Exam Bundle

Store-level net revenue equals the validated reference calculation would demonstrate the separate action Normalize missing-discount semantics according to the data contract, which addresses “Different stores encode null discount fields inconsistently.” It does not verify the requirement “An upstream source changes the amount column from cents to major currency units” following Detect the unit change and apply explicit normalization before aggregation.

That observation is a valid indicator for “A report must compare two monetary datasets with a small specified numeric tolerance” when using Compute per-key differences and flag only values outside the tolerance. The scenario instead tests “An upstream source changes the amount column from cents to major currency units”; the relevant result is A known invoice fixture yields the expected standardized revenue amount.

This signal belongs to the adjacent case “A retrying source double-delivers payments with the same payment event ID,” whose remedy is Deduplicate on the unique payment event ID before aggregation. The observed result for the present requirement is A known invoice fixture yields the expected standardized revenue amount.

Inspect silver expectation metrics and rejected source samples belongs to the separate case “Bronze ingest succeeds but silver quality checks suddenly drop many records,” where An implicated quality rule and source payload explain the losses would be expected. That differs from the condition in this stem: “An expectation suddenly rejects many new source records and operators need its run-level metrics.”

Compare event-log update and flow timings across runs belongs to the separate case “An SLA breach begins after a recent pipeline refresh and needs duration analysis,” where The slow phase is identified from stage-specific timing evidence would be expected. That differs from the condition in this stem: “An expectation suddenly rejects many new source records and operators need its run-level metrics.”

Join usage records to available job tags and identity metadata is appropriate when “A new job increases expenditure and FinOps needs to identify the owner”; it would be verified by The report identifies the responsible job and owning cost center. It does not directly satisfy “An expectation suddenly rejects many new source records and operators need its run-level metrics,” whose relevant acceptance evidence is The event records identify the expectation and failed-record count.

An implicated quality rule and source payload explain the losses pertains to a different control (Inspect silver expectation metrics and rejected source samples) and a different need (Bronze ingest succeeds but silver quality checks suddenly drop many records). It cannot establish that Query the pipeline event log for expectation outcome metrics met the stated goal.

This signal belongs to the adjacent case “Late-data drops rise after a watermark configuration change,” whose remedy is Inspect watermark position and dropped-late-record progress counters. The observed result for the present requirement is The event records identify the expectation and failed-record count.

That observation is a valid indicator for “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots” when using Expose or query the supported event-log table through SQL. The scenario instead tests “An expectation suddenly rejects many new source records and operators need its run-level metrics”; the relevant result is The event records identify the expectation and failed-record count.

The alternative Differentiate task retries from terminal job run failure notifications has a distinct purpose—A retry event should not notify customers as if the entire workflow failed. It would not establish the required outcome (An update failed and engineers need the exact failure text and originating flow); the necessary verification concerns The log identifies the failing flow and recorded exception.

The alternative Expose or query the supported event-log table through SQL has a distinct purpose—An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots. It would not establish the required outcome (An update failed and engineers need the exact failure text and originating flow); the necessary verification concerns The log identifies the failing flow and recorded exception.

Use supported Jobs task repair or selective pipeline retry where applicable is appropriate when “A pipeline failure needs targeted replay without rerunning unaffected upstream tasks”; it would be verified by Only the intended failed task/path reruns and downstream outputs are correct. It does not directly satisfy “An update failed and engineers need the exact failure text and originating flow,” whose relevant acceptance evidence is The log identifies the failing flow and recorded exception.

This signal belongs to the adjacent case “A pipeline failure needs targeted replay without rerunning unaffected upstream tasks,” whose remedy is Use supported Jobs task repair or selective pipeline retry where applicable. The observed result for the present requirement is The log identifies the failing flow and recorded exception.

This signal belongs to the adjacent case “A new pipeline produces no useful quality metrics for rejected data,” whose remedy is Add supported expectations and inspect corresponding event metrics. The observed result for the present requirement is The log identifies the failing flow and recorded exception.

That observation is a valid indicator for “A retry event should not notify customers as if the entire workflow failed” when using Differentiate task retries from terminal job run failure notifications. The scenario instead tests “An update failed and engineers need the exact failure text and originating flow”; the relevant result is The log identifies the failing flow and recorded exception.

This action solves a different issue: “A new pipeline produces no useful quality metrics for rejected data.” Its normal signal is Test invalid rows create observable expectation metrics in event logs. Here the requested behavior is “An SLA breach begins after a recent pipeline refresh and needs duration analysis,” so that signal is not sufficient.

Use the Jobs API run listing with relevant filters and pagination belongs to the separate case “Automation needs to list recent failed runs for a configured job,” where The result includes the desired failed runs without missing later pages would be expected. That differs from the condition in this stem: “An SLA breach begins after a recent pipeline refresh and needs duration analysis.”

