AWS Serverless Microcredential Sample Questions

Use this page if you are tracking AWS serverless credential updates and want an early self-check around production serverless decision-making.

This is an update-watch page. It is not an official AWS exam guide. Until formal public details are available, treat the model below as a practical preparation checklist for serverless design, operations, and troubleshooting.

Candidate preparation model

Sample Exam Questions

Try these 12 original AWS Serverless Microcredential planning questions. They are designed for self-assessment and are not official AWS exam questions.

Question 1

Topic: idempotency

A Lambda function processes payment events from a queue. Occasionally the same event is delivered twice. What design choice best prevents duplicate charges?

• A. Increase the function timeout only
• B. Store and check an idempotency key before applying the payment action
• C. Disable logging
• D. Remove the queue

Best answer: B

Explanation: Distributed event systems can retry or redeliver messages. Idempotency keys let the function recognize repeated events and avoid applying the same side effect twice.

Question 2

Topic: orchestration

A workflow needs approval, a fraud check, payment capture, shipment creation, and compensation if shipment creation fails. Which service pattern best fits?

• A. One untracked script on a laptop
• B. Direct synchronous calls with no retry handling
• C. A Step Functions workflow with explicit states, retries, and compensation logic
• D. An S3 bucket policy only

Best answer: C

Explanation: Multi-step business workflows need visible state, retry rules, and compensating actions. Step Functions is a better fit than hiding the process inside one opaque function.

Question 3

Topic: asynchronous integration

A web API should accept a request quickly while slower downstream processing continues later. What is the best serverless pattern?

• A. Make the API client wait for all downstream systems
• B. Put the accepted request on a queue or event bus and process it asynchronously
• C. Increase client-side retries without changing the backend
• D. Remove authentication

Best answer: B

Explanation: Queues and event buses decouple the user-facing request from slower work. This improves responsiveness and protects downstream systems during bursts.

Question 4

Topic: least privilege

A Lambda function reads one DynamoDB table and writes logs. Which IAM approach is best?

• A. Attach administrator access to simplify debugging
• B. Use a role limited to the required table actions and logging actions
• C. Store AWS root credentials in environment variables
• D. Share one role across every application

Best answer: B

Explanation: Serverless functions should have narrowly scoped execution roles. Least privilege reduces blast radius and makes access easier to audit.

Question 5

Topic: retries and dead-letter queues

A function triggered by a queue fails on malformed messages. The same bad message keeps blocking useful work. What should the team add?

• A. A dead-letter queue or failure destination plus validation and alerting
• B. A larger logo on the dashboard
• C. Unlimited retries with no visibility
• D. A manual spreadsheet of all messages

Best answer: A

Explanation: Bad messages should be isolated for inspection instead of blocking the pipeline indefinitely. DLQs, failure destinations, validation, and alarms improve reliability.

Question 6

Topic: concurrency control

A backend database is overloaded when thousands of events trigger functions at once. Which control helps most?

• A. Remove all connection limits
• B. Disable the database
• C. Use reserved concurrency, queue buffering, or back-pressure to protect the database
• D. Increase user passwords

Best answer: C

Explanation: Serverless scaling can overwhelm downstream dependencies. Concurrency limits, buffers, and back-pressure protect bottleneck services while work continues safely.

Question 7

Topic: API security

An API Gateway endpoint exposes customer data. Which control combination is most appropriate?

• A. Authentication, authorization, input validation, throttling, and logging
• B. Public access with no monitoring
• C. Client-side filtering only
• D. A hidden URL with no auth

Best answer: A

Explanation: APIs that expose sensitive data need identity, authorization, input validation, abuse controls, and observability. Obscurity is not a security control.

Question 8

Topic: observability

Users report intermittent errors across API Gateway, Lambda, and a downstream service. What should the team use to trace requests end to end?

• A. A single success counter
• B. Manual screenshots only
• C. A weekly meeting note
• D. Correlated logs, metrics, traces, and request IDs across the workflow

Best answer: D

Explanation: Serverless applications are distributed. Correlation IDs and traces help connect symptoms across gateways, functions, queues, state machines, and dependencies.

Question 9

Topic: event filtering

An EventBridge rule sends every event in an account to a function, but the function only needs order-cancelled events. What should change?

• A. Remove all events from the account
• B. Add an event pattern that filters only the required events before invoking the target
• C. Increase the function package size
• D. Store unrelated events in environment variables

Best answer: B

Explanation: Filtering at the event bus reduces unnecessary invocations, cost, and noise. The function should not receive broad events it does not need.

Question 10

Topic: cold-start trade-off

A latency-sensitive API has occasional cold-start delays. What is the best first response?

• A. Evaluate runtime, package size, initialization work, memory, and provisioned concurrency for the critical path
• B. Move all code into one massive function
• C. Disable metrics because they add noise
• D. Remove input validation

Best answer: A

Explanation: Cold-start mitigation depends on workload. Smaller packages, lighter initialization, memory tuning, and provisioned concurrency may help when latency requirements justify the cost.

Question 11

Topic: deployment safety

A team deploys a new Lambda version and wants to limit customer impact if errors increase. What release pattern is best?

• A. Replace all production traffic immediately with no alarms
• B. Delete the previous version before testing
• C. Use aliases, gradual traffic shifting, alarms, and rollback
• D. Disable deployment history

Best answer: C

Explanation: Gradual deployments with alarms and rollback reduce blast radius. Versioning and aliases make safe traffic shifting possible.

Question 12

Topic: data storage fit

A serverless application needs low-latency key-value lookups with automatic scaling and event-driven processing. Which service pairing is most natural?

• A. DynamoDB with stream processing where change events are needed
• B. A manually patched database on one instance for every workload
• C. Static website hosting only
• D. A DNS hosted zone as the primary datastore

Best answer: A

Explanation: DynamoDB fits many serverless key-value access patterns, and streams can drive downstream processing when item changes matter.

What to open now

• Need current AWS developer practice? Open AWS DVA-C02 .
• Need architecture trade-offs? Open AWS SAA-C03 .
• Want notifications for this possible microcredential route? Use the Notify me form above.