Snowflake SnowPro Core Practice Test 69
Question 680
For an original SnowPro Core educational scenario, consider: A data steward claims a directory table automatically performs optical character recognition on every staged PDF. Select TWO complementary steps: the targeted correction and its strongest prechange validation.
- Treat directory-table creation as automatic extraction of all PDF text and image captions.
- Change: Implement a supported Snowpipe Streaming client or connector path to ingest rows directly into the intended table.
- Check: Inspect stage LIST, COPY validation or output and load history for the specified file set.
- Check: Compare LIST results, directory table rows and refresh timestamp after adding a controlled file.
- Change: Enable supported directory table metadata and maintain its refresh mechanism to inventory staged objects.
Explanation
Correct Answer
D:Directory tables expose staged file paths and metadata; they do not inherently parse unstructured file content. Acceptance: New staged files are enumerable through governed metadata without inventing extracted textual content.
E:Directory tables expose staged file paths and metadata; they do not inherently parse unstructured file content. Evidence: Compare LIST results, directory table rows and refresh timestamp after adding a controlled file.
Incorrect Answers
A:For this select two assessment, the shortcut changes or bypasses the necessary correctness, governance or operational boundary rather than correcting it. Directory tables expose staged file paths and metadata; they do not inherently parse unstructured file content.
B:Change: Implement a supported Snowpipe Streaming client or connector path to ingest rows directly into the intended table. Directory tables expose staged file paths and metadata; they do not inherently parse unstructured file content.
C:Check: Inspect stage LIST, COPY validation or output and load history for the specified file set. Directory tables expose staged file paths and metadata; they do not inherently parse unstructured file content.
Question 681
A Snowflake engineer is examining this platform situation: A cloud event says a file arrived, but its records are missing from a reporting table after a load window. A proposed shortcut is: Assume a green pipeline dashboard proves every source row was loaded successfully. What should the Snowflake engineer recommend as the narrowly scoped correction? Select ONE best answer.
- Assume a green pipeline dashboard proves every source row was loaded successfully.
- Review COPY load history and file metadata, using a deliberate replay or idempotent merge design when reprocessing is truly required.
- Use supported SELECT transformations in COPY INTO from the staged source with explicit casts and column mapping.
- Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
Explanation
Correct Answer
D:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. Applied case: A cloud event says a file arrived, but its records are missing from a reporting table after a load window.
Incorrect Answers
A:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. For this best change assessment, the shortcut changes or bypasses the necessary correctness, governance or operational boundary rather than correcting it.
B:Review COPY load history and file metadata, using a deliberate replay or idempotent merge design when reprocessing is truly required. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
C:Use supported SELECT transformations in COPY INTO from the staged source with explicit casts and column mapping. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 682
A Snowflake implementation review finds: A cloud event says a file arrived, but its records are missing from a reporting table after a load window. The team proposes: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures. Which platform principle most strongly supports the recommended response? Select ONE best answer.
- Automatic schema evolution has prerequisites and supported behaviors; incompatible type or semantic changes still require intentional governance.
- Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
- Export file count and formatting depend on supported COPY output settings and result volume; exact file counts can vary with parallel execution.
- COPY error handling can abort or skip selected failures, so successful command completion does not guarantee that every source record was accepted.
Explanation
Correct Answer
B:Demonstrable success requires: Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause. This principle explains the proposed technical approach, which is: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
Incorrect Answers
A:Automatic schema evolution has prerequisites and supported behaviors; incompatible type or semantic changes still require intentional governance. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
C:Export file count and formatting depend on supported COPY output settings and result volume; exact file counts can vary with parallel execution. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:COPY error handling can abort or skip selected failures, so successful command completion does not guarantee that every source record was accepted. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 683
The data platform team has not yet approved changes. Current evidence indicates: A cloud event says a file arrived, but its records are missing from a reporting table after a load window. Which evidence best tests the technical diagnosis before approving a disruptive change? Select ONE best answer.
