Amazon AWS Certified Developer - Associate DVA-C02 Practice Test 10
Question 90
An API Gateway route produces 504 responses when an upstream Lambda call takes too long. Which implementation approach BEST addresses this requirement?
- Measure the integration latency and redesign or optimize synchronous work to finish within the API integration limit.
- Set the Lambda function timeout to fifteen minutes and assume API Gateway waits equally long.
- Increase DynamoDB TTL to change HTTP integration deadlines.
- Add random retries to the front end with no timeout diagnosis.
Explanation
Correct Answer
A:Measure the integration latency and redesign or optimize synchronous work to finish within the API integration limit. API integration timeout limits are independent of a Lambda function's own maximum runtime. A confirming test would show: End-to-end traces show the integration completes under the configured limit without gateway timeouts.
Incorrect Answers
B:Set the Lambda function timeout to fifteen minutes and assume API Gateway waits equally long. This suggestion does not establish the required API Gateway Lambda integration timeout; the relevant risk is that a longer Lambda timeout alone does not prevent an earlier API Gateway integration timeout.
C:Increase DynamoDB TTL to change HTTP integration deadlines. This suggestion does not establish the required API Gateway Lambda integration timeout; the relevant risk is that a longer Lambda timeout alone does not prevent an earlier API Gateway integration timeout.
D:Add random retries to the front end with no timeout diagnosis. This suggestion does not establish the required API Gateway Lambda integration timeout; the relevant risk is that a longer Lambda timeout alone does not prevent an earlier API Gateway integration timeout.
Question 91
A checkout Lambda receives the same order-paid event twice when its upstream emitter retries after a timeout. Which implementation approach BEST addresses this requirement?
- Keep processed-event IDs only in a Lambda process variable.
- Disable every retry including recoverable failures.
- Increase the Lambda timeout without tracking event identity.
- Write a stable event ID with a conditional DynamoDB write and skip a side effect when that ID was processed.
Explanation
Correct Answer
D:Write a stable event ID with a conditional DynamoDB write and skip a side effect when that ID was processed. At-least-once message delivery needs application-level idempotency for irreversible side effects. A confirming test would show: Replaying the identical event leaves one committed business operation and one idempotency record.
Incorrect Answers
A:Keep processed-event IDs only in a Lambda process variable. This suggestion does not establish the required idempotent event handling; the relevant risk is that duplicate deliveries bypass a stable deduplication check and cause the same side effect twice.
B:Disable every retry including recoverable failures. This suggestion does not establish the required idempotent event handling; the relevant risk is that duplicate deliveries bypass a stable deduplication check and cause the same side effect twice.
C:Increase the Lambda timeout without tracking event identity. This suggestion does not establish the required idempotent event handling; the relevant risk is that duplicate deliveries bypass a stable deduplication check and cause the same side effect twice.
Question 92
A Lambda-backed application connects to an RDS database and wants to avoid a long-lived database password. Which implementation approach BEST addresses this requirement?
- Enable supported IAM database authentication, generate an auth token with the SDK, and connect over TLS with required database grants.
- Use a Cognito user-pool token as the database password.
- Store the database master password in every function's source code.
- Disable database-side access controls because IAM auth exists.
Explanation
Correct Answer
A:Enable supported IAM database authentication, generate an auth token with the SDK, and connect over TLS with required database grants. IAM database authentication issues short-lived login tokens but database user permissions still control SQL access. A confirming test would show: Authenticated connections accept fresh tokens for the mapped database user and reject unauthorized principals.
Incorrect Answers
B:Use a Cognito user-pool token as the database password. This suggestion does not establish the required IAM authentication and TLS for RDS; the relevant risk is that a correct IAM token without database user setup or TLS configuration can still fail to connect.
C:Store the database master password in every function's source code. This suggestion does not establish the required IAM authentication and TLS for RDS; the relevant risk is that a correct IAM token without database user setup or TLS configuration can still fail to connect.
D:Disable database-side access controls because IAM auth exists. This suggestion does not establish the required IAM authentication and TLS for RDS; the relevant risk is that a correct IAM token without database user setup or TLS configuration can still fail to connect.
Question 93
A mobile client needs temporary upload access to one object key without AWS credentials. Which implementation approach BEST addresses this requirement?
- Use an unsigned GET URL for a private PUT upload.
- Make the entire bucket publicly writable.
- Generate a scoped presigned S3 URL with the intended method, object key, and expiration.
- Embed an IAM user's permanent secret access key in mobile apps.
Explanation
Correct Answer
C:Generate a scoped presigned S3 URL with the intended method, object key, and expiration. Presigned requests delegate bounded use of the signing principal's permissions without exposing long-lived keys. A confirming test would show: A request succeeds before expiry for the intended action and fails after the URL is invalid.
