You save $34.99
An auditor reviews the applied Fabric control: Create a Fabric Environment with the approved runtime and packages and set it as the workspace default. It was selected to address: A data engineering team needs a common Spark runtime and Python library versions across 12 notebooks, without editing each notebook independently. Which evidence demonstrates the intended result?
Environment-backed workspace defaults provide reusable Spark runtime and library configuration.
Domains group analytics content by business ownership rather than granting fine-grained data permissions. It does not show that a new notebook using Workspace default loads the approved library versions.
Outbound protection restricts data egress paths independently of ordinary item-level read permissions. It does not show that a new notebook using Workspace default loads the approved library versions.
Airflow configuration belongs to the workspace/orchestration setup rather than individual Spark SQL statements. It does not show that a new notebook using Workspace default loads the approved library versions.
A Fabric engineer has implemented the following change: Apply deterministic key-based deduplication, such as a window with row_number ordered by version. The original requirement was: A source sends duplicate orders with different ingestion timestamps and only the newest version per order must remain. Which observation best validates that the change works?
Windowed ranking can select one authoritative row from repeated records.
Distributed Spark processing supports large programmable transforms beyond simple portal mapping. It does not show that each order ID appears exactly once with the latest valid version.
Eventhouse supports KQL-based event exploration and real-time analytics. It does not show that each order ID appears exactly once with the latest valid version.
Mirroring depends on source-specific permissions and configuration, not only workspace Reader rights. It does not show that each order ID appears exactly once with the latest valid version.
A Fabric data engineering release must address both of these independent requirements: (1) The organization wants Finance domain contributors to find a certified dataset without making it publicly writable. (2) A team needs to control which external destinations OneLake-enabled workspace items can connect to for sensitive datasets. Which combined plan meets both without substituting an unrelated change?
Step 1 resolves requirement A: Domain metadata and endorsement communicate ownership and trust, not write grants. Step 2 resolves requirement B: Outbound protection restricts data egress paths independently of ordinary item-level read permissions. Both acceptance tests are required: the dataset appears under the domain and unauthorized edits remain denied; and an approved endpoint remains reachable and an unapproved outbound connection is blocked.
Outbound protection restricts data egress paths independently of ordinary item-level read permissions. The other proposed action (Review capacity and Spark job concurrency/resource settings before increasing retries) does not implement the first control. Domain metadata and endorsement communicate ownership and trust, not write grants.
Domain metadata and endorsement communicate ownership and trust, not write grants. The other proposed action (evaluate the supported starter pool for low-latency sessions while preserving a separate batch capacity strategy) addresses a different operation. Outbound protection restricts data egress paths independently of ordinary item-level read permissions.
Domain metadata and endorsement communicate ownership and trust, not write grants. Outbound protection restricts data egress paths independently of ordinary item-level read permissions. The proposed steps (Review capacity and Spark job concurrency/resource settings before increasing retries; evaluate the supported starter pool for low-latency sessions while preserving a separate batch capacity strategy) address other capabilities instead of this required pair.
A shared Fabric backlog includes two mandatory controls. First: Operations needs ten-minute averages recalculated every two minutes over a live signal. Second: An engineer plans to solve a batch nightly sales reconciliation using Eventstream solely because the records are in JSON. Which two-part implementation satisfies them together?
Step 1 resolves requirement A: Overlapping hopping windows recompute aggregates over sliding intervals. Step 2 resolves requirement B: Streaming imposes state and always-on operation costs that a finite batch task may not need. Both acceptance tests are required: a source event participates in the correct overlapping windows; and one complete nightly batch reconciles totals without unnecessary continuous event processing.
Overlapping hopping windows recompute aggregates over sliding intervals. The other proposed action (create a supported OneLake shortcut within the KQL database to the source data) addresses a different operation. Streaming imposes state and always-on operation costs that a finite batch task may not need.
Overlapping hopping windows recompute aggregates over sliding intervals. Streaming imposes state and always-on operation costs that a finite batch task may not need. The proposed steps (Use a supported make-series query with an appropriate default value and time range; create a supported OneLake shortcut within the KQL database to the source data) address other capabilities instead of this required pair.
Streaming imposes state and always-on operation costs that a finite batch task may not need. The other proposed action (Use a supported make-series query with an appropriate default value and time range) does not implement the first control. Overlapping hopping windows recompute aggregates over sliding intervals.
A team is working on two related Fabric work items. Work item A: A pipeline Copy activity fails with an HTTP 401 after a source credential rotation. Work item B: A Dataflow Gen2 refresh fails because the owner of internal staging artifacts left the tenant. Select TWO specific implementation steps.
System-created staging permissions can be tied to an original owner and may outlive a user's access. Confirm that the refresh can access the required staging artifacts without granting unrestricted permissions.
An authentication failure must be fixed at the connection boundary, not by changing data mapping. Confirm that the authenticated copy test succeeds using the rotated credential.
Conflicting source matches prevent unambiguous one-target-row updates.
Shortcut path references are not automatically rewritten when source folders are relocated.
A Fabric data engineer receives a change request: An automated deployment is using an overprivileged owner credential even though the release updates only two selected items. Which action best produces this measurable result: the release can update approved items and is denied on an unrelated workspace?
Least-privilege CI/CD identities reduce lateral impact if automation credentials are compromised. Verify that the release can update approved items and is denied on an unrelated workspace.