- Inspect location, partition definitions, metadata refresh and query against known source files.
- Inspect connector configuration, ingestion channels or stages, offset handling and end-to-end lag.
- Inspect recent load history, file status and error messages with a controlled problem file.
- Test stage preview, file-format properties and sample loaded row values including quoted delimiters.
Explanation
Correct Answer
C:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. If confirmed, the proportionate intervention is: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
Incorrect Answers
A:Inspect location, partition definitions, metadata refresh and query against known source files. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
B:Inspect connector configuration, ingestion channels or stages, offset handling and end-to-end lag. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:Test stage preview, file-format properties and sample loaded row values including quoted delimiters. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 684
A candidate correction has been tested and the platform team requires a post-change acceptance check. Original problem: A cloud event says a file arrived, but its records are missing from a reporting table after a load window. The service owner needs positive and negative results and evidence that legitimate data workflows still function. Which acceptance result demonstrates correct platform behavior without disrupting authorized workflows? Select ONE best answer.
- Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause.
- Each document remains usable as typed semi-structured data with invalid records surfaced.
- Only files in the configured scope trigger ingestion and missing notifications are detectable.
- Recovery loads only missing or intentionally replaced records without accidental duplicates.
Explanation
Correct Answer
A:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. The approved implementation step was: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
Incorrect Answers
B:Each document remains usable as typed semi-structured data with invalid records surfaced. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
C:Only files in the configured scope trigger ingestion and missing notifications are detectable. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:Recovery loads only missing or intentionally replaced records without accidental duplicates. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 685
An implementation sign-off request identifies the following problem: A cloud event says a file arrived, but its records are missing from a reporting table after a load window. Which paired technical change and validation should close the implementation review? Select ONE best answer.
- Change: Define a Snowpipe with its COPY statement and supported AUTO_INGEST event notifications or managed file submissions. Verify: Inspect pipe status, cloud notifications or REST submissions, ingestion history and loaded row latency.
- Change: Use COPY INTO <location> with a supported file format, target stage or secure external location and governed permissions. Verify: Check export query permissions, target path, file format, partitioning and sample file contents.
- Change: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures. Verify: Inspect recent load history, file status and error messages with a controlled problem file.
- Change: Apply supported schema evolution or controlled DDL changes after confirming compatible file formats and data typing. Verify: Compare source schema, table evolution settings, COPY metadata and downstream validation tests.
Explanation
Correct Answer
C:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. The change is successful only when: Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause.
Incorrect Answers
A:This option addresses a different Snowflake technical topic (Snowpipe continuous file ingestion) rather than the documented COPY history and ingestion troubleshooting decision.
B:Change: Use COPY INTO <location> with a supported file format, target stage or secure external location and governed permissions. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:Change: Apply supported schema evolution or controlled DDL changes after confirming compatible file formats and data typing. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 686
A Snowflake engineer is examining this platform situation: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner. A proposed shortcut is: Assume a green pipeline dashboard proves every source row was loaded successfully. What should the Snowflake engineer recommend as the narrowly scoped correction? Select ONE best answer.
- Assume a green pipeline dashboard proves every source row was loaded successfully.
- Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
- Use supported SELECT transformations in COPY INTO from the staged source with explicit casts and column mapping.
- Evaluate the connector's supported ingestion mode and configure Snowpipe Streaming or file-based Snowpipe intentionally.
Explanation
Correct Answer
B:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. This targeted change addresses the stated technical or governance gap without discarding necessary controls.
Incorrect Answers
A:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. Use this instead: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
C:Use supported SELECT transformations in COPY INTO from the staged source with explicit casts and column mapping. Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner.
D:Evaluate the connector's supported ingestion mode and configure Snowpipe Streaming or file-based Snowpipe intentionally. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 687
A Snowflake implementation review finds: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner. The team proposes: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures. Which platform principle most strongly supports the recommended response? Select ONE best answer.
