What causes LAKEHOUSE_FEDERATION_CONNECTION_FAILED?

The secret scope or secret key referenced in the connection definition does not exist or has been rotated. The remote database host is unreachable from Databricks — firewall rule or private-link misconfiguration. Wrong port, account name, or database specified when the connection was created. The remote user account lacks login privileges on the target database

How do I fix LAKEHOUSE_FEDERATION_CONNECTION_FAILED?

Step 1: In Databricks, run SHOW CONNECTIONS and verify the host, port, and user for the failing connection.. Step 2: Test network reachability from a cluster notebook: use socket or telnet to confirm the target host and port respond.. Step 3: Check the secret scope referenced by the connection — use dbutils.secrets.get to confirm the value is correct.. Step 4: Re-create the connection with updated credentials: DROP CONNECTION ; then CREATE CONNECTION with corrected parameters.. Step 5: If using private connectivity, verify the Databricks network configuration includes the target CIDR range.

High severityconnection

Power BI Refresh Error:
LAKEHOUSE_FEDERATION_CONNECTION_FAILED

What does this error mean?

Databricks cannot establish a connection to the foreign data source registered in Lakehouse Federation, typically due to wrong credentials, unreachable host, or a missing network path.

Common causes

1The secret scope or secret key referenced in the connection definition does not exist or has been rotated
2The remote database host is unreachable from Databricks — firewall rule or private-link misconfiguration
3Wrong port, account name, or database specified when the connection was created
4The remote user account lacks login privileges on the target database

How to fix it

1Step 1: In Databricks, run SHOW CONNECTIONS and verify the host, port, and user for the failing connection.
2Step 2: Test network reachability from a cluster notebook: use socket or telnet to confirm the target host and port respond.
3Step 3: Check the secret scope referenced by the connection — use dbutils.secrets.get to confirm the value is correct.
4Step 4: Re-create the connection with updated credentials: DROP CONNECTION <name>; then CREATE CONNECTION with corrected parameters.
5Step 5: If using private connectivity, verify the Databricks network configuration includes the target CIDR range.

Frequently asked questions

Can I update credentials without dropping the connection?

Yes — update the secret value in the secret scope and the connection will pick up the new credential on the next query, because Databricks resolves the secret at query time, not at connection creation time.

Does Lakehouse Federation support all external databases?

As of 2026, supported sources include Snowflake, MySQL, PostgreSQL, SQL Server, Oracle, Redshift, and BigQuery. Each has its own driver configuration requirements.