Data-Engineer-Associate - Governance and Security (15% of the exam) - Section 7.4

Understand Unity Catalog ABAC policies to centrally control row-level filtering and column masking for sensitive data.

Explain how attribute-based access control uses governed tags to apply row filtering and column masking centrally across many tables, and how this differs from attaching a filter or mask to each table individually.

attribute-based access controlgoverned tagsABAC policiescentral policy managementsensitive data

Practice question for this objective

Free sampleGovernance and Securitymedium

A governance team must redact a national_id column that appears in more than forty Unity Catalog tables inside one catalog, and further tables carrying the same column are created every month. The team wants one central definition of the rule rather than a mask attached table by table. Which two properties of Unity Catalog attribute-based access control support that requirement? (Select TWO.)

  • AA policy created at catalog or schema level covers every table in that scope whose columns carry the governed tag, so tables created later fall under it without the policy being edited. Correct
  • BAssigning a governed tag to a column rewrites the stored values in place, so the sensitive figures leave the table files and no reader can reach them through any query path afterwards.
  • CA governed tag is defined centrally with a permitted set of values and controlled assignment, so the policy can target that tag instead of naming a column in every table it protects. Correct
  • DThe policy supersedes the SELECT privilege on the tables it covers, so a reader matched by it stops needing a grant on the catalog and the schema before running a query.
  • EEach table owner has to attach the shared masking function to the tagged column with a statement on that table before the central policy begins to take effect on it.
Unity Catalog attribute-based access control binds a masking rule to a governed tag and a scope, so tagged tables are covered centrally. An attribute-based access control policy names a scope and a governed tag rather than individual tables, so Unity Catalog evaluates it against any column in that scope carrying the tag. Governed tags supply a controlled vocabulary, which is what lets one policy definition follow the classification across many tables and across tables that do not exist yet.

Why A is correct: Correct. An attribute-based access control policy is bound to a scope and to a tag rather than to a fixed list of tables, so a later table whose column carries the tag is governed by the same policy definition.

Why B is wrong: Tempting because tagging feels like an action on the data, but a tag is metadata on the column. Masking is applied when a query reads the column, and the stored values are left as they are.

Why C is correct: Correct. Governed tags give the vocabulary the policy matches on, so the mask follows the classification rather than a hand maintained list of table and column names.

Why D is wrong: Tempting because both features gate what a reader sees, but the privilege model still applies. A reader lacking SELECT with USE CATALOG and USE SCHEMA cannot query the table at all, policy or no policy.

Why E is wrong: Tempting because that is exactly how a per table column mask works, and it is the effort the team is trying to escape. A tag driven policy removes that per table step.

See more Data-Engineer-Associate practice questions, answers explained.

Exam traps in Governance and Security

Answers that look right on this material and are not. Each one is a distractor from a different question in the Data-Engineer-Associate bank for this domain.

  • It applies to any column whose name matches a pattern supplied when the policy is written, so the teams must first rename their columns to one agreed convention.

    Why it is wrong: Tempting because naming conventions are a familiar way to standardise schemas, but the policy keys off an assigned tag, and forcing a rename across several teams is exactly the manual work ABAC removes.

  • Analysts still have to be pointed at a separate object, because the policy publishes a governed view for each tagged table and withdraws access to the table sitting behind it.

    Why it is wrong: Tempting because it mirrors the arrangement being replaced, but the policy adds no second object. Readers keep using the table name they already query.

  • The masking is carried out by the SQL warehouse, so a notebook or a Lakeflow Jobs task that reads the same table by name receives the underlying values in the clear.

    Why it is wrong: Tempting because analysts usually meet the masked column in a warehouse, but enforcement sits in Unity Catalog rather than in one compute product, so every governed read path is treated the same way.

Examworthy is not affiliated with or endorsed by Databricks. Original, blueprint-aligned practice material only.