Skip to content

5 Free AZ-104 Storage Practice Questions

Practise AZ-104 storage with five multi-constraint scenarios from the Pass104 question bank, where every option looks plausible. Every answer includes a worked explanation and Microsoft Learn sources. No account required.

Start with question 1
Work through the set

Four choices per question. Think through your answer, then open the explanation to check it.

5 questions · storage · no timer

What these questions cover

These samples cover selected storage skills in the Microsoft AZ-104 study guide. They are not a complete domain assessment.

Read each scenario, choose an answer, then open the explanation. All questions and answers are free on this page. In the app, the free plan gives you one 6-question exam a day across all domains; domain-targeted practice is a Pro feature.

Question 1 of 5Access tiers and lifecycle management

A general-purpose v2 storage account named finstor1 receives nightly transaction exports as block blobs in the Hot tier. Each export is read frequently during its first 30 days. After that, most exports are never read again, but any export that is requested must be returned within milliseconds. Regulations require each export to be deleted 150 days after creation. You need a lifecycle management policy that minimizes storage costs and never incurs early deletion charges. Which rule should you create?

  1. Move blobs to the Cold tier 30 days after creation, and delete them 150 days after creation
  2. Move blobs to the Cool tier 30 days after creation, and delete them 150 days after creation
  3. Move blobs to the Archive tier 30 days after creation, and delete them 150 days after creation
  4. Move blobs to the Cool tier 30 days after creation, to the Cold tier 90 days after creation, and delete them 150 days after creation
Show answer and explanation

Correct answer: A · Move blobs to the Cold tier 30 days after creation, and delete them 150 days after creation

Cold is an online tier with millisecond time to first byte and lower storage cost than Cool. Its minimum retention is 90 days; moving at day 30 and deleting at day 150 leaves each blob in Cold for about 120 days, so no early deletion charge applies. Hot is only needed while the exports are read frequently.

Why the other choices do not fit

  • B. Cool meets the latency requirement and its 30-day minimum is satisfied, but Cool storage costs more than Cold for 120 days of data that is almost never read.
  • C. Archive has the lowest storage price, but it is offline: a blob must be rehydrated, which takes hours, before it can be read. Deleting it after 120 days in Archive also triggers a prorated early deletion charge for the remaining 60 of the 180 days.
  • D. The stepped rule looks cheaper, but each blob stays in Cold for only 60 days before deletion. Cold is charged for early deletion when a blob leaves it within 90 days, which breaks the no-early-deletion requirement.
Microsoft LearnAccess tiers: minimum retention and early deletion Lifecycle management overview
Question 2 of 5SAS revocation and stored access policies

Four partner applications need read access to blobs in the reports container of storage1. Each partner uses its own SAS, and all SAS tokens are generated by the same automation account. Each partner’s access must remain usable for up to six months, and security must be able to cut off any single partner immediately, without regenerating the storage account keys and without affecting the other three partners. What should you do?

  1. Issue each partner a user delegation SAS, and call Revoke User Delegation Keys when a partner must be cut off
  2. Create one stored access policy per partner on the reports container, and issue each partner a service SAS that references its own policy
  3. Issue each partner an account SAS that references its own stored access policy
  4. Issue each partner an ad hoc service SAS that expires after one hour, and have the automation account reissue it every hour
Show answer and explanation

Correct answer: B · Create one stored access policy per partner on the reports container, and issue each partner a service SAS that references its own policy

A service SAS can reference a stored access policy on the container. Deleting or changing one partner’s policy invalidates only the SAS tokens bound to it, with no key rotation. A container can hold up to five stored access policies, so four partners fit. A policy change can take up to 30 seconds to take effect.

Why the other choices do not fit

  • A. Revoke User Delegation Keys invalidates every user delegation SAS in the account, so all four partners would lose access. A user delegation SAS is also limited to seven days, so it cannot provide six months of access from one token.
  • C. Stored access policies are not supported for account SAS. They work only with service SAS tokens on containers, file shares, queues and tables.
  • D. An ad hoc SAS cannot be revoked before it expires except by rotating the signing key. Stopping reissue leaves the current token valid for up to an hour, so the cut-off is not immediate.
Microsoft LearnDefine a stored access policy Revoke a user delegation SAS
Question 3 of 5Redundancy and secondary reads

A mission-critical application stores data in a Standard general-purpose v2 account named storage5, in a region with availability zones. It must keep reading and writing if one availability zone in the primary region fails. During a region-wide outage it must still be able to read data, without anyone performing an account failover. Which redundancy option should you configure?

