Databricks: Unable to Create Supervisor Agent Despite Enabling Supervisor API (Beta) — Feature Reports Unavailable

Veera-3402 0 Reputation points
2026-10-03T04:17:52.0133333+00:00

Problem description

I am trying to create and save a Supervisor Agent in my Azure Databricks workspace after enabling the Supervisor API (Beta). However, when I attempt to save the agent, I receive an error message stating: "Failed to save agent. Supervisor Agent is not available. To access this feature, please contact sales for assistance." Despite confirming that the Supervisor API (Beta) is enabled and showing as effective in the workspace settings, the save operation continues to fail with the same message.

Environment

The affected environment is an Azure Databricks workspace in East US, with Unity Catalog and serverless features enabled. The Supervisor API (Beta) is enabled and shows as effective. No additional workspace configuration, identity model, or entitlement details are documented.

What I've already tried

I confirmed that the workspace is in East US, Unity Catalog is enabled, and the Supervisor API (Beta) is active and effective. I also verified that agent-related preview features are enabled in the workspace. I attempted to create and save the Supervisor Agent multiple times after enabling the feature, but the operation still results in the same error. Additionally, I checked Azure diagnostics for platform issues and the Activity Log for workspace-specific issues, but no relevant problems were identified. I have not tried any other troubleshooting steps beyond these checks.

Current status

The issue persists despite verifying the workspace configuration and enabling the preview feature. I am seeking guidance on further validation steps to determine if the Supervisor Agent capability is available for my workspace or if there are account-level or region-specific rollout considerations. I would appreciate assistance in identifying the root cause and resolving the issue.

Azure Databricks
Azure Databricks

An Apache Spark-based analytics platform optimized for Azure.


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.