Incorrect Answers
A:Use an unsigned GET URL for a private PUT upload. This suggestion does not establish the required S3 presigned URLs; the relevant risk is that publishing access keys in client code exposes reusable credentials beyond one object operation.
B:Make the entire bucket publicly writable. This suggestion does not establish the required S3 presigned URLs; the relevant risk is that publishing access keys in client code exposes reusable credentials beyond one object operation.
D:Embed an IAM user's permanent secret access key in mobile apps. This suggestion does not establish the required S3 presigned URLs; the relevant risk is that publishing access keys in client code exposes reusable credentials beyond one object operation.
Question 94
A payment-validation worker repeatedly fails on malformed events and should keep good events flowing. Which implementation approach BEST addresses this requirement?
- Use an SNS topic subscription as a replacement for failed-message storage.
- Configure a source queue redrive policy and a dead-letter queue with an appropriate maxReceiveCount.
- Delete all failures before preserving diagnostics.
- Increase batch size indefinitely without isolating failures.
Explanation
Correct Answer
B:Configure a source queue redrive policy and a dead-letter queue with an appropriate maxReceiveCount. A dead-letter queue isolates messages that cannot be processed after bounded delivery attempts. A confirming test would show: Poisoned messages appear in the DLQ after the chosen receives while valid events continue.
Incorrect Answers
A:Use an SNS topic subscription as a replacement for failed-message storage. This suggestion does not establish the required SQS dead-letter queues; the relevant risk is that without redrive, poison messages are repeatedly delivered and hide useful errors.
C:Delete all failures before preserving diagnostics. This suggestion does not establish the required SQS dead-letter queues; the relevant risk is that without redrive, poison messages are repeatedly delivered and hide useful errors.
D:Increase batch size indefinitely without isolating failures. This suggestion does not establish the required SQS dead-letter queues; the relevant risk is that without redrive, poison messages are repeatedly delivered and hide useful errors.
Question 95
A release should pass unit and staging integration tests before any production deploy. Which implementation approach BEST addresses this requirement?
- Trigger production deployment before build artifacts exist.
- Pull arbitrary latest images in production instead of the tested digest.
- Configure the existing pipeline stages to build, test, approve where required, and promote immutable artifacts.
- Bypass failing staging tests automatically.
Explanation
Correct Answer
C:Configure the existing pipeline stages to build, test, approve where required, and promote immutable artifacts. CI/CD stage ordering and approvals reduce uncontrolled deployment changes. A confirming test would show: Pipeline execution history shows the same artifact revision crossing successful gates before production.
Incorrect Answers
A:Trigger production deployment before build artifacts exist. This suggestion does not establish the required CodePipeline gated promotion; the relevant risk is that a pipeline that rebuilds a different artifact after approval breaks artifact provenance.
B:Pull arbitrary latest images in production instead of the tested digest. This suggestion does not establish the required CodePipeline gated promotion; the relevant risk is that a pipeline that rebuilds a different artifact after approval breaks artifact provenance.
D:Bypass failing staging tests automatically. This suggestion does not establish the required CodePipeline gated promotion; the relevant risk is that a pipeline that rebuilds a different artifact after approval breaks artifact provenance.
Question 96
A developer must test a Lambda-backed API endpoint before promoting a build to a shared AWS environment. Which implementation approach BEST addresses this requirement?
- Treat a template lint pass as proof that every business branch behaves correctly.
- Run AWS SAM CLI local API or invoke tests with representative events and runtime-compatible containers.
- Deploy directly to production to inspect each syntax error.
- Unit-test only a text file containing the handler name.
Explanation
Correct Answer
B:Run AWS SAM CLI local API or invoke tests with representative events and runtime-compatible containers. SAM local testing helps validate handler behavior before a cloud deployment but does not replace all integration tests. A confirming test would show: Test output demonstrates the expected HTTP or handler response for valid and invalid fixtures.
Incorrect Answers
A:Treat a template lint pass as proof that every business branch behaves correctly. This suggestion does not establish the required local SAM API testing; the relevant risk is that skipping runtime-compatible local tests can miss event-shape and handler configuration bugs.
C:Deploy directly to production to inspect each syntax error. This suggestion does not establish the required local SAM API testing; the relevant risk is that skipping runtime-compatible local tests can miss event-shape and handler configuration bugs.
D:Unit-test only a text file containing the handler name. This suggestion does not establish the required local SAM API testing; the relevant risk is that skipping runtime-compatible local tests can miss event-shape and handler configuration bugs.
Question 97
A team enables automatic key rotation and wants to know how older ciphertext remains decryptable. Which implementation approach BEST addresses this requirement?
- Use supported KMS automatic key rotation while keeping the original KMS key identity for decryption.
- Change the KMS key ID in every stored old ciphertext without re-encrypting.