This action solves a different issue: “A job unexpectedly exceeds its runtime SLA without failing.” Its normal signal is The on-call group receives a message after the defined duration. Here the requested behavior is “An SLA breach begins after a recent pipeline refresh and needs duration analysis,” so that signal is not sufficient.

A specific incompatible dependency/version is identified pertains to a different control (Review run environment versions and driver exception logs) and a different need (A production job fails only after a library was upgraded). It cannot establish that Compare event-log update and flow timings across runs met the stated goal.

Test invalid rows create observable expectation metrics in event logs would demonstrate the separate action Add supported expectations and inspect corresponding event metrics, which addresses “A new pipeline produces no useful quality metrics for rejected data.” It does not verify the requirement “An SLA breach begins after a recent pipeline refresh and needs duration analysis” following Compare event-log update and flow timings across runs.

The on-call group receives a message after the defined duration pertains to a different control (Configure a duration-based warning or supported monitoring alert) and a different need (A job unexpectedly exceeds its runtime SLA without failing). It cannot establish that Compare event-log update and flow timings across runs met the stated goal.

Pass the intended Databricks CLI profile or host configuration is appropriate when “A CLI script must use a production workspace rather than a developer's default profile”; it would be verified by CLI responses identify the correct production workspace. It does not directly satisfy “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots,” whose relevant acceptance evidence is Historic pipeline updates can be filtered by ID and timestamp.

Schedule alert evaluation after the ETL dependency succeeds belongs to the separate case “An alert should evaluate only after daily ETL completes rather than on stale intermediate data,” where The alert reflects the completed day's data, not an earlier snapshot would be expected. That differs from the condition in this stem: “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots.”

The alternative Add supported expectations and inspect corresponding event metrics has a distinct purpose—A new pipeline produces no useful quality metrics for rejected data. It would not establish the required outcome (An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots); the necessary verification concerns Historic pipeline updates can be filtered by ID and timestamp.

The alert reflects the completed day's data, not an earlier snapshot pertains to a different control (Schedule alert evaluation after the ETL dependency succeeds) and a different need (An alert should evaluate only after daily ETL completes rather than on stale intermediate data). It cannot establish that Expose or query the supported event-log table through SQL met the stated goal.

That observation is a valid indicator for “A new pipeline produces no useful quality metrics for rejected data” when using Add supported expectations and inspect corresponding event metrics. The scenario instead tests “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots”; the relevant result is Historic pipeline updates can be filtered by ID and timestamp.

The response identifies pending, running and terminal update states would demonstrate the separate action Query the supported pipelines API for update details, which addresses “A nightly dashboard should report active production pipeline update status.” It does not verify the requirement “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots” following Expose or query the supported event-log table through SQL.

The alternative Inspect Spark UI spill metrics and task timelines has a distinct purpose—A stage repeatedly spills large intermediates to disk. It would not establish the required outcome (A new pipeline produces no useful quality metrics for rejected data); the necessary verification concerns Test invalid rows create observable expectation metrics in event logs.

Inspect silver expectation metrics and rejected source samples is appropriate when “Bronze ingest succeeds but silver quality checks suddenly drop many records”; it would be verified by An implicated quality rule and source payload explain the losses. It does not directly satisfy “A new pipeline produces no useful quality metrics for rejected data,” whose relevant acceptance evidence is Test invalid rows create observable expectation metrics in event logs.

Compare event-log update and flow timings across runs belongs to the separate case “An SLA breach begins after a recent pipeline refresh and needs duration analysis,” where The slow phase is identified from stage-specific timing evidence would be expected. That differs from the condition in this stem: “A new pipeline produces no useful quality metrics for rejected data.”

That observation is a valid indicator for “The alert query produces no row because ingestion failed and this must be distinguished from zero” when using Define explicit null/no-data handling in the alert query and condition. The scenario instead tests “A new pipeline produces no useful quality metrics for rejected data”; the relevant result is Test invalid rows create observable expectation metrics in event logs.

The arrival rate exceeds processing capacity during the backlog window would demonstrate the separate action Compare input rate and processed-rows rate in streaming progress, which addresses “A stream seems healthy but event processing falls farther behind incoming data.” It does not verify the requirement “A new pipeline produces no useful quality metrics for rejected data” following Add supported expectations and inspect corresponding event metrics.

Historic pipeline updates can be filtered by ID and timestamp would demonstrate the separate action Expose or query the supported event-log table through SQL, which addresses “An engineer needs a persistent queryable audit of pipeline events rather than dashboard screenshots.” It does not verify the requirement “A new pipeline produces no useful quality metrics for rejected data” following Add supported expectations and inspect corresponding event metrics.