100% Latest & Updated Databricks Certified Data Engineer Associate Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 80
A regulated feed must retain unexpected JSON attributes without changing the approved table schema. The team observes: A new field is neither available in defined columns nor in rescued data. Which underlying cause most directly explains this failure?
- Schema handling was too strict at the raw landing boundary for this variable payload.
- The pipeline uses a scheduled batch job instead of a streaming ingestion path.
- Ingestion time was used as a replacement for the source event-time field.
- Rescue behavior or the required rescued-data column was not configured for this source.
Explanation
Correct Answer
D:Rescue behavior or the required rescued-data column was not configured for this source. Set cloudFiles.schemaEvolutionMode to rescue and keep a rescued-data column.
Incorrect Answers
A:If a telemetry source emits JSON lines with heterogeneous optional keys, “Schema handling was too strict at the raw landing boundary for this variable payload” may be the appropriate response. Rescue behavior or the required rescued-data column was not configured for this source.
B:A fraud scoring table must incorporate arriving events with low latency rather than waiting for nightly files. In those circumstances, “The pipeline uses a scheduled batch job instead of a streaming ingestion path” fits. Rescue behavior or the required rescued-data column was not configured for this source.
C:When a vendor's data source sends records with event timestamps older than ingestion time, “Ingestion time was used as a replacement for the source event-time field” could be relevant. Rescue behavior or the required rescued-data column was not configured for this source.
Question 81
A customer source cannot accept newly introduced columns without explicit schema approval. Which design choice best addresses the stated business and technical requirement?
- Use failOnNewColumns rather than silent schema evolution.
- Choose COPY INTO for a simple repeatable bulk import.
- Keep binary or text document payloads linked to their metadata in governed bronze records.
- Target the catalog.schema.table identifier in a COPY INTO command.
Explanation
Correct Answer
A:Use failOnNewColumns rather than silent schema evolution. The stream stops for an unapproved added column until the schema is updated.
Incorrect Answers
B:When a static set of archive files must be loaded once through SQL with minimal pipeline machinery, “Choose COPY INTO for a simple repeatable bulk import” could be relevant. Use failOnNewColumns rather than silent schema evolution.
C:If an application exports documents plus metadata into object storage, “Keep binary or text document payloads linked to their metadata in governed bronze records” may be the appropriate response. Use failOnNewColumns rather than silent schema evolution.
D:A platform team must load raw cloud files to a three-part Unity Catalog table using SQL. In those circumstances, “Target the catalog.schema.table identifier in a COPY INTO command” fits. Use failOnNewColumns rather than silent schema evolution.
Question 82
A customer source cannot accept newly introduced columns without explicit schema approval. Which setup step most directly realizes the proposed design?
- Read the stream with an appropriate source and durable checkpoint, then write to a Delta sink.
- Implement cursor handling, throttling backoff, and idempotent writes to the bronze table.
- Parse the items as an array and use explode for one output row per element.
- Select failOnNewColumns and institute a reviewed schema-change deployment.
Explanation
Correct Answer
D:Select failOnNewColumns and institute a reviewed schema-change deployment. The stream stops for an unapproved added column until the schema is updated.
Incorrect Answers
A:When a fraud scoring table must incorporate arriving events with low latency rather than waiting for nightly files, “Read the stream with an appropriate source and durable checkpoint, then write to a Delta sink” could be relevant. Select failOnNewColumns and institute a reviewed schema-change deployment.
B:If an external service exposes a paginated REST endpoint with rate limits and no managed connector, “Implement cursor handling, throttling backoff, and idempotent writes to the bronze table” may be the appropriate response. Select failOnNewColumns and institute a reviewed schema-change deployment.
C:An order feed supplies JSON objects with nested arrays of items. In those circumstances, “Parse the items as an array and use explode for one output row per element” fits. Select failOnNewColumns and institute a reviewed schema-change deployment.
Question 83
A customer source cannot accept newly introduced columns without explicit schema approval. What is the most relevant verification evidence for this case?
- Historical rows appear first and later changes are reflected without full custom reimports.
- The stream stops for an unapproved added column until the schema is updated.
- Authorized analysts can query loaded rows under the expected catalog and schema.
- A recent test event appears in the target within the required latency budget.
Explanation
Correct Answer
B:The stream stops for an unapproved added column until the schema is updated.
Incorrect Answers
A:If a managed enterprise-source ingestion has an initial backfill followed by periodic change updates, “Historical rows appear first and later changes are reflected without full custom reimports” may be the appropriate response. The stream stops for an unapproved added column until the schema is updated.
