Unit 1: Cloud Fundamentals and Security - Practice Quiz

1 Which AWS service centrally manages backup policies across supported AWS resources?

2 Which Amazon S3 feature automatically copies objects from one AWS Region to another?

3 Which AWS service can direct users to a healthy disaster recovery endpoint through DNS failover?

4 Which AWS service uses state machines to coordinate multiple steps in a recovery workflow?

5 What is the main activity performed before a disaster in the Backup and Restore pattern?

6 Which disaster recovery pattern usually has the lowest ongoing infrastructure cost?

7 What must happen to backed-up data before an application can fully resume after a disaster?

8 Compared with the other listed recovery patterns, Backup and Restore generally has which recovery characteristic?

9 In a Pilot Light architecture, which part of the workload typically remains running in the recovery Region?

10 What usually happens to application servers during Pilot Light recovery?

11 Which statement best describes the recovery environment in the Pilot Light pattern?

12 Pilot Light generally provides faster recovery than which pattern?

13 What is maintained in the recovery Region under a Warm Standby pattern?

14 What is commonly required after failing over to a Warm Standby environment?

15 How does Warm Standby differ from Pilot Light?

16 Compared with Multi-Site Active-Active, Warm Standby usually has which cost characteristic?

17 In a Multi-Site Active-Active configuration, how many sites normally serve production traffic?

18 Which recovery pattern generally provides the shortest recovery time?

19 What is a key data requirement for an Active-Active architecture?

20 What is a common trade-off of Multi-Site Active-Active disaster recovery?

21 A company stores critical objects in an Amazon S3 bucket in one AWS Region. It requires copies to be created automatically in a second Region so that data remains available after a regional disruption. Which solution best meets this requirement?

22 An operations team must coordinate a disaster recovery workflow that restores a database, starts application instances, validates health, and updates DNS. The workflow must track each stage and retry failed tasks. Which AWS service is most suitable for the orchestration?

23 A company wants Route 53 to direct users to a disaster recovery endpoint only when the primary endpoint becomes unhealthy. Which routing policy should it configure?

24 A company needs centrally managed, policy-based backups of Amazon EBS volumes, Amazon RDS databases, and Amazon EFS file systems. It also needs cross-Region backup copies. Which service should it use?

25 A small internal application can tolerate an RTO of 12 hours and an RPO of 24 hours. The company wants the lowest ongoing disaster recovery cost. Which recovery pattern is the best fit?

26 A database is backed up every 6 hours with no transaction-log replication between backups. If the primary Region fails just before the next backup, what is the approximate worst-case RPO?

27 A team can restore its database quickly, but recreating VPCs, load balancers, and application servers manually causes recovery delays and configuration errors. Which improvement most directly reduces this problem?

28 A company copies database backups to a second Region but has never attempted a restore there. Which action provides the strongest evidence that its recovery objective is achievable?

29 A company keeps a continuously replicated database and essential identity services running in its recovery Region. Application servers are created only after a disaster is declared. Which recovery pattern does this describe?

30 During a Pilot Light failover, the replicated database is ready, but the recovery application tier has not yet been deployed. Which action should normally occur before directing production traffic to the recovery Region?

31 A Pilot Light design takes too long to recover because large Amazon EC2 instances must be configured manually after every failover. Which change would most directly reduce the RTO while preserving the Pilot Light approach?

32 Which workload component is the best candidate to remain continuously active in the recovery Region of a Pilot Light architecture?

33 A retailer runs a complete but reduced-capacity copy of its application in a second Region. During a disaster, it scales out that copy and redirects all traffic to it. Which recovery pattern is being used?

34 A Warm Standby environment has enough capacity for only 20% of normal demand. What should the failover runbook do before sending all production traffic to that environment?

35 A business is comparing Pilot Light with Warm Standby. It can accept higher ongoing cost in exchange for a shorter RTO. Why is Warm Standby generally more suitable?

36 A team periodically sends a small percentage of synthetic transactions to its Warm Standby environment. What is the primary recovery benefit of this practice?

37 An application serves users concurrently from two AWS Regions. Route 53 directs requests to healthy endpoints based on latency. Which recovery pattern does this represent?

38 A company is designing an Active-Active shopping cart service across two Regions. Users may update the same cart through either Region. Which issue requires the most careful design?

39 Two active Regions each handle 50% of an application's normal traffic. Either Region must handle all traffic if the other fails. What capacity principle should guide the design?

40 An Active-Active application stores user session state only in the memory of regional web servers. After Route 53 moves a user to another healthy Region, the user is unexpectedly signed out. Which design change best addresses this issue?

