Configure OAuth U2M for Confluence ingestion

This page describes how to configure OAuth user-to-machine (U2M) authentication for Confluence ingestion into Azure Databricks.

Prerequisites

To ingest from Confluence, you need the following:

• A Confluence account with at least View access to the spaces and pages you want to ingest. During OAuth U2M setup, you authorize with this account, and the connector can only read content this account is permitted to view.
• Access to create OAuth applications in your Confluence instance, or permission to edit an existing OAuth app to add the required scopes.
• An OAuth app that grants the required Confluence API scopes. You add these scopes during setup.

Configure Confluence for OAuth

1. Visit https://developer.atlassian.com/console/myapps/.

2. Optionally click Create > OAuth 2.0 integration. To use an existing OAuth app instead, add the required scopes.

3. Provide a name for the integration, then click Create.

4. Visit Permissions > Confluence API (under API scopes) > Add.

5. Click Granular Scopes > Edit Scopes, then add the following scopes to the OAuth app:

   • read:page:confluence
   • read:attachment:confluence
   • read:space:confluence
   • read:label:confluence
   • read:configuration:confluence

6. Go to Authorization and input the redirect URI as follows:

   https://<databricks_workspace_URL>/login/oauth/lakehousefederation.html

   For example:

   https://my-databricks-account.cloud.databricks.com/login/oauth/lakehousefederation.html

7. Go to Settings and record the client ID and the client Secret. You’ll need this to create a Unity Catalog connection.

Next steps

An admin can now use Catalog Explorer to create a connection so that non-admins can create pipelines.