Can't refresh semantic model in Service

[joelmsherman]

Hello,

I have a semantic model in one workspace that is connected to a SQL endpoint of a Fabric Lakehouse in another workspace via import mode.  Note: this is NOT a direct lake connection!

When trying to refresh the semantic model in the Service, I get the following error:

"We cannot refresh this semantic model because this semantic model uses a default data connection without explicit connection credentials. Please replace the default data connection in the semantic model settings with an explicit cloud or gateway data connection."

When when I try to change the Default connection settings from SSO

I get the following New Connection modal:

If I choose an Authentication Method of "Basic", it asks me for a Username and Password.  My Power BI username and password don't work here.

Any help or advice here is much appreciated.

[v-pnaroju-msft]

Hi @joelmsherman,

We are happy to learn that your issue has been resolved. To help others in the community, kindly mark the response that resolved your query as the accepted solution.

For any further assistance or questions, please continue to engage with the Fabric Community.

Best regards,

Pavan

[rajendraongole1]

Hi @joelmsherman - The error indicates that the Power BI Service is unable to refresh the semantic model because the connection to the Fabric Lakehouse SQL endpoint is not properly configured with explicit credentials.

Step 1: Understand the Problem
Default Data Connection Issue: The semantic model uses the default connection settings, which rely on credentials not explicitly provided in the cloud or via a gateway.
Authentication Method: When selecting "Basic" authentication, you need the correct credentials to connect to the SQL endpoint.
Step 2: Verify the SQL Endpoint Credentials
Fabric Lakehouse SQL Endpoint:
By default, Fabric SQL endpoints use Azure Active Directory (AAD) authentication for access.
If the "Basic" option is asking for a username and password, ensure you are not using your Power BI username/password but rather the credentials for an AAD user with access to the Fabric workspace.
Test Locally:
Use a SQL client like SQL Server Management Studio (SSMS) or Azure Data Studio to connect to the SQL endpoint using the same credentials. This will confirm whether the credentials are correct.
Step 3: Configure the Semantic Model Connection
Go to the Power BI Service.
Open the Settings for the dataset/semantic model.
Under Data source credentials, configure the connection:
Authentication Method: Use OAuth2 (recommended) or AAD account if available.
If OAuth2 is not supported, use a Basic authentication method only if the SQL endpoint has a dedicated username and password configured.
Step 4: Set Up a Gateway (if needed)
If the SQL endpoint is not accessible directly via the cloud, you might need to configure a Data Gateway:

Install and configure an On-premises Data Gateway.
Add the data source for the Lakehouse SQL endpoint in the gateway settings.
Configure the Power BI dataset to use the gateway.
Step 5: Test and Refresh
After updating the credentials, trigger a manual refresh in the Power BI Service.
Monitor for any errors during the refresh.

Please check the above detailed information.

[joelmsherman]

OAuth 2.0 worked, after confirming my 365 credentials.  

I'm not sure why Microsoft makes this so complicated; If I connect my Microsoft semantic model to a Microsoft Fabric Lakehouse, one shouldn't have to jump through these hoops to make it all work in the Service.  Ah, the love/hate relationship with these tools!