AZ-305 domain - 13% of the exam

Design Business Continuity Solutions

Design Business Continuity Solutions is 13% of the AZ-305 exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

34
Practice questions
2
Objectives
22
Medium
12
Hard

Where people go wrong

  • Swapping the three Azure SQL Database recovery tiers

    The wrong answers swap the recovery tiers of Azure SQL Database. Active geo-replication and failover groups are offered for an in-region zero-data-loss target, and geo-restore is offered for a sub-minute requirement. Each tier has its own scope. Zone redundancy handles an Availability Zone failure with an RTO typically under 30 seconds and an RPO of 0. Failover groups cover a regional outage with an RTO typically under 60 seconds and an RPO of 0 or more, set by replication lag. Geo-restore takes minutes or hours.

    See it in a sample question
  • Pairing Cosmos DB consistency levels with the wrong RPO

    The wrong answers pair a Cosmos DB consistency level with the wrong recovery point. Session consistency is offered as RPO 0, and Strong consistency is offered on accounts with multiple write regions. Only Strong consistency gives an RPO of 0 during a region outage, and it cannot be configured when multiple write regions are enabled. Session, consistent prefix and eventual consistency carry an RPO of less than 15 minutes, while Bounded Staleness is described by K and T values instead.

    See it in a sample question
  • Mixing up planned, unplanned and Microsoft-managed storage failover

    The wrong answers treat every storage account failover as one operation. A customer-managed planned failover converts the account to GRS, keeps geo-redundancy and demotes the original primary to the new secondary. A customer-managed unplanned failover converts the account to LRS in the new primary region and deletes the original primary. Microsoft-managed failover is a third thing: it is reserved for extreme circumstances and cannot be started for an individual storage account, so a design with an RTO of 1 hour cannot depend on it.

    See it in a sample question
  • Misplacing Azure Site Recovery A2A settings and components

    The wrong answers move Azure Site Recovery A2A settings and components to places the documentation does not put them. The cache storage account sits in the source region, not the target. Crash-consistent recovery points are created every five minutes and that interval cannot be modified; the replication policy sets only the app-consistent snapshot frequency and recovery-point retention, and the app-consistent default is zero hours, meaning disabled. Configuration servers belong to the VMware-to-Azure path, while A2A uses the Mobility service extension on the source VM.

    See it in a sample question
  • Crediting zones and region pairs with cross-region failover

    The wrong answers credit availability zones and paired regions with protection they do not provide. Zones have independent power, cooling and networking but do not protect against a full-region outage. Region pairing staggers planned updates and sets a recovery order, yet deploying to a pair does not automatically provide high availability, disaster recovery or failover, and RBAC and policy are not mirrored. Mission-critical workloads combine multiple zones with multiple regions.

    See it in a sample question

Objectives in this domain

Sample questions from this domain

Free sampleDesign Business Continuity Solutionsmedium

An Azure SQL Database is part of a failover group with a customer-managed failover policy to meet a cross-region DR target, tracked under the WAF Reliability pillar. Which RTO and RPO line up with this configuration in the documented HADR table?

  • ARTO typically less than 30 seconds; RPO always 0 regardless of cross-region replication lag observed.
  • BRTO typically minutes to hours; RPO typically minutes to hours, dependent on the storage replication path.
  • CRTO typically less than 60 seconds; RPO equal to or greater than 0 depending on replication lag. Correct
  • DRTO of 5 minutes guaranteed; RPO of 5 minutes guaranteed by Premium-tier Azure SQL failover groups.
Failover groups target <60 seconds RTO with RPO that depends on replication lag. Azure SQL HADR documentation lists DR using failover groups with customer-managed failover or active geo-replication as RTO typically less than 60 seconds and RPO equal to or greater than 0, with the lag depending on unreplicated data changes at the time of the disruption.

Why A is wrong: 30-second RTO with RPO 0 is the in-region zone-redundant tier; failover groups across regions document <60 seconds RTO and an RPO that depends on replication lag.

Why B is wrong: Minutes-to-hours RTO and RPO describe geo-restore from geo-redundant backups, not failover groups with continuous replication.

Why C is correct: Correct. Azure SQL HADR documentation lists DR using failover groups with customer-managed failover or active geo-replication as RTO typically less than 60 seconds and RPO equal to or greater than 0, with the lag depending on unreplicated data changes at the time of the disruption.

Why D is wrong: Azure SQL HADR documentation does not list a Premium-tier 5/5 guarantee; the documented failover-group targets are <60 seconds RTO and RPO 0+.

Free sampleDesign Business Continuity Solutionshard

A Cosmos DB for NoSQL account is configured with Session consistency, multiple read regions, and a single write region. Following the WAF Reliability pillar, you must report the documented RPO during a region outage. Which value matches the documentation?

  • ARPO is 0 during a region outage because Session consistency replicates synchronously across regions.
  • BRPO equals K and T values where K is the version count and T is the time interval bound.
  • CRPO is 5 minutes during a region outage, matching the ASR crash-consistent recovery-point cadence.
  • DRPO is less than 15 minutes during a documented region outage for Session consistency. Correct
Cosmos Session/Prefix/Eventual RPO during a region outage is documented as less than 15 minutes. Cosmos DB reliability documentation lists RPO during a region outage as less than 15 minutes for Session, consistent prefix, and eventual consistency. Bounded staleness uses K and T values, and Strong consistency is 0. Session falls in the less-than-15-minute bucket.

