affects disk performance, not whether a snapshot is full or incremental. It cannot replace an incremental snapshot configuration.

A full snapshot represents the disk's entire point-in-time contents; it is not equivalent to the incremental snapshot feature. Repeating it can waste recovery storage.

Snapshot redundancy controls placement and resilience of snapshot data, not whether the snapshot captures all blocks or changed blocks. The incremental snapshot type must be selected for change-based disk snapshots.

Caching affects data-path performance and consistency; it does not set the provisioned capacity used for managed-disk billing. No disk shrink would occur.

Resizing a filesystem partition changes guest-level allocation but does not reduce the Azure managed disk’s provisioned disk resource. Azure does not support shrinking that managed disk in place. A smaller replacement disk and a validated migration are required to lower provisioned disk capacity safely.

The Azure disk resource does not support shrinking simply because guest free space exists. Truncating the managed disk could also risk data loss, so the operation is not equivalent to a safe storage migration.

At-rest encryption protects persistent managed-disk data, but the scenario specifically adds temporary disks and the host data path. Those require the supported encryption-at-host feature.

Guest disk encryption can protect guest volumes but does not guarantee protection for all temporary host paths identified in the requirement. Host encryption is the targeted control.

TLS protects application network transport, not the VM temporary disks or host-to-storage disk path. Those are separate data-protection layers.

Encryption at host protects host-side caches, temporary disks and eligible storage paths, but enabling it does not make an unsupported Premium SSD v2 managed disk eligible for double encryption at rest. The requirement is for two independent managed-disk storage encryption layers. Choose a supported disk SKU and compatible encryption set rather than relabeling host encryption as the unsupported feature.

Default server-side encryption provides encryption at rest for Azure managed disks, but the requirement expressly calls for an additional independent infrastructure layer. Relabeling ordinary SSE as double encryption would misstate the implemented protection. The operator must use a disk SKU supporting the extra layer.

A resource-group move changes management scope, not the underlying Azure managed-disk SKU feature matrix. Premium SSD v2 remains unsupported for double encryption regardless of group naming or location metadata. The fix needs a supported disk design, not an ARM organizational change.

Encryption at host adds protection for temporary paths and cache, but by itself does not select an organization-controlled customer key for managed disk SSE.

Azure Disk Encryption encrypts inside the guest using BitLocker or dm-crypt, with keys stored or protected via Key Vault. It does not configure Azure Disk Storage server-side encryption with customer-managed keys on the managed disk resource. The stated compliance boundary specifically requires server-side disk encryption, which is bound through a Disk Encryption Set rather than guest-based ADE.

Granting a VM identity permission to use a Key Vault key does not associate that key with Azure Disk Storage server-side encryption. The managed disk must reference an appropriate Disk Encryption Set whose managed identity can access the selected key. Key Vault permission alone is insufficient to change managed-disk server-side encryption from platform-managed to customer-managed keys.

The two mechanisms are not universally composable. Ignoring their documented restriction can cause the requested configuration to be rejected.

Removing an extension without undoing the guest ADE configuration does not overcome the Azure encryption-at-host compatibility restriction for VMs that have used ADE.

Supported VM size is a separate prerequisite; it does not undo the documented incompatibility for a VM currently or previously configured with Azure Disk Encryption.