MetricSign
EN|NLRequest Access
High severityresource

Power BI Refresh Error:
DatabricksServerlessUnavailableError

What does this error mean?

The dbt-databricks adapter cannot connect because the configured Serverless SQL warehouse is unavailable, over capacity, or the workspace does not have Serverless compute enabled.

Common causes

  • 1The Serverless SQL warehouse is disabled or the workspace has not been enrolled in the Serverless preview
  • 2The workspace-level serverless compute budget is exhausted
  • 3The http_path in the dbt profile points to a non-existent or deleted Serverless warehouse
  • 4The service principal lacks CREATE or USE permission on Serverless warehouses

How to fix it

  1. 1Step 1: Verify the Serverless warehouse exists in the Databricks SQL Warehouses UI and is in Running or Idle state.
  2. 2Step 2: Confirm the http_path in the dbt profile matches the HTTP Path shown in the warehouse Connection Details.
  3. 3Step 3: If Serverless is not available in the workspace, switch to a Pro or Classic warehouse as a fallback.
  4. 4Step 4: Check workspace-level compute quota and request an increase if serverless capacity is exhausted.
  5. 5Step 5: Run dbt debug to validate the warehouse connection from the dbt profile configuration.

Frequently asked questions

Do Databricks Serverless SQL warehouses auto-start like Pro or Classic warehouses?

Yes — Serverless warehouses start on demand and typically become available in under 5 seconds. If a Serverless warehouse is unavailable, it usually indicates a configuration or quota issue rather than a cold-start delay.

How do I switch my dbt-databricks profile from Serverless to a Classic warehouse?

Update the http_path in your dbt profiles.yml to the HTTP Path of the Classic or Pro warehouse. Run dbt debug to confirm connectivity before running production jobs.

Other resource errors