Generic PostgreSQL Connector Does Not Support Scan Rule Set Selection and Automatic Classification in Microsoft Purview

Brishti Saha 0 Reputation points
2026-06-11T08:52:06.6766667+00:00

We are using Microsoft Purview Data Map and have registered a PostgreSQL data source using the Generic PostgreSQL connector (NeonDB).

Environment

  • Microsoft Purview
  • Data Source Type: PostgreSQL (Generic Connector)
  • Source Database: NeonDB (PostgreSQL)
  • Custom classifications created successfully
  • Scan completes successfully
  • Assets and columns are discovered successfully

Issue Description

We are unable to apply automatic classifications to PostgreSQL assets because the scan creation wizard does not provide an option to select a Scan Rule Set.

Expected Behavior

When creating a scan, we expect to see a step allowing selection of:

  • System Scan Rule Set
  • Custom Scan Rule Set containing built-in and custom classifications

This behavior is available for SQL Server scans.

Actual Behavior

For the Generic PostgreSQL connector:

  1. Credential is selected.
  2. Scan trigger is configured.
  3. Wizard immediately moves to the Review screen.
  4. No Scan Rule Set selection page is displayed.

As a result, we cannot associate any Scan Rule Set with the PostgreSQL scan.

Additional Finding

When creating a new Scan Rule Set under:

Data Map → Source Management → Scan Rule Sets

and searching for PostgreSQL-related data source types, only the following options are available:

  • Azure Database for PostgreSQL
  • Amazon RDS (PostgreSQL) (Preview)

There is no option for Generic PostgreSQL.

Our registered source uses the Generic PostgreSQL connector.

This raises the question of whether:

  • Scan Rule Sets are supported for Generic PostgreSQL.
  • Custom classifications are supported for Generic PostgreSQL.
  • Automatic classification is expected to work for Generic PostgreSQL sources.

Impact

Although scans complete successfully and assets are discovered:

  • No Scan Rule Set can be attached.
  • Built-in classifications are not applied.
  • Custom classifications are not applied.
  • Column-level classifications remain empty after rescanning.

This prevents us from demonstrating automated data classification and governance capabilities for PostgreSQL assets.

Questions for Product Team

  1. Is Scan Rule Set selection supported for the Generic PostgreSQL connector?
  2. Are built-in classifications supported for Generic PostgreSQL scans?
  3. Are custom classifications supported for Generic PostgreSQL scans?
  4. Is the absence of Generic PostgreSQL in the Scan Rule Set creation page expected behavior?
  5. If supported, what configuration is required to enable Scan Rule Set selection and automatic classification?

Screenshots Attached

  1. Generic PostgreSQL missing from Scan Rule Set creation options.

User's image

  1. PostgreSQL scan wizard skipping Scan Rule Set selection.

User's image

  1. PostgreSQL scan review page.

User's image

  1. SQL Server scan wizard showing Scan Rule Set selection.

User's image

  1. PostgreSQL asset with no classifications.

image(240).png

  1. PostgreSQL columns showing empty column-level classifications.

image(241).png

Microsoft Security | Microsoft Purview
0 comments No comments

2 answers

Sort by: Most helpful
  1. Manoj Kumar Boyini 18,355 Reputation points Microsoft External Staff Moderator
    2026-07-04T10:10:55.93+00:00

    Hi @Brishti Saha

    The behavior you're seeing is expected for the PostgreSQL connector.

    The supported capabilities matrix indicates that the generic PostgreSQL data source supports metadata extraction (servers, databases, schemas, tables, columns, and views) and lineage, but does not support automatic classifications. Because of this:

    • The scan wizard does not provide a Scan Rule Set selection step.
    • Generic PostgreSQL is not available when creating Scan Rule Sets under Data Map → Source Management → Scan Rule Sets.
    • Built-in and custom classifications are not automatically applied during scans for the generic PostgreSQL connector, even though the scan successfully discovers assets and columns.

    Your screenshots show that metadata discovery is working as expected, and the absence of Scan Rule Set selection and automatic classifications is consistent with the current documented capabilities.

    If this capability is important for your scenario, we recommend submitting a feature request through the Microsoft Feedback portal so the Product Group can evaluate adding automatic classification and Scan Rule Set support for the generic PostgreSQL connector in a future release.

    References:

    Please let us know if you have any questions or need further assistance.

    If this solution helped resolve your issue, please consider clicking ‘Accept Answer’ or giving it an upvote to help others find it easily.

    Was this answer helpful?

    0 comments No comments

  2. AI answer

    2026-06-11T08:52:39.11+00:00

    Scan rule set selection and automatic classification are supported only for PostgreSQL sources that are explicitly listed as supported data sources in Microsoft Purview, such as Azure Database for PostgreSQL. The documented PostgreSQL connector supports:

    • Scanning PostgreSQL versions 8.x to 16.x
    • Extracting technical metadata (server, databases, schemas, tables, views, columns)
    • Fetching static lineage between tables and views

    The documentation for this connector does not describe support for:

    • Associating scan rule sets with a Generic PostgreSQL connector
    • Applying system or custom classifications during scans for Generic PostgreSQL

    Automatic classification in Microsoft Purview works by:

    1. Registering a supported data source.
    2. Creating and associating a scan rule set with that source.
    3. Running a scan so that the selected scan rule set’s classifications are applied to assets.

    This flow is documented generically and relies on the source-specific article to define whether scan rule sets and classifications are available for that source. For PostgreSQL, the documented connector is Azure Database for PostgreSQL, not a generic PostgreSQL connector.

    Classification behavior and scan rule sets:

    • Scan rule sets define which system and custom classifications are evaluated during a scan.
    • Custom classifications are not included in any default scan rules; a custom scan rule set must be created and used in the scan for automatic assignment.
    • Automatic classification applies to columns (table assets themselves are not automatically classified), and requires sufficient distinct data values in each column.

    Given the context:

    1. Scan rule set selection is documented for supported sources like Azure Database for PostgreSQL, not for a Generic PostgreSQL connector.
    2. Built-in classifications and custom classifications are applied only when a scan rule set is associated with the scan; this is tied to supported source types.
    3. Custom classifications require a custom scan rule set that includes them and a source type that supports scan rule sets.
    4. The absence of a “Generic PostgreSQL” option when creating scan rule sets is expected based on the documented list of supported data sources.
    5. For PostgreSQL classification to work as documented, the source must be one of the supported PostgreSQL data sources (for example, Azure Database for PostgreSQL) and a scan rule set must be selected when configuring the scan.

    For PostgreSQL assets where automatic classification is not available (such as Generic PostgreSQL in this scenario), classifications can still be applied manually in the Purview governance portal. Manually applied classifications are retained in subsequent scans and are not removed even if classification rules no longer apply.


    References:

    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.