Amazon AWS Certified Solutions Architect - Associate SAA-C03 Practice Test 29

Question 280

A monthly job is scheduled with cron in EventBridge Scheduler but runs at the wrong local hour after a daylight-saving change. The scheduler is healthy. What configuration explains it?

Use a fixed UTC cron offset throughout the year while the business expects local wall-clock time through daylight-saving changes.
The EC2 default subnet changed the UTC time used by the Scheduler service.
The schedule uses an unintended time zone or UTC interpretation rather than the business calendar’s required zone.
The target Lambda runtime version alone controls all daylight-saving transitions.

Explanation

Correct Answer

Scheduler supports time-zone configuration; validate the selected zone and DST behavior, not simply the cron minute.

Incorrect Answers

A fixed UTC offset drifts relative to a region with DST. EventBridge Scheduler can evaluate an appropriate named time zone; test DST behavior and calendar assumptions.

Managed schedule time-zone settings are separate from EC2 networking.

The invocation schedule is evaluated by Scheduler before the target runs.

Question 281

A Step Functions workflow reserves payment before completing its eligibility-check Task because both states branch from a Parallel node. Which design defect exists?

Step Functions state machines cannot run service-integrated tasks.
Enable a Step Functions Retry policy on the reservation Task but leave its Parallel branch independent of the eligibility Task, assuming retries create a prerequisite dependency.
All Step Functions Standard workflows guarantee that incorrectly ordered state definitions will be corrected automatically.
The state machine runs dependent operations concurrently rather than gating reservation on successful validation.

Explanation

Correct Answer

A prerequisite must be sequenced, and a retry/compensation path must handle actual transition failures.

Incorrect Answers

Step Functions supports service integrations and sequential states.

Retry policies repeat failing states; they do not alter the branch dependency graph. To prevent premature reservations, define eligibility success before the reservation state and compensate failures deliberately.

Execution semantics follow the defined state graph; they do not infer business dependencies.

Question 282

A payment approval remains pending for two days in an Express Step Functions workflow configured for a few-minute execution window. Why does the approval expire?

SQS Standard queues guarantee every Step Functions Express execution persists indefinitely.
A Lambda timeout of fifteen minutes makes every Express execution last one year.
Express workflows are short-running; a long-lived approval must use a suitable durable Standard workflow design.
Increase the Express workflow logging level to ALL and assume detailed execution events extend its five-minute execution-duration quota to multiple days.

Explanation

Correct Answer

Express execution duration is bounded and is not an approval wait engine for days.

Incorrect Answers

A queue does not extend a state machine’s allowed execution duration.

Lambda invocation timeouts cannot override workflow-type limits.

Logging improves observability but does not change the maximum execution time supported by an Express state machine. A Standard workflow is appropriate for a multiday approval.

Question 283

An SQS-triggered Lambda has reserved concurrency set to zero. Messages arrive and persist but there are no successful handler invocations. What is the primary capacity cause?

An S3 object retention period prevents Lambda initialization.
Zero reserved concurrency prevents the function from running concurrent invocations.
Every SQS event source needs CloudFront before Lambda can poll it.
Increase the SQS event source mapping batch size while retaining Lambda reserved concurrency of zero, expecting bigger batches to permit new invocations.

Explanation

Correct Answer

Reserved concurrency both caps and reserves function execution; setting it to zero intentionally throttles invocations.

Incorrect Answers

S3 retention does not determine Lambda reserved concurrency.

Lambda has native SQS event-source mappings.

A batch size changes records per invocation but cannot lift a function-level reserved concurrency limit of zero. The function remains throttled.

Question 284

An SQS event source mapping sends a batch of 10 jobs to Lambda. The function processes nine, fails on one, and the entire batch is received again. What explains duplicate successful work?

SQS automatically removes every successful record from a failed batch without handler reporting.
The queue changed from standard to FIFO solely because the function returned an error.
Configure a higher SQS batch window but throw one unhandled exception, expecting the event source mapping to infer exactly which nine records succeeded.
Without supported partial-batch failure reporting, the entire failed batch is retried; handlers must be idempotent.

Explanation

Correct Answer

SQS/Lambda batch acknowledgement is at batch granularity unless partial batch response is configured and returned correctly.

Incorrect Answers

Without per-record failure protocol, the mapping cannot infer which records succeeded.

Batch failures do not mutate queue type.

Batch windows change invocation grouping, not record-level success reporting. ReportBatchItemFailures must be enabled and valid message IDs returned to prevent whole-batch retry.

Question 285

A public API Gateway REST endpoint is billed for large volumes of unauthenticated requests because the team added a URL token only in its application code after the integration. What boundary is misplaced?

Authentication or authorization must be enforced at a supported API Gateway authorizer/resource-policy layer before business integration.
Configure a usage plan API key alone and treat possession of that key as verified caller identity with per-user business authorization.
A WAF logging rule alone proves every caller has business authorization.
Changing Lambda memory usage automatically revokes invalid API tokens.

Explanation

Correct Answer

Rejecting unauthorized calls only in downstream code consumes unnecessary resources and fails to enforce an ingress identity policy.

