McAfee-Secured Website

Exam Bundle

Exam Code: AZ-104

Exam Name Microsoft Azure Administrator

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Azure Administrator Associate

Microsoft AZ-104 Bundle $44.99

100% Microsoft AZ-104 Guaranteed Success With Testking Training Materials

Get Real AZ-104 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

  • Questions & Answers

    AZ-104 Questions & Answers

    606 Questions & Answers

    The ultimate exam preparation tool, Questions & Answers mirrors the actual AZ-104 exam while allowing you to hone and measure every aspect of your test-taking ability.

  • AZ-104 Video Course

    AZ-104 Video Course

    132 Video Lectures

    AZ-104 Video Course is developed by Microsoft Professionals to help you pass the AZ-104 exam.

    Description

    This course will improve your knowledge and skills required to pass Microsoft Azure Administrator exam.
  • Study Guide

    AZ-104 Study Guide

    458 PDF Pages

    Developed by industry experts, this 458-page guide spells out in painstaking detail all of the information you need to ace AZ-104 exam.

Microsoft Azure Administrator (AZ-104) Practice Tests

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Total Cost: $194.97
Bundle Price: $149.98

Purchase Individually

  • Questions & Answers

    Questions & Answers

    606 Questions

    $124.99
  • AZ-104 Video Course

    Video Course

    132 Video Lectures

    $39.99
  • Study Guide

    Study Guide

    458 PDF Pages

    $29.99

Microsoft AZ-104 Practice Test 34

Question 330

A storage team retains many snapshots of a large frequently modified block blob. Finance asks whether snapshots are free because the base blob already exists. What cost consideration matters?

  1. Snapshot storage can incur charges for changed data blocks and relevant access or operations, depending on the storage implementation.
  2. Charge the full original blob capacity again for every snapshot, even when all its blocks are unchanged.
  3. Only snapshot creation operations are billable and retained snapshot blocks never accrue capacity cost.
  4. Shorten the base blob lifecycle policy and accumulated snapshots are automatically free.

Explanation

Correct Answer

A:

Snapshot cost can increase as blocks differ from the base and as snapshots accumulate. Retention policy should reflect recovery needs and cost.

Incorrect Answers

B:

Billing is based on relevant unique or changed stored blocks and operations, not automatically a full charge for every unchanged copy.

C:

Retained snapshot data can consume billable storage in addition to relevant operations.

D:

Lifecycle choices do not exempt independently retained snapshot data from applicable capacity charges.

Question 331

A Bicep template declares an integer parameter for VM instance count. A pipeline passes a descriptive text string such as many. What should the developer expect?

  1. Set the parameter default to 1 and the invalid explicitly supplied string is silently ignored.
  2. Add the @secure() decorator to the integer parameter to convert the word into a valid instance count.
  3. Convert the value using int(many) and ARM will infer an integer for an English word.
  4. Deployment validation or compilation should reject the value because it does not satisfy the declared parameter type.

Explanation

Correct Answer

D:

Bicep parameters have declared types, and incompatible values do not satisfy the deployment contract. Use a validated numeric input.

Incorrect Answers

A:

A submitted incompatible value does not become valid merely because the parameter has a numeric default.

B:

Secure parameter handling concerns secrecy, not conversion of arbitrary descriptive text to integers.

C:

The Bicep or ARM integer conversion functions cannot infer a numeric value from an arbitrary word.

Question 332

A maintained Bicep module accumulates obsolete parameters and hard-coded values after several refactors. What developer tool can help find maintainability issues before deployment?

  1. Run ARM deployment validation and this will report all unused Bicep parameters and maintainability diagnostics.
  2. Run Bicep decompilation on the maintained module and generated conversion reports all unused declarations.
  3. Run az deployment group what-if and rely on its resource delta to identify every unused source parameter.
  4. Run the Bicep linter and review supported analyzer diagnostics in the CI pipeline.

Explanation

Correct Answer

D:

Bicep analysis rules identify many style, maintainability and suspicious usage issues before deploying. Each diagnostic still needs appropriate engineering judgment.

