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 19

Question 180

A file-share administrator creates frequent share snapshots but management needs protection from accidental removal of the storage account itself. What additional design consideration is essential?

  1. Create additional snapshots in the same storage account and treat their number as protection against deletion of the containing account.
  2. Switch the storage account to GRS and use its asynchronous secondary replica as the sole recovery strategy for deliberate account deletion.
  3. Use supported Azure Files backup or independently managed retention protection rather than assuming in-account snapshots survive every account-level loss event.
  4. Enable share soft delete and rely on its retention policy as an independently isolated copy after the entire storage account is removed.

Explanation

Correct Answer

C:

Share snapshots are useful point-in-time versions but have a different isolation and deletion model than independently governed backup recovery points.

Incorrect Answers

A:

More snapshots improve recovery options for file changes but do not place the copies outside the same storage-account failure or deletion boundary.

B:

Storage redundancy protects against supported infrastructure failures; it is not independently governed backup retention for destructive account-level actions.

D:

Share soft delete is scoped to recoverable deleted shares inside the existing storage-account protection boundary. It is not an independent backup of a deleted entire account.

Question 181

A Bicep module accepts a VM size parameter, but only three validated SKUs are approved for deployment. What template-level guard can reject unsupported parameter values?

  1. Put the allowed SKUs in a comment and ARM will enforce the comment.
  2. Use a suitable allowed-values constraint on the parameter and enumerate the approved sizes.
  3. Place the approved SKU list in the @description parameter decorator, and treat the prose description as an enforced ARM constraint.
  4. Set a Bicep @minLength constraint on the SKU string instead of checking exact permitted values.

Explanation

Correct Answer

B:

An allowed-values constraint validates permitted inputs at deployment. It is useful alongside governance controls that enforce broader organizational rules.

Incorrect Answers

A:

Source comments document intent but are not executed by the template engine as parameter constraints.

C:

The description is documentation; enforcement requires an allowed-values constraint or equivalent typed validation.

D:

Minimum string length does not restrict input to the organization’s three approved VM sizes.

Question 182

A storage account must exist before a diagnostic resource can reference its ID in the same Bicep deployment. What usually creates the resource ordering relationship?

  1. Adding a comment that says deploy storage first creates an executable dependency.
  2. Place both resources in the same Bicep module with matching deployment tags but avoid referencing their symbols.
  3. Make the diagnostic resource alphabetically follow the storage account in the template source.
  4. Referencing the resource symbol's properties normally creates an implicit dependency in Bicep.

Explanation

Correct Answer

D:

Bicep can infer dependency relationships from resource references, avoiding unnecessary manual dependsOn entries. Validate the actual deployed resource links.

Incorrect Answers

A:

A comment cannot instruct ARM to await a resource without a real reference or dependency.

B:

Being in the same module or sharing tags does not by itself create an ordering dependency between independent resources.

C:

Source-code declaration order and resource names do not guarantee deployment sequence. Resource references or explicit dependencies provide the relationship.

Question 183

A template contains a policy assignment at management group scope but a pipeline attempts to run it as a normal resource-group deployment without a module scope. What must be corrected?

  1. Keep the deployment at resource-group scope and assign an identically named resource-group policy resource.
  2. Use the appropriate management group deployment scope or supported scoped module with authorized deployment identity.
  3. Set the deployment to subscription scope but the management-group target is inferred from the policy assignment name.
  4. Give the pipeline Contributor on the resource group and role authorizes policy writes at an unrelated management group.

Explanation

Correct Answer

B:

Resource scopes and deployment scopes must align; management-group governance resources cannot be created by pretending they are child resource-group objects.

Incorrect Answers

A:

A resource-group deployment cannot directly substitute for a management-group-scoped policy assignment without the appropriate target scope.

C:

Subscription scope is different from management-group scope. The correct deployment scope must be established explicitly.

D:

A resource-group-scoped Contributor grant does not by itself authorize management-group-scope deployment and policy assignment changes.

Question 184

A deployment what-if output reports repeated changes to provider-generated fields even though the authored Bicep inputs are stable. How should a reviewer interpret this?

  1. Review the specific resource changes and computed-property limitations; do not treat every predicted diff as guaranteed destructive change.
  2. Ignore all changes without inspecting them because what-if can display noise.
  3. Execute the deployment immediately to see whether production will fail.
  4. Remove all resource IDs from the template to suppress every what-if difference.

Explanation

Correct Answer

A:

Provider-calculated values can yield ambiguous change previews. Human review should distinguish real desired-state edits from unsupported or noisy comparisons.

Incorrect Answers

B:

Some changes are real and important; dismissing all output defeats the change review gate.

C:

Running production first makes the preview requirement meaningless and may cause downtime.

D:

Resource identifiers are required for correct scoping and cannot be discarded to silence warnings safely.

Question 185

A team stores production and staging nonsecret settings in separate Bicep parameter files. A reviewer wants to include a database password in each file as a plain string. Which adjustment is appropriate?

  1. Keep public configuration in parameter files but pass secrets using secure parameters and a protected secret source.
  2. Add the secret as a description comment in the template for easier auditing.
  3. Encode the password in base64 and commit the encoded string to public source control.
  4. Put the secret in the parameters file because its extension always encrypts all values.

Explanation

Correct Answer

A:

Parameter files are useful for environment configuration, but plaintext secret values remain sensitive. Protected runtime or Key Vault references prevent unnecessary exposure.

Incorrect Answers

B:

Comments may be visible to source readers and are not a safe credential transport mechanism.

C:

Base64 is reversible encoding, not encryption, and does not protect a committed password.