  1. GZRS
  2. RA-GZRS
  3. RA-GRS
  4. GRS
Show answer and explanation

Correct answer: B · RA-GZRS

RA-GZRS replicates synchronously across three zones in the primary region, so reads and writes continue through a zone failure. It also replicates to a secondary region that applications can read from at any time through the secondary endpoint, with no failover. Secondary replication is asynchronous, so the latest writes may not be there yet.

Why the other choices do not fit

  • A. GZRS survives a zone failure, but its secondary region cannot be read until a failover is performed.
  • C. RA-GRS allows reads from the secondary without failover, but its primary copies are LRS in a single datacenter. A zone failure affecting that datacenter can interrupt writes in the primary region.
  • D. GRS keeps LRS in the primary region and does not expose the secondary for reads until failover, so it fails both requirements.
Microsoft LearnAzure Storage redundancy Read access to data in the secondary region
Question 4 of 5Azure File Sync topology

Litware has 120 branch offices. Each branch has one Windows Server 2022 file server and its own Azure file share, share1 through share120. Each branch server must use Azure File Sync to cache only its own branch share. All resources are in one Azure subscription. What should you deploy to meet the requirement with the fewest Storage Sync Services?

  1. One Storage Sync Service with 120 sync groups, each containing one cloud endpoint and that branch’s server endpoint
  2. One Storage Sync Service with one sync group containing 120 cloud endpoints and 120 server endpoints
  3. Two Storage Sync Services, each with 60 registered branch servers and their 60 sync groups
  4. One Storage Sync Service per branch, 120 in total, each with a single sync group
Show answer and explanation

Correct answer: C · Two Storage Sync Services, each with 60 registered branch servers and their 60 sync groups

Each share needs its own sync group, because a sync group has exactly one cloud endpoint. A Storage Sync Service supports up to 200 sync groups but only 100 registered servers, and each server can be registered with only one service. 120 servers therefore need two services, each holding the sync groups for its own servers.

Why the other choices do not fit

  • A. 120 sync groups fit within the 200-group limit, but 120 registered servers exceed the hard limit of 100 servers per Storage Sync Service.
  • B. A sync group can contain only one cloud endpoint. One group with 120 shares is not possible, and it would not keep each branch limited to its own share.
  • D. This works technically per branch, but it uses far more services than needed. It also exceeds the hard limit of 100 Storage Sync Services per subscription.
Microsoft LearnAzure File Sync scale targets Plan an Azure File Sync deployment
Question 5 of 5Object replication

storage1 in East US and storage2 in West Europe have blob versioning enabled, and an object replication policy replicates container1 in storage1 to container1 in storage2. Before each update, an application creates a snapshot of the blob it is about to change. During a disaster recovery drill you find that none of the snapshots exist in storage2. You must be able to recover earlier states of the blobs from storage2 with the least administrative effort. What should you do?

  1. Add a second object replication policy between storage1 and storage2 with a rule that includes snapshots
  2. Schedule an AzCopy job that copies the snapshots from storage1 into container1 in storage2
  3. Run an Azure Automation runbook that creates snapshots of the blobs in storage2 on a schedule
  4. Recover earlier states from the previous blob versions that object replication has already copied to storage2
Show answer and explanation

Correct answer: D · Recover earlier states from the previous blob versions that object replication has already copied to storage2

With versioning enabled, every update creates a previous version that holds the pre-update state. Object replication copies the current version and previous versions to the destination, so earlier states are already in storage2. They receive new version IDs there, but the content is preserved. No configuration change is needed.

Why the other choices do not fit

  • A. Object replication never replicates snapshots, whatever the rules say. Only one replication policy is allowed for each source and destination account pair.
  • B. While the replication rule is active, writes to the destination container fail with 409 Conflict. The job would also add ongoing effort for data that versions already cover.
  • C. Scheduled snapshots capture only the state that exists at each run, so updates between runs are missed, and the runbook is extra work to maintain. The replicated versions already hold every earlier state.
Microsoft LearnObject replication: versions and snapshots Blob versioning

Continue your AZ-104 preparation

Review the concepts

Choose your next practice set

Read AZ-104 passing score explained to understand what practice results can tell you and plan your next study session.

Browse all AZ-104 guides and study resources.

Want to focus on storage? Pro lets you choose domains for targeted practice. The free daily exam includes questions across all domains.

These original practice questions are adapted from the Pass104 question bank. They are not actual Microsoft exam questions and do not reproduce or predict exam difficulty or results. The selected answers and explanations include linked Microsoft Learn documentation. The public samples were last reviewed on 26 September 2026. Azure services and exam objectives can change, so use the references for the latest details. Pass104 is independent of Microsoft.