Amazon AWS Certified Solutions Architect - Associate SAA-C03 Practice Test 36

Question 350

A relational database needs automatic availability after a primary instance/AZ outage, but analytic read scaling is not required. Which RDS deployment meets that requirement?

An ordinary read replica with no controlled promotion or failover procedure.
Add an RDS snapshot policy with hourly automated backups and assume those restore points replace automatic failover during an AZ outage.
A larger single-AZ database instance without standby.
An RDS Multi-AZ DB instance deployment with a synchronous standby and automated failover.

Explanation

Correct Answer

A Multi-AZ DB instance standby protects availability; it is not a read-only application endpoint.

Incorrect Answers

Asynchronous replicas are not identical automatic Multi-AZ HA standbys.

Backups support historical restore but require provisioning and a controlled cutover, rather than automatic standby promotion in a traditional Multi-AZ DB instance.

More resources do not remove the AZ single point of failure.

Question 351

An application needs automated RDS database failover plus additional SQL read throughput from two read-capable standby instances. Which supported Multi-AZ topology is intended?

Use the conventional Multi-AZ DB instance standby as the live reporting reader.
Create a traditional Multi-AZ DB instance with one passive standby and assume it accepts analytics reads as one of the two required readers.
Use an RDS Multi-AZ DB cluster with writer endpoint and reader endpoint supported for the selected engine/configuration.
Run all read and write traffic through a Route 53 TXT record.

Explanation

Correct Answer

The cluster variant has reader instances that both serve reads and can become failover targets, unlike an ordinary single-standby DB instance deployment.

Incorrect Answers

That standby cannot normally serve read-only app traffic.

The passive standby in a traditional DB-instance deployment is not available to application queries, unlike a supported Multi-AZ DB cluster with two readers.

DNS TXT is not a DB connection endpoint and does not create readers.

Question 352

An Aurora cluster has a writer and three reader replicas. The team wants an explicit preference among healthy promotion candidates during writer failure. What control applies?

Assign a higher Aurora reader instance class to the preferred writer candidate but leave all candidates at default equal promotion tiers, assuming size alone guarantees the selected priority.
Configure supported Aurora failover priorities for reader instances and validate automatic writer promotion behavior.
Assign every reader a public IP and let random clients pick a primary.
Delete all readers to force fast failover.

Explanation

Correct Answer

Aurora uses promotion tiers and other eligibility criteria to choose a replacement writer.

Incorrect Answers

Sizing affects eligibility and performance; the configured promotion tier determines preference among eligible readers. With equal priorities Aurora applies other selection criteria.

Client routing does not replace Aurora-managed cluster writer promotion.

Removing candidates reduces availability.

Question 353

Reporting queries should not slow an RDS transactional primary, but production also requires its existing Multi-AZ standby. Which arrangement is appropriate?

Send reporting directly to the conventional Multi-AZ standby.
Turn off Multi-AZ because adding a read replica always gives identical automatic failover.
Retain Multi-AZ failover and add an appropriate read replica for reporting, accounting for replication lag.
Route reporting to an S3 static website instead of SQL.

Explanation

Correct Answer

The primary/standby protects availability; separate asynchronous read replicas offload eligible read traffic.

Incorrect Answers

A normal RDS Multi-AZ DB instance standby is not readable by applications.

A read replica does not replace the standard Multi-AZ availability contract.

Static HTTP hosting cannot run arbitrary reporting SQL on RDS.

Question 354

A customer portal runs in two AWS Regions. DNS should normally direct users to the primary but switch to the secondary when an application health check detects primary failure. Which policy is appropriate?

Use a fixed weighted 100:0 policy without monitoring any endpoint health.
Use one CloudFront CNAME with no origin/region failure design.
Use Route 53 failover records with relevant health checks for primary and secondary targets.
Use Route 53 weighted records set to 100 and 0 without health checks, expecting the inactive regional destination to be selected automatically when the primary fails.

Explanation

Correct Answer

Failover routing selects the secondary when the configured primary health becomes unhealthy; consider DNS cache behavior.

Incorrect Answers

Static weights alone do not implement failover health assessment.

A CNAME by itself is not health-aware regional failover.

Fixed weights are not a health-based primary/secondary selection policy. Use a failover policy with application-relevant health checks and account for DNS caching.

Question 355

A team is performing a controlled regional canary rollout and wants 5% of DNS answers to use the new region even while both regions are healthy. Which Route 53 routing policy best expresses that?

Use weighted routing records with appropriate 95:5 weights and any necessary health checks.
Use strict geolocation routing to force every region to receive exactly 5%.
Use DNS failover records and assume primary health produces a 95:5 split.
Choose Route 53 latency-based routing and assume it returns precisely 5% of global DNS answers from the new Region regardless of resolver geography.

Explanation

Correct Answer

Weighted policy splits eligible DNS answers by configured proportions; it is different from geography-based selection.

Incorrect Answers

Geolocation selects by geographic rule, not a global percentage.

Failover is primarily an active/standby decision, not weighted canary distribution.