Incorrect Answers

A:

Deployment validation examines deployability and parameter/schema requirements, but Bicep lint rules are the suitable static diagnostics for unused declarations and style issues.

B:

Decompilation converts suitable ARM JSON to Bicep; it is not the source-linting pass that detects unused Bicep parameters and style diagnostics.

C:

What-if previews resource-level change, not all local source-code maintainability or unused-declaration lint findings.

Question 333

A release pipeline chooses a production Bicep parameter file whose using declaration references a different template than the file being deployed. What should be checked?

  1. Rename the parameter file without updating its using reference or submitted values.
  2. Any Bicep parameters file can supply any deployment without validation.
  3. Match the parameters file to the intended Bicep template and validate parameter names and types before execution.
  4. Force deployment validation to skip parameter schema checks for the mismatched using declaration.

Explanation

Correct Answer

C:

Parameter files bind deployment inputs to a declared template contract. Wrong references or names can cause validation failure or unintended deployment values.

Incorrect Answers

A:

A filename change does not repair the template reference and parameter contracts inside a bicepparam file.

B:

Parameter files are not universal untyped bags of arbitrary configuration.

D:

Skipping checks would not resolve the underlying mismatch and is not the supported repair for a wrong template binding.

Question 334

An ARM JSON template generates a storage account name from a prefix and a unique suffix. An operator copies the template to a second resource group. What should they verify?

  1. Derive the suffix only from the resource-group name and two subscriptions cannot contain a resource group with that same name.
  2. Azure automatically allows duplicate globally unique storage account names across subscriptions.
  3. Remove all uniqueness expressions and rely on incremental deployment mode to rename a colliding account.
  4. Resource name expressions and deployment scope produce a valid unique name for the target account, including provider naming rules.

Explanation

Correct Answer

D:

ARM expressions are evaluated at deployment and may depend on scope or parameters. Globally named services must satisfy uniqueness and naming constraints.

Incorrect Answers

A:

Resource-group names can recur across subscriptions, while storage account names have wider uniqueness requirements. The expression must be evaluated against actual scope and global provider constraints.

B:

Storage account names are globally unique within Azure's relevant namespace, not just locally unique to a subscription.

C:

Incremental deployment mode does not automatically create a globally distinct resource name when a collision occurs.

Question 335

A team runs ARM what-if before updating a SQL database and claims it proves no customer table rows will be modified by the application's post-deployment migration scripts. What is the limitation?

  1. Enable complete deployment mode to make what-if include the future application migration scripts.
  2. Treat the absence of a predicted Azure SQL resource change as proof no migration script can alter data.
  3. Run the ARM validation endpoint as a substitute for testing the database migration in staging.
  4. ARM what-if predicts supported infrastructure resource changes, not arbitrary application database data mutations performed later.

Explanation

Correct Answer

D:

Infrastructure previews do not replace database migration testing or backups. Application data-plane operations require independent change review.

Incorrect Answers

A:

Complete mode affects resource removal semantics; it does not analyze arbitrary later SQL data migrations.

B:

Resource delta output does not establish that later SQL data-plane scripts are safe.

C:

Validation focuses on deployment inputs and resources, not SQL row-level effects of application migration logic.

Question 336

A department takes a daily managed disk snapshot and assumes it can fail over the application instantly to a different region at any moment. What protection distinction should be made?

  1. Copy a recent managed-disk snapshot into the recovery region and treat that copy as meeting a near-zero recovery time objective without an orchestrated VM, network, or data replication design.
  2. Set a shorter managed-disk snapshot schedule to provide automated regional recovery orchestration.
  3. Disk snapshots preserve selected point-in-time disk states, while regional failover requires an appropriate replicated compute and recovery design.
  4. Create a Recovery Services vault in the secondary region and a running replica appears without protection configuration.

Explanation

Correct Answer

C:

Snapshots are useful recovery artifacts but not a complete orchestrated DR environment. RTO and RPO must be designed separately.

Incorrect Answers

A:

A copied snapshot is a recovery artifact, not a running application. Recovery still requires compatible compute, networking, orchestration, and an RPO/RTO plan; a snapshot alone does not guarantee immediate failover.