C:When a platform team must load raw cloud files to a three-part Unity Catalog table using SQL, “Authorized analysts can query loaded rows under the expected catalog and schema” could be relevant. The stream stops for an unapproved added column until the schema is updated.
D:If a fraud scoring table must incorporate arriving events with low latency rather than waiting for nightly files, “A recent test event appears in the target within the required latency budget” may be the appropriate response. The stream stops for an unapproved added column until the schema is updated.
Question 84
A customer source cannot accept newly introduced columns without explicit schema approval. The team observes: The pipeline continues accepting undocumented fields into the approved schema. What is the most likely mechanism behind the observed defect?
- The ingestion was left on an automatic-evolution policy contrary to the approval requirement.
- The REST client fetched only the initial page and ignored continuation tokens.
- The job watches cloud files instead of connecting to the database through ODBC.
- The run started without a reliable completion signal for the monthly export.
Explanation
Correct Answer
A:The ingestion was left on an automatic-evolution policy contrary to the approval requirement. Select failOnNewColumns and institute a reviewed schema-change deployment.
Incorrect Answers
B:When an external service exposes a paginated REST endpoint with rate limits and no managed connector, “The REST client fetched only the initial page and ignored continuation tokens” could be relevant. The ingestion was left on an automatic-evolution policy contrary to the approval requirement.
C:If a SQL Server source is accessible only through a company-approved ODBC route, “The job watches cloud files instead of connecting to the database through ODBC” may be the appropriate response. The ingestion was left on an automatic-evolution policy contrary to the approval requirement.
D:A finance feed is delivered as one vetted CSV export at the end of each month. In those circumstances, “The run started without a reliable completion signal for the monthly export” fits. The ingestion was left on an automatic-evolution policy contrary to the approval requirement.
Question 85
A streaming ingestion restarts after maintenance and must continue from its last committed progress. Which approach most directly satisfies the requirement?
- Rely on COPY INTO's file-level ingestion history.
- Prefer managed Lakeflow Connect over custom API extraction when source features fit.
- Reuse the stream's durable checkpoint with the same logical source and sink.
- Use a continuously updating Structured Streaming ingestion pattern.
Explanation
Correct Answer
C:Reuse the stream's durable checkpoint with the same logical source and sink. Previously committed files are not replayed simply because the job process restarted.
Incorrect Answers
A:When a data owner receives daily Parquet exports and wants previously copied files skipped on reruns, “Rely on COPY INTO's file-level ingestion history” could be relevant. Reuse the stream's durable checkpoint with the same logical source and sink.
B:If a common SaaS application has a vetted managed connector with built-in ingestion state, “Prefer managed Lakeflow Connect over custom API extraction when source features fit” may be the appropriate response. Reuse the stream's durable checkpoint with the same logical source and sink.
D:When a fraud scoring table must incorporate arriving events with low latency rather than waiting for nightly files, “Use a continuously updating Structured Streaming ingestion pattern” could be relevant. Reuse the stream's durable checkpoint with the same logical source and sink.
Question 86
A streaming ingestion restarts after maintenance and must continue from its last committed progress. Which configuration or engineering action would most directly implement the requirement?
- Capture corrupt-record evidence and apply an explicit reject or quarantine path.
- Schedule a Lakeflow Job to read and validate the completed monthly export.
- Restore the original checkpointLocation rather than creating an empty path.
- Configure the supported connector connection and its governed destination.
Explanation
Correct Answer
C:Restore the original checkpointLocation rather than creating an empty path. Previously committed files are not replayed simply because the job process restarted.
Incorrect Answers
A:If a customer source mixes valid records with a small number of malformed JSON documents, “Capture corrupt-record evidence and apply an explicit reject or quarantine path” may be the appropriate response. Restore the original checkpointLocation rather than creating an empty path.
B:A finance feed is delivered as one vetted CSV export at the end of each month. In those circumstances, “Schedule a Lakeflow Job to read and validate the completed monthly export” fits. Restore the original checkpointLocation rather than creating an empty path.
D:If a common SaaS application has a vetted managed connector with built-in ingestion state, “Configure the supported connector connection and its governed destination” may be the appropriate response. Restore the original checkpointLocation rather than creating an empty path.
Question 87
A streaming ingestion restarts after maintenance and must continue from its last committed progress. Which test would establish that the change worked as intended?
- Previously committed files are not replayed simply because the job process restarted.
- The connector can enumerate and ingest the selected ERP objects without broader rights.
- Re-executing the same COPY INTO statement does not reload already tracked unchanged source files.
- The successful batch count plus quarantine count reconciles to input events.
