AWS Certified Solutions Architect - Associate SAA-C03 Practice Test 56
Question 550
A production EC2 worker in AZ-a has an EBS root/data setup and becomes unavailable when the AZ fails. A runbook says 'launch EC2 in AZ-b and attach the original EBS volumes,' with no snapshots or image process. What should the runbook include?
- Use reproducible AMIs/launch templates and recover data from suitable snapshots or independent stores in the surviving AZ.
- Remove all EBS snapshots to save cost because snapshotless restoration is more reliable during regional incidents.
- Change the original AZ's display label, which causes AWS to physically move attached volumes into another AZ.
- Launch a blank EC2 instance with no software and assume it inherits every database block from the failed AZ automatically.
Explanation
Correct Answer
A:A replacement host requires boot configuration plus usable data copies; zone-scoped original volumes cannot be moved directly across the failed boundary.
Incorrect Answers
B:Deleting recovery artifacts makes reconstructing unavailable zone-scoped volumes much harder or impossible.
C:Zone names are not a data migration mechanism and do not change the placement of existing EBS volumes.
D:EC2 launches do not teleport instance-specific local or EBS data without a recovery artifact.
Question 551
Route 53 failover records send customers to a secondary endpoint after the primary health check fails. Yet some browsers continue trying the primary for two minutes. A colleague claims DNS failover must switch every client instantaneously. What explains the lag?
- All Route 53 failover answers forcibly close existing client connections at the precise second the health status changes.
- Increasing record TTL to a day guarantees that clients immediately fetch the new secondary answer.
- DNS answers can be cached according to record TTL and resolver behavior, so plan tolerable failover delay.
- Disabling secondary endpoint health checks permanently makes the primary DNS record expire instantly at all clients.
Explanation
Correct Answer
C:DNS routing changes affect new lookups, while resolvers and clients may retain earlier responses until their cache behavior permits refresh.
Incorrect Answers
A:DNS services cannot guarantee immediate refresh of every external resolver cache or established client socket.
B:Long TTLs can prolong stale routing because old answers remain cached longer.
D:Health-check configuration does not forcibly invalidate worldwide resolver caches on demand.
Question 552
A Route 53 primary health check probes a static /healthz page that always returns 200, even while payment transactions fail because the database is unreachable. DNS never fails over. What should the health check reflect?
- Increase DNS TTL until Route 53 starts examining SQL transaction contents automatically.
- Disable both primary and secondary health checks, forcing every user to assume the primary is always healthy.
- Keep the static page returning 200 because a listening web socket proves payment transactions can commit.
- Probe a meaningful application-readiness endpoint that detects critical dependencies and returns an unhealthy result when the business service cannot function.
Explanation
Correct Answer
D:A transport or static-page check can miss a broken core dependency. Readiness should encode the availability condition required for useful customer traffic.
Incorrect Answers
A:TTL controls cache behavior and does not change what the configured health probe validates.
B:Removing health evidence defeats automated failover when critical application components fail.
C:The process may serve a trivial endpoint while the payment database or other dependency is down.
Question 553
An application has three EC2 targets within one AWS Region and also a separate regional DR site. A teammate expects Route 53 failover to eject an unhealthy individual EC2 target from the ALB target group every second. Which control belongs where?
- Use ALB target-group health checks for individual targets and Route 53 policies for DNS-level endpoint or regional switching.
- Set the ALB idle timeout to one second to perform cross-Region DNS switching automatically.
- Use only Route 53 failover records and disable every target-group health check on the regional ALB.
- Use an S3 object lifecycle rule to remove failed EC2 targets from the ALB when their disks fill.
Explanation
Correct Answer
A:Load balancer health operates within a target group; DNS failover answers endpoint names across broader failure domains.
Incorrect Answers
B:An idle HTTP connection setting is not a DNS health and regional failover policy.
C:DNS cannot substitute for checking each backend that shares one load-balancer endpoint.
