Fabric OneLake

Write Delta or Iceberg tables to a Microsoft Fabric Lakehouse via OneLake.

Supported features

Feature	Status	Remarks
Read/write Delta tables	✅	Fabric discovers them in the Lakehouse automatically
Read/write Iceberg tables	✅	Through OneLake's Iceberg REST catalog
Table descriptions in OneLake Catalog	🟡	Only supported for Delta tables
Column descriptions in OneLake Catalog	❌️
Read/write views	❌️

Prerequisites

• A Fabric workspace on a Fabric capacity (F or P SKU, or a Fabric trial). The capacity must be running: OneLake is unavailable while it's paused.
• A Lakehouse in that workspace with schemas enabled, which is the default for new Lakehouses.

Installation

pip install 'dbt-polars[fabric]'          # Delta tables
pip install 'dbt-polars[fabric,iceberg]'  # Delta and Iceberg tables

Profile configuration

Example

my_project:
  target: dev
  outputs:
    dev:
      type: polars
      catalog_type: fabric
      schema: dev
      workspace: my_workspace
      lakehouse: my_lakehouse
      default_table_format: delta

To combine this catalog with others, put the same parameters in an entry of catalogs, with type: fabric and a name, see Catalog overview.

Parameter Reference

The following parameters are supported, in addition to the common profile parameters:

• catalog_type - (Required) Must be fabric.

• workspace - (Required) Name or ID of the Fabric workspace.

• lakehouse - (Required) Name or ID of the Lakehouse. It must have schemas enabled, which is the default for new Lakehouses. Use IDs when a name contains spaces or special characters.

• schema - (Required) Default schema for models. dbt-polars creates missing schemas. The Lakehouse's default schema dbo can't be dropped.

• default_table_format - (Optional) Format of new tables: delta or iceberg. Defaults to delta. Models can override it by setting table_format.

• Authentication parameters - (Optional) Passed to azure.identity.DefaultAzureCredential, which tries environment variables, workload identity, the Azure CLI, managed identity and more, in that order. Locally, az login is enough. Delta and Iceberg tables use the same credentials. Commonly used:

  • exclude_managed_identity_credential - Skip managed identity, e.g. on a machine where it isn't the identity you want.
  • managed_identity_client_id - Client ID of a user-assigned managed identity.
  • exclude_interactive_browser_credential - Set to false to allow a browser login.

  For a service principal, set AZURE_CLIENT_ID, AZURE_CLIENT_SECRET and AZURE_TENANT_ID in the environment.

Model configuration

The following custom model parameters are available for Fabric:

• table_format - (Optional) delta or iceberg. Overrides the profile's default_table_format. When the existing table has another format, a table model recreates it and an incremental model requires --full-refresh.

  {{ config(materialized='table', table_format='iceberg') }}

Required permissions

The principal dbt-polars authenticates as needs these permissions:

Permission	On	Needed for
Read item permission, e.g. the Viewer workspace role or a shared Lakehouse	the Lakehouse	Accessing the Lakehouse at all
Read OneLake security role	the schemas or tables dbt-polars reads	Reading tables, e.g. sources
ReadWrite OneLake security role	the Lakehouse's Tables folder	Creating schemas
ReadWrite OneLake security role	the schemas dbt-polars writes to	Writing tables

Granting the Contributor workspace role, or higher, covers all of these.