Microsoft AZ-104 Exam Dumps, Practice Test Questions

An ARM resource tag labels the storage account as a management-plane resource; it is not equivalent to an index tag stored on each blob. Lifecycle rule filters operate on supported blob types, prefixes, and blob index tags. An account-level tag would not identify which individual files carry Class=Archive.

Prefix matching selects blob names and virtual folders, not the requested metadata classification. The stem explicitly says the Class value must work independently of the blob path, so a prefix-only rule would miss correctly labeled blobs in other paths and could include unlabeled blobs with similar names.

An age threshold can govern when a lifecycle action runs, but by itself it cannot distinguish Class=Archive from unrelated objects of the same age. It would apply the transition to every otherwise eligible blob. Combine age conditions with a matching blob index tag filter so classification, not merely time, determines membership.

Change Feed records mutation events, not a complete point-in-time inventory of all existing objects and historical versions. Blob Inventory reports are designed to produce the requested enumerated compliance snapshot.

Storage Explorer is suitable for interactive inspection, but an ad hoc listing does not create a scheduled machine-readable inventory across containers and version metadata. Blob inventory reports do.

Lifecycle management controls object tiering and deletion; it does not generate an inventory report merely because its actions are disabled. Configure Blob inventory reporting explicitly.

Blob versioning records writes/updates, not a new blob version for every read. It cannot replace Blob service last-access tracking for a daysAfterLastAccessTimeGreaterThan lifecycle condition.

Modification time measures writes, not reads. An actively downloaded but unchanged blob could be incorrectly tiered as stale.

Account-level read metrics do not supply per-blob last-access timestamps for lifecycle evaluation. Last-access time tracking must be enabled and used at the Blob service level.

Hot-tier blobs are normal lifecycle candidates. Eligibility is governed by the rule filters, supported blob type, and age condition.

Azure lifecycle management runs asynchronously under the configured policy without needing each object to be touched by a client. Manual per-object actions are not an activation requirement.

Re-saving a policy does not convert asynchronous lifecycle processing into an immediate transaction. Examine processing timing and logs rather than treating a five-minute delay as a defective rule.

Even a data-plane Owner identity cannot bypass an active legal hold protecting the blob. Authorization and immutability are separate enforcement boundaries.

Neither an access-tier change nor a separate time-based retention configuration cancels an active legal hold. Lifecycle deletion remains blocked until that hold is lawfully cleared.

Immutable storage protection also applies to direct Delete Blob requests. Removing the lifecycle rule cannot remove the active legal hold, so an authorized release of the hold is required first.

The creation-time condition measures the age of the blob since it was created, even when it was modified more recently. A recently edited old blob could meet this condition before the required post-modification wait has elapsed. The requested lifecycle rule should instead evaluate daysAfterModificationGreaterThan.

Last access and last modification are different events: downloads may refresh an access timestamp without changing blob contents, and uploads may modify data without a later read. Even with supported last-access tracking enabled, that condition measures access recency, not the business-required modification age.

The last-tier-change condition is designed for controlling TierToArchive after rehydration, not as a general modification-age measure for TierToCool. Moving between tiers also differs from modifying blob data. For a current blob transition measured from its latest write, use the supported modification-based age condition.

Hard-coded values force manual source edits whenever the environment changes. That makes deployments harder to repeat and increases the risk of deploying test properties into production.

Copying entire files creates two independently maintained definitions for the same design. Differences can drift over time even though the resource topology should be identical.

A Bicep variable is defined within the template and still requires a source change to vary per environment. Parameters are the intended caller-supplied contract.

Tags do not resolve the actual resource instance for Bicep property references. The new resource still needs a valid existing resource reference.

Complete mode may delete out-of-template resources and is unrelated to referencing an existing object safely. It is more destructive than the needed declaration.

An ordinary resource declaration may attempt to deploy or update the resource and can conflict with naming or location constraints. The requirement explicitly forbids redeployment.

Template outputs and ordinary strings can expose sensitive values in deployment records. This increases the precise secret-leakage risk being prevented.

Resource tags are inspectable metadata and are not a secret store. They are inappropriate for deployment credentials.

Embedding the password in source code defeats secure parameter handling and risks disclosure in source control and reviews.

Duplicating declarations can work initially but allows drift and makes future updates inconsistent across environments. It does not meet the reuse objective.

ARM/Bicep parameter files supply inputs for deployments; they do not encapsulate executable resource declarations to be instantiated repeatedly at different scopes. A Bicep module is the reusable declaration unit, with environment-specific parameters passed by the caller.

Bicep outputs expose values produced by one deployment, such as resource IDs or endpoints; they do not themselves redeploy the configuration in a new target resource group. Packaging the stable diagnostics resources in a parameterized Bicep module avoids repeating declarations.

Outputs affect returned deployment values, not the creation of underlying resources. The diagnostic VM would still be provisioned and billed in test.

Parameters can customize names, but without conditional resource creation the VM still exists in both environments, violating the explicit no-VM test requirement.

Blocking connectivity does not remove VM compute charges. The stated requirement is to omit the optional resource, not merely hide its network access.