Environment comparison reveals additions, deletions and changed definitions before an update. However, this alone does not show that the release can update approved items and is denied on an unrelated workspace.
Stage separation isolates release scopes so testing does not modify the production workspace. However, this alone does not show that the release can update approved items and is denied on an unrelated workspace.
Deployment pipelines promote supported item definitions between bound environments. However, this alone does not show that the release can update approved items and is denied on an unrelated workspace.
An auditor reviews the applied Fabric control: Choose a supported stream enrichment/join pattern and define time/state limits. It was selected to address: A traffic team needs to join near-real-time events to contextual reference data with bounded state. Which evidence demonstrates the intended result?
Streaming joins must control state growth and match event keys appropriately.
Query acceleration caches relevant shortcut data for faster queries while retaining source storage. It does not show that a controlled key joins to the correct enrichment and idle state remains bounded.
Materialized aggregations can precompute repeated summaries to reduce query cost. It does not show that a controlled key joins to the correct enrichment and idle state remains bounded.
Streaming imposes state and always-on operation costs that a finite batch task may not need. It does not show that a controlled key joins to the correct enrichment and idle state remains bounded.
The operations team has applied: Write KQL with summarize count() by bin(event_time, 1h) and relevant dimensions. Its purpose was: A KQL dashboard needs hourly request totals from an Eventhouse event table. Which of the following provides the strongest direct evidence of success?
KQL binning groups events into fixed time buckets.
A stream can distribute selected events to more than one supported consumer. It does not show that a controlled test dataset yields the expected hourly counts by application.
Separating invalid records from typed valid rows allows streaming jobs to continue safely. It does not show that a controlled test dataset yields the expected hourly counts by application.
Custom microbatch side effects can rerun; checkpoint alone does not make arbitrary writes exactly once. It does not show that a controlled test dataset yields the expected hourly counts by application.
The project has two blockers. Blocker A: A team needs to distinguish a Dataflow Gen2 save-validation failure from a refresh execution failure. Blocker B: An Eventhouse ingestion pipeline has rising latency but no application exception. What pair of actions resolves both blockers?
Step 1 resolves requirement A: Editing/publishing validation and scheduled refresh are distinct operational states. Step 2 resolves requirement B: Freshness degradation can occur before a hard ingestion failure. Both acceptance tests are required: the incident report names the actual failed phase and the corresponding diagnostic; and an alert triggers when measured ingestion delay exceeds the agreed objective.
Editing/publishing validation and scheduled refresh are distinct operational states. The other proposed action (inspect Spark job monitoring and the current concurrency/queue state) addresses a different operation. Freshness degradation can occur before a hard ingestion failure.
Editing/publishing validation and scheduled refresh are distinct operational states. Freshness degradation can occur before a hard ingestion failure. The proposed steps (Inspect Query Insights and execution history for the warehouse or SQL endpoint; inspect Spark job monitoring and the current concurrency/queue state) address other capabilities instead of this required pair.
Freshness degradation can occur before a hard ingestion failure. The other proposed action (Inspect Query Insights and execution history for the warehouse or SQL endpoint) does not implement the first control. Editing/publishing validation and scheduled refresh are distinct operational states.
The project has two blockers. Blocker A: A team uses the Promoted endorsement on a model but describes it as independent compliance certification. Blocker B: A user sees a table through a notebook but an unauthorized third-party reader should not bypass its OneLake row filters. What pair of actions resolves both blockers?
Step 1 resolves requirement A: Promotion is a discoverability signal; certification requires an authorized organizational review process. Step 2 resolves requirement B: OneLake can block unsupported engines when a table requires enforcement of granular row or column restrictions. Both acceptance tests are required: the displayed label and review evidence agree with the actual governance decision; and the supported engine returns filtered rows and an unsupported path cannot bypass them.
OneLake can block unsupported engines when a table requires enforcement of granular row or column restrictions. The other proposed action (Correlate audit events with pipeline run identifiers and supported lineage records) does not implement the first control. Promotion is a discoverability signal; certification requires an authorized organizational review process.
Promotion is a discoverability signal; certification requires an authorized organizational review process. The other proposed action (use the applicable Microsoft Fabric/Microsoft Purview audit log search with principal, time and operation filters) addresses a different operation. OneLake can block unsupported engines when a table requires enforcement of granular row or column restrictions.
Promotion is a discoverability signal; certification requires an authorized organizational review process. OneLake can block unsupported engines when a table requires enforcement of granular row or column restrictions. The proposed steps (Correlate audit events with pipeline run identifiers and supported lineage records; use the applicable Microsoft Fabric/Microsoft Purview audit log search with principal, time and operation filters) address other capabilities instead of this required pair.
After a production change, the runbook records: Version and promote the supported environment and notebook together with valid references. The previous problem was: A notebook is pinned to a specific runtime but its Git commit omits the required linked environment item. What test result confirms effective remediation?
Notebook definitions can depend on environment-level libraries and Spark versions.
Automated release operations require explicit identity and API permissions independent of interactive workspace ownership. It does not show that the promoted notebook starts with its required package and runtime configuration.
Stage separation isolates release scopes so testing does not modify the production workspace. It does not show that the promoted notebook starts with its required package and runtime configuration.
Fabric Git integration versions item definitions so changes can be reviewed and traced. It does not show that the promoted notebook starts with its required package and runtime configuration.
Please check your mailbox for a message from support@examlabs.com and follow the directions.