Why A is wrong: Session consistency replication is asynchronous across regions; RPO 0 is only achieved with Strong consistency on a multi-region single-write account.

Why B is wrong: K and T describe Bounded Staleness, not Session; Session RPO during a region outage is less than 15 minutes.

Why C is wrong: Cosmos RPO is not derived from ASR cadence; the documented Session RPO is less than 15 minutes.

Why D is correct: Correct. Cosmos DB reliability documentation lists RPO during a region outage as less than 15 minutes for Session, consistent prefix, and eventual consistency.

Free sampleDesign Business Continuity Solutionshard

Your AZ-305 storage DR design uses RA-GZRS and you trigger a customer-managed unplanned failover during a regional outage to meet a WAF Reliability pillar RTO of 1 hour. After the failover completes, what is the documented redundancy state of the storage account in the new primary region?

  • AThe storage account remains RA-GZRS in the new primary region; geo-redundancy is preserved with no further action.
  • BThe storage account is converted to GRS in the new primary region and automatically re-paired to a different secondary region.
  • CThe storage account is converted to ZRS in the new primary region to retain zone-level resilience.
  • DThe storage account is automatically converted to locally redundant storage (LRS) in the new primary region; geo-redundancy is lost. Correct
Customer-managed unplanned failover converts the storage account to LRS; geo-redundancy must be re-enabled manually. Storage disaster recovery documentation states that initiating a customer-managed unplanned failover automatically converts the storage account to locally redundant storage (LRS) within the new primary region and deletes the original primary. Geo-redundancy is lost; re-enabling GRS or RA-GRS incurs a re-replication charge.

Why A is wrong: Customer-managed unplanned failover specifically converts the new primary to LRS; this is documented as the unplanned failover behaviour, regardless of starting tier.

Why B is wrong: Customer-managed planned failover preserves geo-redundancy; the unplanned variant does not, and it lands on LRS rather than GRS.

Why C is wrong: Documented unplanned-failover behaviour is conversion to LRS, not ZRS; re-enabling zone-redundancy requires explicit reconfiguration.

Why D is correct: Correct. Storage disaster recovery documentation states that initiating a customer-managed unplanned failover automatically converts the storage account to locally redundant storage (LRS) within the new primary region and deletes the original primary.

Free sampleDesign Business Continuity Solutionshard

Your AZ-305 ASR A2A design replicates VMs from Australia East to Australia Southeast with a target RPO of 5 minutes and RTO of 30 minutes, defended through the WAF Reliability pillar. The networking team wants to host the cache storage account in the secondary region to save inter-region egress. Where does ASR require the cache storage account to live?

  • AIn the source region, because writes are immediately transferred to the cache before being processed and sent to the target. Correct
  • BIn the target region, because the cache stages data for the replica managed disks closer to the failover footprint.
  • CEither region, provided the cache and target storage accounts share the same redundancy tier.
  • DIn a third Azure region geographically between source and target to minimize round-trip latency on the cache write.
ASR cache storage lives in the source region by design, regardless of the target region. The A2A architecture article specifies a cache storage account in the source network. During replication, VM disk writes are sent to the cache first, then processed and replicated to target storage or replica managed disks. The cache is what keeps production-application impact minimal.

Why A is correct: Correct. The A2A architecture article specifies a cache storage account in the source network.

Why B is wrong: ASR puts the cache in the source region; the replica managed disks (with the -ASRReplica suffix) live in the target region but the cache does not.

Why C is wrong: The architecture explicitly requires the cache in the source region; it is not a choice based on redundancy parity.

Why D is wrong: No third region is documented; the cache stays local to the source network for minimal impact on production applications.

Free sampleDesign Business Continuity Solutionsmedium

Your AZ-305 design uses Australia East and Australia Southeast for cross-region redundancy with an RPO of 15 minutes and RTO of 1 hour, audited against the WAF Reliability pillar. Which benefit of region pairing is most relevant to the patching strategy?

  • AAzure synchronously replicates resources between paired regions, so updates land at exactly the same moment in both.
  • BAzure mirrors all RBAC and policy changes from the primary to the paired region within 5 minutes during patching cycles.
  • CAzure strives to stagger planned system updates across region pairs so the pair is not updated simultaneously. Correct
  • DAzure shifts customer workloads automatically to the paired region while updates roll out, then shifts them back.
Paired regions get sequentially staggered planned updates, not simultaneous ones. The region-pairs documentation states that Azure strives to stagger planned system updates across region pairs so that paired regions are not updated simultaneously. This protects designs that rely on the pair as their resilience pattern from a faulty update hitting both halves at once.

Why A is wrong: Paired regions do not provide synchronous replication; deployment to a region pair does not automatically replicate resources.

Why B is wrong: There is no documented automatic RBAC/policy mirroring between paired regions; that has to be designed in.

Why C is correct: Correct. The region-pairs documentation states that Azure strives to stagger planned system updates across region pairs so that paired regions are not updated simultaneously.

Why D is wrong: Microsoft-managed failover between paired regions is only used in catastrophic situations, not as a routine patching mechanism.

Other domains in this exam

See also the AZ-305 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by Microsoft. Original, blueprint-aligned practice material only.