100% Latest & Updated Amazon AWS Certified Data Engineer - Associate DEA-C01 Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 320
An S3 prefix contains `.json.gz` objects but the Glue crawler labels them as raw JSON and downstream parsers fail. Which engineering change best addresses the underlying issue?
- Iterate with the supported LastEvaluatedKey mechanism and checkpoint progress.
- Restore a compatible snapshot/checkpoint pair and reconcile changes before resuming writes.
- Configure the supported classifier and decompression-aware source layout.
- Reassess partition keys/granularity and apply efficient columnar file layout.
Explanation
Correct Answer
C:Configure the supported classifier and decompression-aware source layout. It directly addresses the case cause (The crawler classifier or file-format handling does not match actual data encoding.) and should be verified by: A test compressed object creates the expected typed catalog columns.
Incorrect Answers
A:That change addresses a DynamoDB archival pipeline uses Scan without pagination and misses records beyond the first page; it would not repair this incident.
B:That change addresses a cross-region disaster recovery table is restored from an old snapshot, and CDC replication starts from an incompatible new checkpoint; it would not repair this incident.
D:That change addresses an Athena table uses daily partitions on a timestamp with millions of near-empty partition folders; it would not repair this incident.
Question 321
A team applies a reviewed fix for this situation: An S3 replication configuration excludes delete markers, but a team assumes target data will reflect source deletes instantly. Which observed outcome would most convincingly validate the fix?
- Checkouts commit atomically while analytic queries remain isolated from OLTP load.
- A controlled delete produces the documented target behavior within accepted lag.
- Representative key lookups meet the expected latency and consistency requirements.
- A test with a banner row keeps numeric fields correctly typed and records the reject.
Explanation
Correct Answer
B:A controlled delete produces the documented target behavior within accepted lag. This verifies the result of review supported replication/delete-marker configuration and establish reconciliation monitoring.
Incorrect Answers
A:This would verify another situation: A Redshift analytics warehouse holds transactional checkout rows that require thousands of short atomic row updates per second. It does not validate the current case.
C:This would verify another situation: An online session store uses S3 object listing as a low-latency millisecond key lookup service and response times are erratic. It does not validate the current case.
D:This would verify another situation: A crawler changes a decimal sales column to string after encountering a nonnumeric banner row in one export. It does not validate the current case.
Question 322
After resolving the following incident, which ongoing safeguard most directly prevents recurrence? A Glue job reading nested JSON flattens arrays incorrectly and duplicates parent order totals.
- Check service response enums when revising orchestration logic.
- Run IaC diff/change-set reviews and drift checks before release.
- Apply schema and timestamp quality tests at ingress.
- Add grain and cardinality assertions after each flattening step.
Explanation
Correct Answer
D:Add grain and cardinality assertions after each flattening step. This safeguards against a recurring cause: The transformation changes record granularity without an explicit explode/join design.
Incorrect Answers
A:That safeguard addresses another case: A workflow branches on a job-status string but overlooks the Cancelled terminal state, leaving executions stuck. It is not the control for this recurring failure.
B:That safeguard addresses another case: A serverless pipeline deploys with hand-edited IAM and S3 settings that differ among test and production accounts. It is not the control for this recurring failure.
C:That safeguard addresses another case: A data ingestion service accepts API JSON but passes inconsistent event time zones to downstream partitions. It is not the control for this recurring failure.
Question 323
A multi-account data pipeline sends operational alerts through an SNS topic that no engineer subscribes to. Select TWO complementary measures that correctly repair and validate this failure.
- Separate ingestion capacity budgets and backpressure with appropriate queues or schedules.
- A real on-call endpoint receives a test failure with actionable context.
- A representative compressed event reaches the expected normalized target once.
- Configure validated subscriptions, policies and controlled test notifications.
- Mark an alarm as working solely because Publish returned success.
Explanation
Correct Answers
B:A real on-call endpoint receives a test failure with actionable context. This is the observable result required before accepting the correction.
D:Configure validated subscriptions, policies and controlled test notifications. The underlying cause is: Publishing events is not equivalent to having a viable incident notification path.
Incorrect Answers
A:That corrective action is for a different case: A large historical data export shares the same small API rate limit as near-real-time orders and delays critical events. It does not resolve this incident.