- Delete the old key as soon as rotation is enabled.
- Store key material in an S3 public object as the rotation backup.
Explanation
Correct Answer
A:Use supported KMS automatic key rotation while keeping the original KMS key identity for decryption. KMS-managed rotation retains key identity and the old material required to decrypt previous ciphertext. A confirming test would show: Older ciphertext remains decryptable using protected historical key material and new operations use current material.
Incorrect Answers
B:Change the KMS key ID in every stored old ciphertext without re-encrypting. This suggestion does not establish the required KMS rotation scope; the relevant risk is that deleting the KMS key during rotation makes ciphertext inaccessible after deletion.
C:Delete the old key as soon as rotation is enabled. This suggestion does not establish the required KMS rotation scope; the relevant risk is that deleting the KMS key during rotation makes ciphertext inaccessible after deletion.
D:Store key material in an S3 public object as the rotation backup. This suggestion does not establish the required KMS rotation scope; the relevant risk is that deleting the KMS key during rotation makes ciphertext inaccessible after deletion.
Question 98
A Lambda producer invokes an internal Lambda consumer and must avoid shared static API keys. Which implementation approach BEST addresses this requirement?
- Make every target endpoint publicly invokable to avoid IAM trust.
- Treat a browser-supplied user ID as an AWS service principal.
- Place a permanent admin API key in both Lambda source files.
- Use IAM-authenticated AWS SDK calls with a narrowly scoped identity policy and resource policy when required.
Explanation
Correct Answer
D:Use IAM-authenticated AWS SDK calls with a narrowly scoped identity policy and resource policy when required. AWS service-to-service calls use IAM identity and resource permissions according to API requirements. A confirming test would show: CloudTrail and target logs show requests authorized only for the permitted source principal.
Incorrect Answers
A:Make every target endpoint publicly invokable to avoid IAM trust. This suggestion does not establish the required service-to-service IAM authorization; the relevant risk is that a backend-to-backend call can fail when identity access exists but a required resource policy denies the principal.
B:Treat a browser-supplied user ID as an AWS service principal. This suggestion does not establish the required service-to-service IAM authorization; the relevant risk is that a backend-to-backend call can fail when identity access exists but a required resource policy denies the principal.
C:Place a permanent admin API key in both Lambda source files. This suggestion does not establish the required service-to-service IAM authorization; the relevant risk is that a backend-to-backend call can fail when identity access exists but a required resource policy denies the principal.
Question 99
A checkout service writes an order and immediately reads it from the same Region's base table. Which implementation approach BEST addresses this requirement?
- Request a strongly consistent read for the DynamoDB table or LSI where supported.
- Use ConsistentRead on a GSI Query and assume it is supported.
- Cache the previous response to obtain stronger consistency.
- Increase the item's TTL to force immediate visibility.
Explanation
Correct Answer
A:Request a strongly consistent read for the DynamoDB table or LSI where supported. Strongly consistent reads are available on DynamoDB tables and LSIs in a Region, not GSIs. A confirming test would show: An immediate post-write read returns the latest committed item in the supported read path.
Incorrect Answers
B:Use ConsistentRead on a GSI Query and assume it is supported. This suggestion does not establish the required DynamoDB strong read consistency; the relevant risk is that an eventually consistent read may briefly return a stale value after a successful write.
C:Cache the previous response to obtain stronger consistency. This suggestion does not establish the required DynamoDB strong read consistency; the relevant risk is that an eventually consistent read may briefly return a stale value after a successful write.
D:Increase the item's TTL to force immediate visibility. This suggestion does not establish the required DynamoDB strong read consistency; the relevant risk is that an eventually consistent read may briefly return a stale value after a successful write.
Question 100
An ECS service unexpectedly runs a different build because the pipeline reuses the mutable latest tag. Which implementation approach BEST addresses this requirement?
- Change only the ECS service description without updating task image references.
- Rebuild source from the main branch during a rollback.
- Pin deployment manifests to an immutable image digest or unique tested release tag.
- Pull latest without recording its digest.
Explanation
Correct Answer
C:Pin deployment manifests to an immutable image digest or unique tested release tag. Immutable artifact references make releases and rollbacks reproducible. A confirming test would show: The running task image digest matches the tested build artifact after deployment and rollback.
Incorrect Answers
A:Change only the ECS service description without updating task image references. This suggestion does not establish the required immutable container image tags; the relevant risk is that a mutable latest tag can point to different image bytes on separate deployments.
B:Rebuild source from the main branch during a rollback. This suggestion does not establish the required immutable container image tags; the relevant risk is that a mutable latest tag can point to different image bytes on separate deployments.
D:Pull latest without recording its digest. This suggestion does not establish the required immutable container image tags; the relevant risk is that a mutable latest tag can point to different image bytes on separate deployments.