D:S3 lifecycle applies to object storage, not registered load balancer target health.
Question 554
A global SaaS application has fully deployed endpoints in Asia and Europe. It wants users to be directed to the Region AWS predicts offers the lowest network latency, subject to healthy endpoints. Which Route 53 policy should be considered?
- Use a 50/50 weighted policy because it always predicts which Region will have the lowest RTT for each user.
- Use Route 53 latency-based routing with appropriate health checks and regional application readiness.
- Use a geolocation policy without evaluating the actual latency path and claim it guarantees minimum RTT worldwide.
- Publish one fixed EC2 address in Europe for every client, even when the Asian endpoint is nearer and healthy.
Explanation
Correct Answer
B:Latency routing chooses among named regional endpoints using AWS latency measurements for the client network rather than predetermined traffic percentages.
Incorrect Answers
A:Weights express traffic proportions, not client-specific latency preference.
C:Geographic assignment is a different routing objective and does not necessarily give the lowest network latency.
D:A single endpoint ignores the requirement for location-sensitive latency improvement.
Question 555
A retail company uses Route 53 latency-based routing across US and EU endpoints. A customer writes an order in the US Region, then a subsequent request goes to EU, which lacks that order. Why did the routing policy not solve the problem?
- Set the Route 53 DNS TTL to 10 seconds and expect a later response from either Region to have the latest US order record.
- Enable Route 53 health checks and route all healthy users to either regional endpoint without a shared or replicated order state.
- Latency DNS routing selects endpoints, but the application still needs an explicit cross-Region data consistency and replication model.
- Add CloudFront origin failover in front of the two independent regional databases and rely on cache routing to synchronize orders.
Explanation
Correct Answer
C:Cross-Region DNS latency routing chooses a reachable low-latency endpoint but has no role in replicating order state. To satisfy immediate confirmation after a US write, use an explicit consistent read path, session affinity to the writer or a data topology with the required cross-Region consistency semantics.
Incorrect Answers
A:Short DNS TTLs can change which endpoint a resolver chooses, but cannot make regional databases replicate or provide session read-after-write consistency.
B:Health-based routing identifies reachable application endpoints; it does not synchronize transaction data that exists only in another Region.
D:CloudFront edge/origin routing is not a transactional cross-Region database replication mechanism; origin choice cannot make a missing order exist.
Question 556
A global service uses latency routing between three Regions, but users are sometimes directed to an endpoint that is returning 503 after a regional failure. Which Route 53 configuration helps avoid selecting unavailable endpoints?
- Turn off CloudWatch alarms on the failed Region so its endpoint stops producing 503 responses.
- Use latency policies without any health awareness and assume a failed application always has the lowest measured latency.
- Increase traffic weights on all records, which repairs failed web applications without dependency changes.
- Associate suitable health checks or health-evaluation settings with the routed regional records.
Explanation
Correct Answer
D:Latency preference should be constrained by endpoint health, so known failed regional destinations are excluded where supported.
Incorrect Answers
A:Suppressing telemetry does not restart the unavailable application or adjust DNS health routing.
B:Latency estimates do not prove a business endpoint is actually serving valid requests.
C:Weighting does not implement health-based exclusion of endpoints that are returning errors.
Question 557
A company deploys a replacement API and wants exactly 10% of new DNS responses routed to it during a canary rollout, independent of which Region is geographically closer. Which Route 53 policy best matches the requested distribution?
- Use latency-based routing and expect it to deliver exactly 10% to the new version regardless of network conditions.
- Use geolocation routing with a list of ten countries and assume the countries always represent 10% of all DNS traffic.
- Use Route 53 weighted routing with appropriately configured record weights and monitor the rollout.
- Use Route 53 failover routing and mark the old version always unhealthy to send only one in ten users to the new version.
Explanation
Correct Answer
C:Weighted records approximate an intended DNS response split; latency policies optimize location rather than a deliberate canary percentage.