C:This would validate a different case: A streaming producer writes compressed messages into a format the Lambda consumer assumes is plain UTF-8. It does not verify the outcome required here.
E:Mark an alarm as working solely because Publish returned success. That shortcut conflicts with the stated requirement.
Question 324
A Kinesis consumer reports healthy CPU yet its IteratorAge grows for hours during normal traffic. Which technical dependency most directly explains this behavior?
- The schedule target's Region, ARN, role, or cross-account authorization is misconfigured.
- Consumption lag is growing despite adequate compute metrics and may indicate insufficient read/processing throughput.
- The SDK client ignores paginated object listing.
- The join excludes the population that must remain in the denominator.
Explanation
Correct Answer
B:Consumption lag is growing despite adequate compute metrics and may indicate insufficient read/processing throughput. The corresponding fix is to examine iterator age, shard utilization, downstream latency and consumer parallelism.
Incorrect Answers
A:This diagnosis addresses another AWS case: An EventBridge schedule was configured in one Region but the receiving Step Functions state machine lives in another without a reviewed target design. The dependency here is different.
C:This diagnosis addresses another AWS case: A daily Python ingestion script calls S3 ListObjectsV2 only once and misses most keys after data growth. The dependency here is different.
D:This diagnosis addresses another AWS case: An Athena view compares monthly cohorts using an inner join that drops customers with no activity, overstating retention. The dependency here is different.
Question 325
A Redshift query joins external and local datasets but expects Spectrum to automatically overwrite remote S3 data during UPDATE. Which engineering change best addresses the underlying issue?
- Tune supported export parallelism and downstream compaction based on file size and latency.
- Configure the supported classifier and decompression-aware source layout.
- Use a supported ETL/write path to update the authoritative source, then refresh catalog metadata.
- Add supported partition projection or update Glue partitions in the ingestion workflow.
Explanation
Correct Answer
C:Use a supported ETL/write path to update the authoritative source, then refresh catalog metadata. It directly addresses the case cause (External-table query capability is confused with supported table mutation semantics.) and should be verified by: The remote data change is visible through the supported external query path.
Incorrect Answers
A:That change addresses a warehouse unload creates thousands of tiny output files, raising downstream S3 request and processing overhead; it would not repair this incident.
B:That change addresses an S3 prefix contains `.json.gz` objects but the Glue crawler labels them as raw JSON and downstream parsers fail; it would not repair this incident.
D:That change addresses an Athena query of yesterday's S3 partition returns zero rows because its table metadata lacks that partition; it would not repair this incident.
Question 326
A team applies a reviewed fix for this situation: A batch ETL task combines USD and EUR amounts under the same column and computes invalid totals. Which observed outcome would most convincingly validate the fix?
- An injected failure leaves no orphaned temporary cluster and preserves incident evidence.
- Only the matching authorized callback advances the waiting workflow.
- A representative compressed event reaches the expected normalized target once.
- A mixed-currency fixture reproduces independently computed accepted totals.
Explanation
Correct Answer
D:A mixed-currency fixture reproduces independently computed accepted totals. This verifies the result of preserve currency metadata, convert only at an explicit approved rate and aggregation grain.
Incorrect Answers
A:This would verify another situation: A workflow that provisions a temporary staging cluster leaks resources after a later transform fails. It does not validate the current case.
B:This would verify another situation: A Step Functions Standard workflow waits for an external vendor callback but the task token is lost in logs. It does not validate the current case.
C:This would verify another situation: A streaming producer writes compressed messages into a format the Lambda consumer assumes is plain UTF-8. It does not validate the current case.
Question 327
A workflow branches on a job-status string but overlooks the Cancelled terminal state, leaving executions stuck. Select TWO complementary measures that correctly repair and validate this failure.
- Injected cancelled and failed jobs exit with correctly classified outcomes.
- Minimize input fields, mask prohibited values, and apply approved data-handling and output-validation controls.
- Model all documented terminal statuses with success, retryable, and failure branches.
- Treat any status other than FAILED as success.
- Athena scanned bytes and cost decrease while results stay equivalent.
Explanation
Correct Answers
A:Injected cancelled and failed jobs exit with correctly classified outcomes. This is the observable result required before accepting the correction.