Incorrect Answers

API keys and usage plans are not strong authentication/authorization boundaries; use appropriate IAM, JWT/Cognito or Lambda authorizers and resource policies.

Security filtering/logging is not equivalent to caller permission enforcement.

Function sizing does not implement an API authorization decision.

Question 286

An ALB listener rule routes /orders/* requests to the payments target group because a broad /* catch-all has a higher precedence than the path-specific rule. Why?

Listener rules are evaluated in configured priority order; the catch-all matches first and hides the specific rule.
An Application Load Balancer cannot evaluate URL paths at all.
Configure both /orders/* and /* listener path conditions but give the catch-all rule higher priority because its wildcard is more general.
RDS replication lag changes the order of ALB rules.

Explanation

Correct Answer

Give the intended path rule precedence, then validate target health and matching path conditions.

Incorrect Answers

ALB supports application-layer host and path routing.

ALB listener rule priority determines evaluation order; a broad wildcard above a narrow path can shadow it. The intended specific rule needs earlier priority.

Database replication does not set listener priority.

Question 287

An ECS Fargate service restarts because the tasks require scratch space exceeding the configured ephemeral storage. Memory metrics are healthy. What needs correction?

Provision sufficient supported Fargate ephemeral storage for task scratch requirements and validate disk usage.
Increase ECS service desiredCount without addressing task scratch storage.
Change the SQS visibility timeout because Fargate has no file system.
Increase Fargate task memory from 4 to 8 GiB while keeping ephemeralStorage at the same too-small GiB value, assuming RAM expands the local writable task filesystem.

Explanation

Correct Answer

Fargate task storage limits are independent of CPU and memory reservations.

Incorrect Answers

More equally undersized tasks still run out of local working space.

Visibility is a queue setting and does not enlarge local task storage.

Task RAM and ephemeral disk are independent quotas. Out-of-space failures require changing supported task scratch storage or streaming/cleanup design, not only memory.

Question 288

An EKS Horizontal Pod Autoscaler increases pod replicas, but new pods remain Pending with Insufficient cpu in the scheduler. What limiting layer is missing?

Node capacity must also scale or be provisioned to host the requested pods.
Increase the Route 53 TTL so pods receive more allocatable CPU.
Reconfigure S3 versioning to choose pod placement.
Increase only Kubernetes HPA maxReplicas and assume this provisions more EC2 node allocatable CPU despite the absence of a compatible node autoscaler.

Explanation

Correct Answer

Pod-level replication cannot create EC2 or Fargate node capacity by itself; inspect requests and node autoscaling.

Incorrect Answers

DNS cache lifetime does not create node resources.

Object versioning is unrelated to Kubernetes scheduling.

HPA changes requested pods but does not add worker-node resources. Nodes must provide enough allocatable CPU/memory to schedule them.

Question 289

An EC2 Auto Scaling group launches instances in two subnets, but one subnet has exhausted available IP addresses. Scaling activities report insufficient free IPs in that AZ. What is the direct cause?

The root user password expires after each scaling action.
Increase the EC2 Auto Scaling maximum group size while leaving the exhausted subnet CIDR and same VPC subnet selections unchanged.
CloudFront origin cache capacity determines EC2 private IPv4 allocation.
The chosen subnet cannot supply an additional private IP for new instances.

Explanation

Correct Answer

Auto Scaling needs usable subnet capacity; add address space or a healthy alternate subnet/AZ as the topology permits.

Incorrect Answers

Account credentials are unrelated to available VPC IPv4 addresses.

A larger scaling limit only authorizes additional desired instances; it does not create additional private IP addresses in the selected subnet.

A CDN cache does not allocate host subnet addresses.

Question 290

A cache-aside service writes a new product price to DynamoDB but continues returning yesterday’s value from ElastiCache until the TTL expires. Which data-consistency step is missing?

Increase ElastiCache key TTL to avoid inconsistency.
Add more EC2 worker threads without changing cache invalidation.
Turn up ElastiCache read replica count without invalidating the old price key that every cache replica still contains.
Invalidate or update the corresponding cache key when the authoritative value changes.

Explanation

Correct Answer

Cache-aside reads can serve stale entries until expiration unless writes trigger targeted invalidation or refresh.

Incorrect Answers

Longer TTL increases how long obsolete values can persist.

Concurrency does not correct a stale cache item.

Additional cache replicas can increase read capacity but do not make stale entries current. Coordinate write-through or targeted invalidation with authoritative data changes.

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

Question 280

Explanation

Correct Answer

Incorrect Answers

Question 281

Explanation

Correct Answer

Incorrect Answers

Question 282

Explanation

Correct Answer

Incorrect Answers

Question 283

Explanation

Correct Answer

Incorrect Answers

Question 284

Explanation

Correct Answer

Incorrect Answers

Question 285

Explanation

Correct Answer

Incorrect Answers

Question 286

Explanation

Correct Answer

Incorrect Answers

Question 287

Explanation

Correct Answer

Incorrect Answers

Question 288

Explanation

Correct Answer

Incorrect Answers

Question 289

Explanation

Correct Answer

Incorrect Answers

Question 290

Explanation

Correct Answer

Incorrect Answers

AWS Certified Solutions Architect - Associate SAA-C03 Practice Tests