41 A company replicates an Amazon Aurora database to a secondary Region using an Aurora global database. During a regional failure, the secondary cluster is promoted successfully, but application recovery still exceeds the RTO because clients continue using the original database endpoint. Which design most directly resolves this orchestration gap?

42 A regulated workload requires immutable, logically isolated recovery points that remain available even if the production AWS account is compromised by an administrator. Which AWS-native design best satisfies this requirement?

43 An application writes an order to Amazon DynamoDB and then asynchronously publishes an event. During a regional disaster, DynamoDB global tables preserve the order, but some events are missing in the recovery Region. Which redesign most effectively preserves the order-to-event relationship across failover?

44 A disaster recovery workflow uses AWS Step Functions to promote a database, scale compute, update DNS, and run validation. Operators may retry the workflow after a timeout whose completion state is unknown. Which property is most important for preventing retries from corrupting the recovery state?

45 A 40 TiB application is rebuilt in a recovery Region from nightly snapshots. Snapshot creation meets an 8-hour RPO, but restoring storage and rehydrating application caches takes 14 hours against a 6-hour RTO. Which change addresses the violated objective without changing the backup interval?

46 A database receives writes at an approximately uniform rate. Full backups occur every 24 hours and transaction logs are archived every 15 minutes. After a disaster, the latest usable log archive is timestamped 10:45, while the failure occurred at 10:57. Ignoring clock skew, what is the effective data-loss window?

47 Infrastructure is recreated from AWS CloudFormation during a Backup and Restore recovery. The stack consistently fails because an S3 bucket name is globally unique and remains owned by the unavailable primary account configuration. Which design best improves deterministic recovery?

48 A company copies encrypted Amazon RDS snapshots to a recovery Region. During a disaster exercise, the snapshots are visible but cannot be restored because the original customer-managed KMS key is unavailable in that Region. Which preparation would have prevented this failure?

49 A Pilot Light environment keeps an asynchronously replicated database running in the recovery Region, while application servers are created only after a disaster. A test shows that database promotion takes 4 minutes, infrastructure deployment takes 11 minutes, instance bootstrapping takes 9 minutes, and validation takes 6 minutes. If the stages are strictly sequential, what RTO can the design realistically claim?

50 A Pilot Light design stores current application data in the recovery Region but creates the VPC, load balancers, IAM roles, and compute fleet only during failover. Which risk most directly threatens its RTO despite successful data replication?

51 A Pilot Light architecture uses an Amazon RDS read replica in another Region. After promotion during a regional outage, the application writes to the promoted database. The original primary later returns. Which failback strategy best avoids silently discarding recovery-Region writes?

52 A company claims a 20-minute RTO for its Pilot Light design because its Auto Scaling group can launch the required instances in 12 minutes. During a test, newly launched instances cannot retrieve deployment artifacts because the artifact bucket exists only in the failed Region. What architectural principle was violated?

53 A Warm Standby environment continuously handles 10% of production traffic and is sized for 20% of peak load. Scaling to full capacity takes 18 minutes, but the business RTO is 10 minutes. Which adjustment most directly aligns the architecture with the RTO?

54 A Warm Standby system uses asynchronous database replication. At disaster declaration, monitoring reports a replication lag of 90 seconds. DNS failover completes in 20 seconds, and application scaling completes in 5 minutes. Which statement correctly distinguishes its recovery objectives?

55 A standby Region is running a reduced but functional application stack. Its database is current, but background workers are disabled to avoid duplicate processing. During failover, web traffic is redirected before workers are enabled, causing accepted requests to remain unprocessed. Which orchestration sequence is most appropriate?

56 A Warm Standby exercise succeeds under average traffic but fails at peak because the recovery Region reaches an Elastic Load Balancing quota before Auto Scaling finishes. Which testing practice would have exposed this weakness most reliably?

57 Two active Regions accept updates to the same customer profile through DynamoDB global tables. Concurrent writes update different attributes of the same item, but one update disappears after replication convergence. Which change best reduces this conflict risk?

58 An active-active API uses Route 53 latency-based routing across two Regions. A Region remains reachable and returns HTTP 200 responses from /health, but its dependency on a payment provider is broken, so all payment requests fail. Which health design most effectively prevents traffic from continuing to reach that Region?

59 An active-active application processes messages from independent regional queues. A producer retry and cross-Region rerouting can cause the same business event to appear in both queues. Which consumer design is required to preserve exactly-once business effects despite at-least-once delivery?

60 A two-Region active-active service has independent availability probabilities of per Region for a given interval. The service remains available if at least one Region is available, and shared dependencies are ignored. What is the theoretical service availability?