C:Model all documented terminal statuses with success, retryable, and failure branches. The underlying cause is: The state machine lacks complete terminal-status handling.
Incorrect Answers
B:That corrective action is for a different case: An ETL pipeline calls a Bedrock model on full records containing account secrets merely to extract public product categories. It does not resolve this incident.
D:Treat any status other than FAILED as success. That shortcut conflicts with the stated requirement.
E:This would validate a different case: An ETL job writes CSV to a large analytics lake and Athena scans every column for two-field reports. It does not verify the outcome required here.
Question 328
A Lambda data extractor assumes an EC2 instance profile is available when running as a Lambda function. Which technical dependency most directly explains this behavior?
- At-rest encryption does not secure data transmission on an external path.
- The published data access layer does not implement required masking or view restrictions.
- The runtime identity is supplied by the Lambda execution role, not a host instance profile.
- Attribute-based permissions have a policy scope broader than the approved audience.
Explanation
Correct Answer
C:The runtime identity is supplied by the Lambda execution role, not a host instance profile. The corresponding fix is to configure the intended Lambda execution role with scoped API permissions.
Incorrect Answers
A:This diagnosis addresses another AWS case: An S3 export writes regulated data to an unencrypted partner endpoint despite source-at-rest SSE-KMS protection. The dependency here is different.
B:This diagnosis addresses another AWS case: A Redshift view exposes unmasked customer SSNs to a general BI role despite a requirement for partial disclosure. The dependency here is different.
D:This diagnosis addresses another AWS case: An S3 table is tagged `classification=restricted`, but its Lake Formation LF-Tag grant allows a broad organizational unit with no business need. The dependency here is different.
Question 329
A team needs ACID updates and time-travel queries against a governed S3 lake but writes plain Parquet files with no table transaction metadata. Which technical dependency most directly explains this behavior?
- Raw object files do not by themselves provide open-table transaction semantics.
- Concurrent metadata changes lack a controlled version/change process.
- Physical metadata alone does not express authoritative business meaning and stewardship.
- The catalog partition listing is not synchronized with newly written S3 prefixes.
Explanation
Correct Answer
A:Raw object files do not by themselves provide open-table transaction semantics. The corresponding fix is to design a supported Apache Iceberg table with compatible catalog and compute engines.
Incorrect Answers
B:This diagnosis addresses another AWS case: Two developers independently edit Glue Data Catalog schema types during release, and a job fails after the last writer wins. The dependency here is different.
C:This diagnosis addresses another AWS case: An analytics team relies on Glue technical table names to determine which datasets are permitted for finance reporting. The dependency here is different.
D:This diagnosis addresses another AWS case: An Athena query of yesterday's S3 partition returns zero rows because its table metadata lacks that partition. The dependency here is different.
Question 330
An S3 lake access point exists but a cross-account principal still receives AccessDenied when its IAM and bucket policies are not aligned. Select THREE complementary measures that correctly repair and validate, and prevent recurrence of this failure.
- Ignore explicit deny rules because the access-point name looks correct.
- Review the required access-point and identity/bucket permissions for the approved path.
- Apply supported column-level permissions or a governed aggregate view and deny raw salary access.
- A sample output meets organizational reidentification risk requirements.
- Use policy-simulator and negative-access tests during access changes.
- The intended cross-account role reads only the allowed lake dataset.
Explanation
Correct Answers
B:Review the required access-point and identity/bucket permissions for the approved path. The underlying cause is: Permission evaluation spans identity, access-point, bucket and organizational controls.
E:Use policy-simulator and negative-access tests during access changes. This protects against the following risk: Permission evaluation spans identity, access-point, bucket and organizational controls.
F:The intended cross-account role reads only the allowed lake dataset. This is the observable result required before accepting the correction.
Incorrect Answers
A:Ignore explicit deny rules because the access-point name looks correct. That shortcut conflicts with the stated requirement.
C:That corrective action is for a different case: A Lake Formation table exposes a private salary column through a broad Athena SELECT grant when an analyst needs only aggregated headcount. It does not resolve this incident.
D:This would validate a different case: A tokenization pipeline uses a deterministic unkeyed hash on low-entropy postal identifiers and calls them irreversibly anonymous. It does not verify the outcome required here.
AWS Certified Data Engineer - Associate DEA-C01 Practice Tests