Azure Storage redundancy practice questions
Fifteen free Azure Storage practice questions with explained answers, written as short scenarios about choosing and operating the redundancy option of a storage account. Redundancy decides how many copies of the data exist and where they sit, from LRS in one datacenter to RA-GZRS across zones and regions. The topic is tested in AZ-900, AZ-104 and AZ-305.
How to use these questions
Answer each question before opening the explanation, and commit to one letter. Underline the constraint word in the stem, such as “datacenter”, “region”, “read during the outage” or “lowest cost”, because it separates neighbouring options. Read why each wrong option fails, since the same distractors return on real exams. The guide Azure Storage redundancy: LRS to RA-GZRS holds the decision table behind the answers.
Warm-up: choosing a redundancy option
Question 1. A storage account holds thumbnail images that an application regenerates from the original photos within minutes. The business wants the lowest possible storage cost and accepts a rebuild if a datacenter is lost. Which redundancy option should be chosen?
- A. LRS
- B. ZRS
- C. GRS
- D. RA-GRS
Show answer
Answer: A
LRS keeps three copies in a single datacenter, is the cheapest option and protects against drive, server and rack failures. That fits data that can be rebuilt easily. ZRS, GRS and RA-GRS all cost more and add protection (zone or region loss) that the requirement does not ask for.
Question 2. Invoice PDFs must stay available if one datacenter in the primary region fails. For governance reasons the data must not be replicated outside that region. Which redundancy option meets both requirements?
- A. LRS
- B. ZRS
- C. GRS
- D. GZRS
Show answer
Answer: B
ZRS copies the data synchronously across three availability zones in one region, so it survives the loss of a zone and has no copy outside the region. LRS sits in one datacenter and does not survive its loss. GRS and GZRS both replicate to a secondary region, which the governance rule forbids.
Question 3. A status dashboard must keep reading blobs during a complete outage of the primary region, without an administrator starting a failover first. Zone redundancy in the primary region is not required, and the cost should be as low as possible. Which option fits?
- A. GRS
- B. ZRS
- C. RA-GRS
- D. RA-GZRS
Show answer
Answer: C
Only the read-access variants expose the secondary region for reads before a failover, and RA-GRS is cheaper than RA-GZRS because the primary region uses LRS. GRS keeps the secondary unreadable until a failover. ZRS has no second region. RA-GZRS also works but adds zone redundancy that was not requested, so it is not the lowest cost.
Question 4. Which statement correctly describes how a write is replicated in a storage account that uses GRS?
- A. Synchronously to both regions, so no data can ever be lost
- B. Asynchronously within the primary region and synchronously to the secondary region
- C. Across availability zones in both regions
- D. Synchronously within the primary region, then asynchronously to the secondary region
Show answer
Answer: D
With GRS the write is committed synchronously to the primary region by using LRS, and then copied asynchronously to the secondary region, where it is again stored by using LRS. It is not synchronous to the secondary region, which is why recent writes can be lost in a disaster. It is not asynchronous inside the primary region. The copies are not spread over zones in the secondary region, which always uses LRS.
Question 5. One storage account holds a blob container for archives that needs GZRS and an Azure Files share for scratch data that needs only LRS. Which approach meets both needs at the correct redundancy?
- A. Use two storage accounts, one with GZRS and one with LRS
- B. Use one GZRS account and set LRS on the file share
- C. Use one LRS account and set GZRS on the container
- D. Use one account and a lifecycle rule to switch redundancy per container
Show answer
Answer: A
The redundancy setting belongs to the storage account and applies to every service in it, so resources with different redundancy needs belong in separate accounts. Setting GZRS on the one account would also apply it to the scratch share and cost more. A per-container redundancy setting does not exist, and a lifecycle rule moves blobs between access tiers, not between redundancy options.
Scenarios: geo-replication, protection and limits
Question 6. A company runs a GRS storage account in West Europe. The compliance team asks for the secondary copy to be placed in a specific region of its own choice. What is the correct answer?
- A. Select the secondary region in the redundancy blade when creating the account
- B. This is not possible, because the secondary region is fixed by the primary region
- C. Open a support request, which can reassign the secondary region
- D. Switch to RA-GRS, which allows a custom secondary region
Show answer
Answer: B
The secondary region is determined by the primary region and cannot be changed. The portal offers no secondary-region picker for GRS, so choosing the primary region effectively chooses the pair. A support request does not change this, switching to RA-GRS adds read access but keeps the same pair, and customer-managed failover swaps roles between the two regions rather than selecting a new one.
Question 7. A faulty application release overwrote thousands of blobs in a GZRS account with corrupt content. The team needs to return to the earlier content of each blob. Which feature must have been enabled to make that possible?
- A. Upgrading the account from GZRS to RA-GZRS
- B. A customer-managed failover to the secondary region
- C. Blob versioning
- D. Container soft delete
Show answer
Answer: C
Blob versioning keeps earlier versions when a blob is overwritten, so a previous version can be restored. Redundancy, including RA-GZRS, copies the overwrite to every replica, so changing it restores nothing. A failover promotes the secondary, which holds the same overwritten data. Container soft delete protects deleted containers, not overwritten blobs.
