PostgreSQL metadata collector

  • Release version: Australia
  • Updated March 12, 2026
  • 2 minutes to read

    • Provides read-only access to metadata from a PostgreSQL database.

    The collector harvests metadata for PostgreSQL tables and columns, making them searchable and discoverable in the data catalog. Supports both self-hosted PostgreSQL instances and managed PostgreSQL instances, such as those hosted on AWS RDS.

    Authentication supported

    Username and password authentication and AWS IAM authentication.

    Metadata cataloged

    The PostgreSQL collector catalogs the following information.

    Note:
    All versions of overloaded functions and stored procedures are cataloged. Each version has its own title in the catalog but a distinct identifier.
    Table 1. Cataloged metadata
    Object Information cataloged
    Columns Name, Description, JDBC type, Column Type, Is Nullable, Default Value, Key type (Primary, foreign), column size, column index
    Table Name, description, primary key, schema
    Table Index Index Cardinality, Column name, Index Type, Index Name, is non Unique, Ordinal Position, Pages, Sort Sequence
    Views Name, description, SQL definition
    Materialized View Name, description, SQL definition
    Schema Identifier, Name
    Database Type, name, identifier, server, port, environment, JDBC URL
    Function Name, Description, Function Type
    Stored Procedure Name, Description, Stored Procedure Type
    When profiling and sampling parameters are enabled, the following additional column information is cataloged:
    Note:
    The user/role must have read access to data to be able to harvest profiling information (column statistics).
    Table 2. Profiling and sampling specific information
    Object Information cataloged
    Column
    • Average Length (sample)
    • Average Value (sample)
    • Data Distribution
    • Distinct Values
    • Estimated Distinct Values
    • Estimated Non-null Values
    • Maximum Length (sample)
    • Maximum Value (sample) sorted numerically or alphabetically (z-a)
    • Minimum Length (sample)
    • Minimum Value (sample) sorted numerically or alphabetically (a-z)
    • Non-null Values (sample)
    • Sample String Values (first 5 items in a column)
    Table
    • Row Count
    • Sample Count (Target sample size)

    Relationships between objects

    Catalog pages show relationships between the following data asset types:

    Table 3. Relationships between harvested data asset pages
    Data asset page Relationship
    Table Columns, Table Indexes
    Columns Table
    Schema
    • Database that contains Schema
    • Table that is part of Schema
    • View that is part of Schema
    • Materialized View that is part of Schema
    View
    • Schema that contains Views
    • Columns that are part of Views
    Materialized View
    • Schema that contains Materialized Views
    • Columns that are part of Materialized Views

    Lineage for PostgreSQL

    The collector identifies column-level lineage for Views and Materialized Views, tracing data sources from view columns to source table columns across SQL expressions and subqueries.

    Establishes relationships between Views and source Table columns that:

    • Sort rows (SQL ORDER BY)
    • Filter rows (SQL WHERE and HAVING clauses)
    • Aggregate rows (SQL GROUP BY)
    Note:
    View-to-table relationships can be established transitively through column-level relationships. Variable statement lineage isn’t supported.