Incorrect Answers
A:Latency routing is based on measured client-to-Region paths and does not enforce an arbitrary fixed deployment ratio.
B:Country counts do not represent a controlled fraction of incoming client queries.
D:Failover is primary-secondary health behavior, not a probabilistic deployment split.
Question 558
An application already uses an AWS Global Accelerator standard accelerator with healthy endpoints in two AWS Regions. The operations team needs to reduce the portion of NEW client connections sent to the first Region during maintenance, but existing long-lived TCP connections must not be deliberately terminated. Which adjustment should be made?
- Lower the traffic dial percentage on the first Region’s Global Accelerator endpoint group and confirm that the setting applies to new connections.
- Remove both Regions from the accelerator because traffic dials affect only endpoint health checks, not new connection routing.
- Change an S3 bucket replication rule because that governs which regional load balancer receives inbound TCP connections.
- Change only the endpoint weights in the other Region and claim the percentage of total listener traffic directed to the first Region is now fixed.
Explanation
Correct Answer
A:Global Accelerator standard endpoint groups have a per-Region traffic dial from 0 to 100. It limits the portion of newly established traffic already directed toward that group; changing it does not terminate connections that are established.
Incorrect Answers
B:Removing endpoints disrupts service and ignores the supported percentage-control mechanism for new flows.
C:S3 object replication settings are unrelated to Global Accelerator client connections and regional traffic distribution.
D:Endpoint weights distribute traffic among endpoints INSIDE their group; they do not directly cap the fraction of traffic allocated to another Region.
Question 559
A payments service uses Amazon Application Recovery Controller (ARC) routing controls to direct traffic to one of two independently healthy Regions. Operators must be able to change routing-control states during failover drills but must never leave both Regions disabled at the same time. What control should the resilience design enforce?
- Use a CloudFront cache policy with a minimum TTL of zero; that policy rejects ARC changes that disable both regional cells.
- Configure an EC2 Auto Scaling cooldown in each Region, because cooldown periods validate ARC routing-control state transitions.
- Use a Route 53 latency routing record without an ARC rule, since latency-based DNS will prevent administrators from turning both routing controls off.
- Create an ARC routing-control assertion safety rule that requires at least one routing control in the panel to remain On whenever states change.
Explanation
Correct Answer
D:An ARC assertion safety rule evaluates the proposed control-panel state and rejects transitions that violate a defined invariant, such as all Regions being Off at once.
Incorrect Answers
A:A CloudFront cache policy controls HTTP caching, not ARC routing-control state-change validation. It cannot reject unsafe Region switch operations.
B:Auto Scaling cooldown settings affect instance capacity changes in a Region. They do not enforce global routing-control invariants during manual recovery operations.
C:Latency routing helps choose destinations from DNS records, but it does not stop an operator from changing two ARC routing controls into an unsafe all-Off configuration.
Question 560
Global Accelerator sends users to both US and EU backend ALBs. If the US Region fails, all US user traffic may shift to the EU region. The EU ALB's targets are sized for only its normal users. What failure readiness issue exists?
- Assume the accelerator automatically multiplies EU EC2 and relational database capacity without service quotas or scaling configuration.
- Reserve or rapidly provision enough healthy secondary-region application and database capacity to absorb redirected traffic.
- Configure the EU ALB to respond 200 to all health checks while dropping excess business requests.
- Disable all autoscaling in the EU Region because static capacity always grows automatically after DNS failure.
Explanation
Correct Answer
B:Traffic steering can move connections but cannot create backend throughput or database headroom in a Region that is undersized for failover.
Incorrect Answers
A:Global Accelerator routes traffic and does not bypass downstream provisioning or database limits.
C:False health makes routing decisions misleading and does not expand usable application capacity.
D:Without capacity planning, a surviving Region can saturate during a global traffic shift.