Question 8. A team wants to move rarely used blobs to the archive access tier to cut cost. The accounts below are all general-purpose v2. Which account redundancy supports the archive tier?
- A. ZRS
- B. GZRS
- C. RA-GZRS
- D. GRS
Show answer
Answer: D
The archive tier is supported for LRS, GRS and RA-GRS accounts only. It is not supported for ZRS, GZRS or RA-GZRS accounts. Of the options, only GRS appears in the supported list.
Question 9. An architect plans to use Azure Files for SMB shares and lists redundancy options for the design review. Which option cannot be used for Azure Files?
- A. RA-GZRS
- B. GZRS
- C. ZRS
- D. GRS
Show answer
Answer: A
Azure Files does not support the read-access variants RA-GRS and RA-GZRS. It supports LRS, ZRS, GRS and GZRS for standard shares. So RA-GZRS is the option that cannot be used, while GZRS, ZRS and GRS are all valid.
Question 10. Virtual machines in one region use Premium SSD managed disks. The disks must survive the failure of one availability zone, and no geo-replication of the disks is wanted. Which redundancy option applies to the managed disks?
- A. GRS
- B. GZRS
- C. ZRS
- D. RA-GRS
Show answer
Answer: C
Managed disks support only LRS and ZRS, and ZRS disks keep their data across zones in the region. GRS, GZRS and the read-access variants are not available for managed disks. LRS would not survive a zone failure.
Edge cases and traps: failover and data loss
Question 11. A GRS account in the primary region is unavailable during a regional disaster. An administrator starts a customer-managed unplanned failover. What is the state of the account after the failover completes?
- A. LRS in the new primary region, with geo-redundancy lost
- B. GRS, with the original primary region now acting as the secondary
- C. GRS, unchanged, because the pair is fixed
- D. RA-GRS, because the new primary is readable from both regions
Show answer
Answer: A
After an unplanned failover the secondary becomes the new primary and the account is converted to LRS in that region, so geo-redundancy is lost until GRS is configured again. The data copy in the original primary region is deleted. Keeping GRS with the old primary demoted to secondary describes a planned failover. RA-GRS is not configured by a failover.
Question 12. The Last Sync Time of a GRS account reads 09:40. At 10:05 the primary region fails and the administrator performs a customer-managed unplanned failover. Which statement is correct?
- A. Nothing is lost, because geo-replication is synchronous
- B. Only data older than 24 hours is lost
- C. Nothing is lost, because the secondary region uses LRS
- D. Data written before 09:40 is available, and later writes may be lost
Show answer
Answer: D
Data and metadata written before the Last Sync Time are available on the secondary. Data written after it may not have been copied yet and can be lost, because replication to the secondary is asynchronous. Replication is not synchronous, so the claim that nothing is lost is wrong. The local redundancy inside the secondary region protects against hardware failure, not against missing replication. The loss is not limited to data older than a day.
Question 13. An RA-GRS account named contosodata serves blobs from contosodata.blob.core.windows.net. During a primary region outage the application must read the same blobs from the secondary region. Which endpoint should it use?
- A. contosodata.blob.core.windows.net, retried until it responds
- B. contosodata-secondary.blob.core.windows.net
- C. A second account named contosodata2 in the paired region
- D. contosodata.secondary.blob.core.windows.net
Show answer
Answer: B
The secondary endpoint adds the suffix -secondary to the account name, here contosodata-secondary.blob.core.windows.net, and uses the same account keys. A separate account in the paired region does not exist, because the secondary is not a second account. The primary endpoint is unavailable during the outage. Reading from the secondary needs no redirect from a Front Door.
Question 14. A team creates a premium block blob storage account for low-latency workloads and wants GRS so the data survives a regional outage. What is the outcome?
- A. GRS can be selected, but without read access to the secondary
- B. GZRS can be selected instead, which is supported for premium accounts
- C. Geo-redundancy is not available, so data must be copied to another region with a tool such as AzCopy
- D. GRS is available after enabling a customer-managed failover
Show answer
Answer: C
Premium block blob accounts support LRS and ZRS only, and they do not support geo-redundancy. GRS therefore cannot be selected, and a customer-managed failover is not possible for such accounts. Copying the data to another account in a different region with AzCopy is the documented alternative. RA-GZRS and GZRS are geo-redundant options and are equally unavailable.
Question 15. A web app must keep serving read requests from the secondary region during a regional outage without anyone starting a failover, and the primary region must also tolerate the loss of one availability zone. Which is the lowest-cost option that meets both requirements?
- A. GZRS
- B. RA-GRS
- C. GRS
- D. RA-GZRS
Show answer
Answer: D
Zone resilience in the primary region requires ZRS there, which means GZRS or RA-GZRS. Reading from the secondary before a failover requires a read-access variant. Only RA-GZRS satisfies both. GZRS has the zone resilience but the secondary stays unreadable until a failover. RA-GRS is readable but its primary region uses LRS. GRS has neither feature in the way the stem needs.
Where to go next
Review the theory in Azure Storage redundancy: LRS to RA-GZRS, then continue with Azure RBAC practice questions and NSG and VNet practice questions. The free 20-question AZ-104 practice test mixes storage with the rest of the exam, and multiple-choice exam tips cover how to handle the constraint words.