- Export file count and formatting depend on supported COPY output settings and result volume; exact file counts can vary with parallel execution.
- External tables expose metadata about data kept in cloud storage; they are not identical to internally managed loaded tables.
- Snowflake tracks staged-file loading metadata to help avoid accidental repeat loads, but forcing reloads can create duplicate rows.
- Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Explanation
Correct Answer
D:Demonstrable success requires: Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause. Applied case: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner.
Incorrect Answers
A:Export file count and formatting depend on supported COPY output settings and result volume; exact file counts can vary with parallel execution. Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner.
B:External tables expose metadata about data kept in cloud storage; they are not identical to internally managed loaded tables. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
C:Snowflake tracks staged-file loading metadata to help avoid accidental repeat loads, but forcing reloads can create duplicate rows. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 688
The data platform team has not yet approved changes. Current evidence indicates: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner. Which evidence best tests the technical diagnosis before approving a disruptive change? Select ONE best answer.
- Test stage preview, file-format properties and sample loaded row values including quoted delimiters.
- Inspect a staged JSON file, COPY output, TYPEOF and nested path values in loaded rows.
- Inspect recent load history, file status and error messages with a controlled problem file.
- Check client-channel status, offset or commit acknowledgement and measured end-to-end event latency.
Explanation
Correct Answer
C:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. Confirm afterward that: Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause.
Incorrect Answers
A:Test stage preview, file-format properties and sample loaded row values including quoted delimiters. Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner.
B:Inspect a staged JSON file, COPY output, TYPEOF and nested path values in loaded rows. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:Check client-channel status, offset or commit acknowledgement and measured end-to-end event latency. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 689
A candidate correction has been tested and the platform team requires a post-change acceptance check. Original problem: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner. The service owner needs positive and negative results and evidence that legitimate data workflows still function. Which acceptance result demonstrates correct platform behavior without disrupting authorized workflows? Select ONE best answer.
- Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause.
- Additive supported changes load safely and incompatible changes receive a review rather than silent corruption.
- Accepted rows meet the documented threshold and rejected data is retained for investigation.
- Analysts can query approved external data while file ownership and metadata freshness remain explicit.
Explanation
Correct Answer
A:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. The result must be validated using the documented dataset or access context.
Incorrect Answers
B:Additive supported changes load safely and incompatible changes receive a review rather than silent corruption. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
C:Accepted rows meet the documented threshold and rejected data is retained for investigation. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
D:Analysts can query approved external data while file ownership and metadata freshness remain explicit. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
Question 690
For an original SnowPro Core educational scenario, consider: Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner. Select TWO complementary steps: the targeted correction and its strongest prechange validation.
- Change: Use COPY INTO <location> with a supported file format, target stage or secure external location and governed permissions.
- Change: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
- Assume a green pipeline dashboard proves every source row was loaded successfully.
- Check: Inspect recent load history, file status and error messages with a controlled problem file.
- Check: Check client-channel status, offset or commit acknowledgement and measured end-to-end event latency.
Explanation
Correct Answer
B:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. Acceptance: Operators can explain each expected file as loaded, pending or rejected and resolve the actual cause.
D:Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination. Required intervention: Correlate COPY_HISTORY, pipe diagnostics, stage listings and target row checks to isolate load versus transformation failures.
Incorrect Answers
A:Change: Use COPY INTO <location> with a supported file format, target stage or secure external location and governed permissions. Operations sees successful Snowpipe status yet a subset of files has parsing errors and no alert reaches the owner.
C:For this select two assessment, the shortcut changes or bypasses the necessary correctness, governance or operational boundary rather than correcting it. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
E:Check: Check client-channel status, offset or commit acknowledgement and measured end-to-end event latency. Ingestion observability requires per-file and per-row evidence; event receipt alone does not prove records reached the destination.