Explanation
Correct Answer
A:Previously committed files are not replayed simply because the job process restarted.
Incorrect Answers
B:When an ERP table is replicated through a managed connector but its ingestion account lacks read access, “The connector can enumerate and ingest the selected ERP objects without broader rights” could be relevant. Previously committed files are not replayed simply because the job process restarted.
C:If an operations team wants a SQL-only one-time load of historical CSV objects stored in an approved S3 prefix, “Re-executing the same COPY INTO statement does not reload already tracked unchanged source files” may be the appropriate response. Previously committed files are not replayed simply because the job process restarted.
D:A customer source mixes valid records with a small number of malformed JSON documents. In those circumstances, “The successful batch count plus quarantine count reconciles to input events” fits. Previously committed files are not replayed simply because the job process restarted.
Question 88
A streaming ingestion restarts after maintenance and must continue from its last committed progress. The team observes: A restart unexpectedly reprocesses older source files. Which configuration or design gap best explains the symptom?
- The managed ingestion schedule or ongoing sync was never activated.
- The connector destination was set to the wrong catalog or schema.
- The pipeline treated an array as one scalar field instead of exploding its elements.
- The checkpoint location changed, making the restarted stream a new query.
Explanation
Correct Answer
D:The checkpoint location changed, making the restarted stream a new query. Restore the original checkpointLocation rather than creating an empty path.
Incorrect Answers
A:When a managed enterprise-source ingestion has an initial backfill followed by periodic change updates, “The managed ingestion schedule or ongoing sync was never activated” could be relevant. The checkpoint location changed, making the restarted stream a new query.
B:If an audit team needs stable ownership of data ingested from a managed enterprise application, “The connector destination was set to the wrong catalog or schema” may be the appropriate response. The checkpoint location changed, making the restarted stream a new query.
C:An order feed supplies JSON objects with nested arrays of items. In those circumstances, “The pipeline treated an array as one scalar field instead of exploding its elements” fits. The checkpoint location changed, making the restarted stream a new query.
Question 89
A JSON source contains additional nested keys that are not part of a strongly typed downstream contract. Which architecture or engineering decision best fits this scenario?
- Keep unexpected fields in a rescued-data column for later triage.
- Use a bounded REST extraction in a scheduled notebook with explicit pagination and retry handling.
- Evaluate a governed partner connector compatible with the required source and volume.
- Preserve the nested data at bronze and explicitly flatten items for the serving table.
Explanation
Correct Answer
A:Keep unexpected fields in a rescued-data column for later triage. The unexpected nested key and source path can be inspected in the rescue output.
Incorrect Answers
B:An external service exposes a paginated REST endpoint with rate limits and no managed connector. In those circumstances, “Use a bounded REST extraction in a scheduled notebook with explicit pagination and retry handling” fits. Keep unexpected fields in a rescued-data column for later triage.
C:When a proprietary application has no native managed connector but a supported third-party integration, “Evaluate a governed partner connector compatible with the required source and volume” could be relevant. Keep unexpected fields in a rescued-data column for later triage.
D:If an order feed supplies JSON objects with nested arrays of items, “Preserve the nested data at bronze and explicitly flatten items for the serving table” may be the appropriate response. Keep unexpected fields in a rescued-data column for later triage.
Question 90
A JSON source contains additional nested keys that are not part of a strongly typed downstream contract. Which implementation step is most appropriate before production use?
- Capture corrupt-record evidence and apply an explicit reject or quarantine path.
- Create the required governed Delta target and use COPY INTO to load the approved cloud path.
- Read the stream with an appropriate source and durable checkpoint, then write to a Delta sink.
- Configure schema rescue and inspect _rescued_data for the source record.
Explanation
Correct Answer
D:Configure schema rescue and inspect _rescued_data for the source record. The unexpected nested key and source path can be inspected in the rescue output.
Incorrect Answers
A:A customer source mixes valid records with a small number of malformed JSON documents. In those circumstances, “Capture corrupt-record evidence and apply an explicit reject or quarantine path” fits. Configure schema rescue and inspect _rescued_data for the source record.
B:When a platform team must load raw cloud files to a three-part Unity Catalog table using SQL, “Create the required governed Delta target and use COPY INTO to load the approved cloud path” could be relevant. Configure schema rescue and inspect _rescued_data for the source record.
C:If a fraud scoring table must incorporate arriving events with low latency rather than waiting for nightly files, “Read the stream with an appropriate source and durable checkpoint, then write to a Delta sink” may be the appropriate response. Configure schema rescue and inspect _rescued_data for the source record.
Certified Data Engineer Associate Practice Tests