Latency-based routing selects the lowest-latency eligible target for the querying resolver rather than enforcing an explicit 95:5 distribution.

Question 356

A media site has a primary HTTP origin and a backup HTTP origin. It wants CloudFront to switch for supported error/status outcomes without rewriting clients. What CloudFront feature can help?

Set the viewer cache TTL to zero and assume an origin group appears automatically.
Configure an origin group with primary/secondary members and supported failover status codes and request-method limitations.
Route POST write transactions through CloudFront origin failover and assume automatic replay is always supported.
Use a private S3 bucket name as proof that HTTP failover is configured.

Explanation

Correct Answer

CloudFront origin failover can retry eligible requests to the secondary on specified primary failures.

Incorrect Answers

Caching settings alone do not configure failover destinations.

Origin failover has method restrictions and must not be treated as universal write replay.

An origin name does not provide a secondary origin group and trigger rules.

Question 357

A multinational retail system needs local DynamoDB read/write access in multiple Regions, and is able to tolerate brief remote replication delay. Which design meets that multi-region write requirement?

Use DynamoDB global tables configured for MREC, with application writes directed to the local regional replica and conflict policy understood.
Use one single-Region table and claim strongly consistent reads across continents without network calls.
Use RDS read replicas and call them active DynamoDB write replicas.
Use CloudFront static caching to make DynamoDB writes local and conflict-free.

Explanation

Correct Answer

MREC replicates asynchronously and supports active regional writes with last-writer-wins conflict handling.

Incorrect Answers

A regional table provides no local replica in other Regions.

Relational RDS replicas do not implement DynamoDB global tables.

A CDN is not a multi-region database writer.

Question 358

A critical Aurora system needs low-latency reads in a second Region and a managed cross-region recovery option. Which Aurora capability matches the database tier?

Enable only Multi-AZ replicas within the primary Region and call it cross-region recovery.
Use a Route 53 weighted record without provisioning a second database.
Use Aurora Global Database with a suitable secondary Region and tested regional switchover/failover procedure.
Treat an hourly S3 CSV export as a low-lag global SQL reader.

Explanation

Correct Answer

Global Database uses dedicated cross-region replication for read locality and managed disaster recovery workflows.

Incorrect Answers

Multi-AZ availability addresses zone failures, not regional loss.

DNS alone cannot create data in another Region.

Snapshots do not offer live query replication or equivalent RPO.

Question 359

The compliance archive is stored in an S3 bucket in one Region. Policy demands replicated copies in a geographically separated Region with defined eligible objects. What S3 facility helps?

Use S3 Same-Region Replication as proof of a remote regional copy.
Enable S3 bucket versioning in the source and destination but never configure a replication rule, expecting those historical object versions to appear in the remote Region.
Assume S3 versioning automatically replicates objects across Regions.
Configure S3 Cross-Region Replication with versioning, suitable IAM role and destination encryption permissions.

Explanation

Correct Answer

Asynchronous CRR copies qualifying source versions to a destination Region and must be tested against the RPO.

Incorrect Answers

Same-Region replication does not survive loss of that entire Region.

Versioning enables separate revisions within each bucket; S3 Cross-Region Replication requires an explicit rule, supported IAM role and suitable encryption access.

Versioning retains object versions locally; replication is a separate rule.

Question 360

An analyst overwrote an important S3 CSV, but the bucket needs an easy recovery path to an earlier object revision. Which property must have been enabled before overwrite?

S3 Transfer Acceleration, which keeps every previous file version.
S3 Versioning, followed by retrieval/restore of the previous object version.
Configure same-Region S3 replication only after an object overwrite, expecting the destination to reconstruct every preconfiguration revision automatically.
A Route 53 failover alias for the S3 bucket.

Explanation

Correct Answer

Versioning preserves historical object versions and delete markers, subject to lifecycle and object-lock conditions.

Incorrect Answers

Transfer Acceleration speeds certain transfers but is not version retention.

Replication typically covers eligible new writes after the rule is enabled. The lost earlier version requires it to have existed under versioning or another prior backup.

DNS routing cannot reconstruct an earlier object body.

Pass Your Amazon AWS Certified Solutions Architect - Associate SAA-C03 Exam Easy!

Question 350

Explanation

Correct Answer

Incorrect Answers

Question 351

Explanation

Correct Answer

Incorrect Answers

Question 352

Explanation

Correct Answer

Incorrect Answers

Question 353

Explanation

Correct Answer

Incorrect Answers

Question 354

Explanation

Correct Answer

Incorrect Answers

Question 355

Explanation

Correct Answer

Incorrect Answers

Question 356

Explanation

Correct Answer

Incorrect Answers

Question 357

Explanation

Correct Answer

Incorrect Answers

Question 358

Explanation

Correct Answer

Incorrect Answers

Question 359

Explanation

Correct Answer

Incorrect Answers

Question 360

Explanation

Correct Answer

Incorrect Answers

AWS Certified Solutions Architect - Associate SAA-C03 Practice Tests