B:

More frequent snapshots can improve point-in-time recovery granularity but do not create failover orchestration.

D:

A vault alone does not replicate, start, and route an application to a secondary-region VM.

Question 337

An autoscale policy adds and removes VMSS instances every few minutes as CPU moves slightly above and below a threshold. The team wants to reduce unstable scaling. Which setting should be reviewed?

  1. Set the autoscale minimum and maximum equal to the current count, preventing future scaling.
  2. Adjust the autoscale cooldown, thresholds and evaluation windows to dampen repeated scale actions.
  3. Shorten both scale-in and scale-out cooldowns so that each fluctuation causes an even faster reaction.
  4. Change the autoscale metric from CPU percentage to disk capacity without examining the workload trigger.

Explanation

Correct Answer

B:

Cooldown and hysteresis reduce rapid alternating scale-in and scale-out decisions. Capacity still needs workload-based tuning.

Incorrect Answers

A:

Locking capacity eliminates elasticity rather than improving stability of permitted scale decisions.

C:

Shorter cooldowns can exacerbate oscillation instead of dampening it.

D:

An unrelated metric switch does not resolve overly aggressive thresholds and evaluation windows for the actual CPU pattern.

Question 338

A VMSS automatically scales on CPU but finance requires at least two and no more than twelve workers. What should the administrator configure?

  1. Set the autoscale profile default to two but leave the maximum at its prior value.
  2. Configure an autoscale rule that scales out by one instance but omit the minimum and maximum bounds.
  3. Set the scale-set instance count to twelve manually and disable autoscale entirely.
  4. Set the autoscale profile minimum and maximum instance counts to two and twelve respectively.

Explanation

Correct Answer

D:

Autoscale profiles support capacity bounds independent of the trigger threshold. They prevent scaling outside approved instance ranges.

Incorrect Answers

A:

Default capacity is not an upper bound, so the VMSS could still exceed twelve workers.

B:

A scale action specifies the adjustment, not the authorized capacity limits.

C:

A fixed count enforces only one size; it does not preserve CPU-driven scaling between two and twelve.

Question 339

A VM is being retired, but its managed data disk must be retained for a later forensic restore. Which resource deletion option should the operator review?

  1. Set only the OS disk deletion option to Detach and the setting applies to separately attached data disks.
  2. Delete the VM using its default disk deletion settings, and attempt to change the data disk delete option after the VM disappears.
  3. Configure the data disk's supported delete option to Detach and verify retention before deleting the VM.
  4. Rely on a resource-group CanNotDelete lock as the only retention control while deleting the VM.

Explanation

Correct Answer

C:

Managed disk delete options can determine whether a disk is removed with the VM. Explicit detach/retention settings protect the required evidence.

Incorrect Answers

A:

Disk deletion behavior is configured for each attached disk resource; OS disk retention does not automatically retain the data disk.

B:

If the configured VM deletion behavior removes attached disks, changing disk retention after the deletion cannot recover the removed forensic disk.

D:

A lock that prevents deletion can block the retirement itself; it does not replace choosing the data disk detach behavior for a successful VM deletion.

Question 340

A VM SKU is offered in the region, but deployment fails because the subscription has reached the relevant VM family vCPU quota. What should an operator do?

  1. Create a capacity reservation for the same VM family and it automatically raises the regional vCPU quota.
  2. Assign Contributor on the target resource group to clear a regional VM-family quota exhaustion error.
  3. Change the availability zone while retaining the same exhausted family quota without checking other limits.
  4. Check the applicable regional VM family quota and request an approved increase or choose a supported size within quota.

Explanation

Correct Answer

D:

Service quotas can block deployment even when a SKU is regionally available. Capacity and quota are separate limits requiring appropriate checks.

Incorrect Answers

A:

Capacity reservations address available compute allocation; they do not by themselves increase subscription family vCPU quota.

B:

RBAC can authorize deployment actions but cannot increase a subscription quota already exhausted.

C:

Zone placement alone does not remove regional family vCPU quota constraints.