D:

File format by itself does not guarantee encryption of literal secret content.

Question 186

An operator maintains a stateful scale-set instance for a data migration, and autoscale may remove instances during a quiet period. Which supported VMSS configuration can protect that particular instance from scale-in while work runs?

  1. Mark the whole scale-set model with an Azure resource lock to protect only a selected instance from scale-in.
  2. Increase the minimum instance count without protecting the specific migration instance from selection.
  3. Disable scale-in for every instance indefinitely rather than protecting the one actively being maintained.
  4. Enable supported instance scale-in protection for the specific scale-set instance where the orchestration model supports it.

Explanation

Correct Answer

D:

Instance protection can exclude designated instances from scale-in operations in supported VM Scale Sets. It should be removed after maintenance so capacity can be managed normally.

Incorrect Answers

A:

A management lock on a scale set is not the instance-level scale-in protection setting that prevents automatic removal of the selected VM.

B:

A minimum count can preserve fleet capacity while still allowing a particular instance to be removed. This task needs instance-level protection.

C:

Stopping all scale-in may prevent one removal but is broader than necessary and conflicts with the intended autoscaling behavior.

Question 187

A latency-sensitive database writes transaction logs to a dedicated managed data disk. The application vendor advises avoiding read/write host caching for durable log writes. Which VM disk setting should the operator review?

  1. Set an appropriate disk host caching mode such as None according to the database workload guidance.
  2. Use ReadOnly disk host caching for transaction-log writes to ensure the write path has read cache acceleration.
  3. Enable ReadWrite host caching because it should always improve database transaction-log durability.
  4. Increase the disk tier while keeping a caching mode that conflicts with the database vendor’s durability advice.

Explanation

Correct Answer

A:

Host caching changes the I/O path for managed disks and should match workload durability guidance. Transactional log disks often require different caching from read-heavy data disks.

Incorrect Answers

B:

ReadOnly caching does not match vendor guidance to avoid host caching for durable log writes; None is commonly appropriate for log disks.

C:

ReadWrite caching can affect write durability and is often contraindicated for transaction logs; vendor-specific requirements must govern.

D:

Higher provisioned IOPS do not fix an incorrect host-caching mode for sensitive transaction-log writes.

Question 188

A company deploys short-lived stateless build workers that should boot rapidly and can be recreated after host failure. No state may be retained on the OS disk. Which Azure VM OS disk design could be appropriate when SKU supports it?

  1. Place durable build outputs on the ephemeral OS disk so they survive host reimage and deallocation.
  2. Attach a premium managed OS disk to every worker and it offers the same host-local ephemeral OS characteristics.
  3. Use an ephemeral OS disk on an unsupported VM SKU, and the feature is available regardless of VM size.
  4. Evaluate ephemeral OS disks for the image and VM size, while storing essential state externally.

Explanation

Correct Answer

D:

Ephemeral OS disks can reduce provisioning overhead for replaceable stateless compute, but local OS state is not durable across certain events.

Incorrect Answers

A:

Ephemeral OS disks are designed for recoverable/rebuildable systems; any state that must survive reimage needs durable external storage.

B:

A durable managed OS disk has different lifecycle and placement properties. It can be valid in other designs but not the requested ephemeral worker option.

C:

Ephemeral OS disks depend on compatible image, placement and VM size constraints; they cannot simply be enabled on any SKU.

Question 189

A Linux Azure VM is not completing boot, and operators cannot SSH into it. Which Azure diagnostic should they enable or inspect to capture the early boot screen and serial output?

  1. Use Azure Monitor platform CPU metric charts as the primary record of the Linux boot-screen and serial-console failure text.
  2. Install Azure Monitor Agent inside the failed guest and inspect its newly collected syslog before the operating system can initialize.
  3. Open SSH on the NSG and rely on successful TCP access to capture the failed kernel boot screen.
  4. Use Azure VM boot diagnostics with supported storage settings, and inspect the boot screenshot and serial log.

Explanation

Correct Answer

D:

Boot diagnostics records early guest boot evidence outside a normal SSH session. It can reveal kernel or configuration errors preventing sign-in.

Incorrect Answers

A:

Platform CPU telemetry gives numeric performance measurements, not the serial log or screenshot needed when the guest fails before SSH access is possible.

B:

Guest agents depend on an adequately booted guest environment; they may not start when a kernel or early-boot fault prevents the OS from loading. Boot diagnostics can provide earlier evidence.

C:

Permitting port 22 changes network filtering but cannot restore a failed Linux boot sequence or produce its early serial-console evidence.

Question 190

A VM deployment succeeds but its Custom Script Extension reports failed status. The script downloads from a private endpoint that the VM cannot resolve. What should be investigated?

  1. Increase the VM extension’s execution timeout without fixing DNS resolution of the private download hostname.
  2. Validate the extension's VM-side DNS, network reachability, execution identity and script logs, not only the ARM resource success.
  3. Rerun ARM resource deployment validation alone and it proves guest extension script downloads succeeded.
  4. Move the extension to another resource group while leaving the VM-side DNS and private endpoint access unchanged.

Explanation

Correct Answer

B:

Extension execution occurs in the guest and depends on its network and permissions. A successful VM resource creation does not guarantee the bootstrap task succeeded.

Incorrect Answers

A:

A longer timeout cannot resolve a private endpoint hostname that the guest cannot resolve. VM-side DNS and network access should be checked.

C:

Template validation and resource creation do not prove the extension can resolve and reach its private endpoint in the guest OS.

D:

Resource-group placement does not correct guest DNS or routing for a custom script download.