Skip to content
Cheat sheets

Data Engineer Professional · Domain 7: Data Governance · 5% of the exam

Task 7.1: Demonstrate understanding of Unity Catalog tags and comments as mechanisms for adding metadata to securable objects to improve data discoverability.

Tags and governed tags, COMMENT ON and AI-generated comments, the privileges to set them, searching by tag, and the information schema views.

Study it

  • Tags and comments for discovery

    Lesson coming

Sample questions

Try each one before opening the answer. Every option is explained, with the Databricks documentation page that proves it.

Question 1 · choose 2

A data steward plans to tag the table sales.orders and 30 of its columns with domain and sensitivity tags to make them easier to find. Which TWO statements about these tags are correct? (Choose TWO.)

  1. ATag keys are case-insensitive: Domain and domain are one tag
  2. BA single table or column can carry at most 50 tags
  3. CA tag key such as data-domain is valid
  4. DTag values can hold up to 4,096 characters
  5. EEach column must be tagged with its own ALTER TABLE statement
Show the answer and why

  • ATag keys are case-insensitive: Domain and domain are one tag

    Incorrect

    Tag keys are case-sensitive.

  • BA single table or column can carry at most 50 tags

    Correct

    Unity Catalog allows a maximum of 50 tags on a single securable object, such as a table or column.

  • CA tag key such as data-domain is valid

    Incorrect

    Hyphens are among the characters not allowed in tag keys.

  • DTag values can hold up to 4,096 characters

    Incorrect

    Tag keys and values are limited to 256 characters.

  • EEach column must be tagged with its own ALTER TABLE statement

    Correct

    Tags cannot be assigned to several columns in one ALTER TABLE command, unlike comments.

Tags are key and optional value pairs on securable objects that improve search and discovery. Their constraints include case-sensitive keys, 50 tags per object, 1,000 column tags per table, 256-character keys and values, banned characters in keys and one column per tagging command.

Databricks documentation

Task 7.1Report a problem with this question

Question 2 · choose 1

A data steward who does not own main.sales.orders must add tags to it. Which privileges do they need?

  1. ASELECT on the table, with USE SCHEMA and USE CATALOG
  2. BAPPLY TAG on the table plus USE SCHEMA and USE CATALOG
  3. CMODIFY on the table, with USE SCHEMA and USE CATALOG
  4. DBROWSE on the catalog and USE SCHEMA on the schema
Show the answer and why

  • ASELECT on the table, with USE SCHEMA and USE CATALOG

    Incorrect

    Reading data does not allow changing tags.

  • BAPPLY TAG on the table plus USE SCHEMA and USE CATALOG

    Correct

    Non-owners need APPLY TAG on the object and usage on the parent schema and catalog.

  • CMODIFY on the table, with USE SCHEMA and USE CATALOG

    Incorrect

    MODIFY lets users change the table's data, not its tags.

  • DBROWSE on the catalog and USE SCHEMA on the schema

    Incorrect

    BROWSE lets users see metadata, not change it.

To add tags, you must own the object or have APPLY TAG on it plus USE SCHEMA and USE CATALOG on its parents. Governed tags also need ASSIGN on the tag.

Databricks documentation

Task 7.1Report a problem with this question

Question 3 · choose 1

A data engineer who does not own the table main.sales.orders wants to add descriptions to its columns. Which privileges let them edit the comments?

  1. AMODIFY and SELECT on the table, plus usage on its parents
  2. BAPPLY TAG on the table and on its columns
  3. CBROWSE on the catalog, with USE SCHEMA on its schema
  4. DREAD METADATA on the table, with usage on its parents
Show the answer and why

  • AMODIFY and SELECT on the table, plus usage on its parents

    Correct

    Table and column comments need ownership or MODIFY and SELECT on the table, plus usage on its parents.

  • BAPPLY TAG on the table and on its columns

    Incorrect

    That privilege does not cover editing comments.

  • CBROWSE on the catalog, with USE SCHEMA on its schema

    Incorrect

    BROWSE allows viewing comments, not editing them.

  • DREAD METADATA on the table, with usage on its parents

    Incorrect

    Reading metadata does not allow changing it.

Editing comments on a table or column requires ownership, or MODIFY and SELECT on the table with USE CATALOG and USE SCHEMA on its parents.

Databricks documentation

Task 7.1Report a problem with this question

Question 4 · choose 1

A steward with MANAGE on a materialized view tries to add a description to it and is refused. Why?

  1. AComments are not supported on materialized views at all
  2. BThe steward also needs APPLY TAG on the view
  3. CComments can be added only from a classic cluster, not a warehouse
  4. DOnly the owner can edit the comment of a materialized view
Show the answer and why

  • AComments are not supported on materialized views at all

    Incorrect

    Materialized views can carry comments.

  • BThe steward also needs APPLY TAG on the view

    Incorrect

    That privilege does not cover a view's comment.

  • CComments can be added only from a classic cluster, not a warehouse

    Incorrect

    Any Unity Catalog compute, including SQL warehouses, can add comments.

  • DOnly the owner can edit the comment of a materialized view

    Correct

    For views and materialized views, MANAGE is insufficient to edit comments.

For catalogs, schemas and many other objects, owners or users with MANAGE can edit comments. For views and materialized views, only the owner can.

Databricks documentation

Task 7.1Report a problem with this question

Question 5 · choose 1

A team wants to tag each customer table with the name and phone number of the customer's account manager for discovery. What does Databricks warn?

  1. ATags are encrypted, so personal data in them is safe
  2. BTags are visible only to the table owner and metastore admins
  3. CTags are plain text and may be replicated globally; keep PII out
  4. DTags with contact details expire after 30 days
Show the answer and why

  • ATags are encrypted, so personal data in them is safe

    Incorrect

    Tag data is stored as plain text.

  • BTags are visible only to the table owner and metastore admins

    Incorrect

    Tags support search and discovery by other users.

  • CTags are plain text and may be replicated globally; keep PII out

    Correct

    Tag names, values and descriptors should not contain personal or sensitive information.

  • DTags with contact details expire after 30 days

    Incorrect

    Tags do not expire.

Tag data is stored as plain text and may be replicated globally, so tags must not contain personal or sensitive information.

Databricks documentation

Task 7.1Report a problem with this question

Practice domain 7 →Practice all domains →