Cloud Connection - Semantic Model

[Mikerhode_44]

Here is my current problem. I created a semantic model off of a Fabric Data Warehouse table. I created a connection on power bi via Fabric / Semantic model. When I look at the cloud settings on the Semantic model from DW vs power bi semantic model for the same table they are different and this is the root cause for my error and why it's not syncing. If I set the Maps to: Personal Cloud Connection it doesn't give me the OnPremiseServiceException error but would the semantic model still refresh?

Thanks,

Mike

[v-sshirivolu · Accepted solution]

Hi @Mikerhode_44 ,
Thanks for reaching out to Community Forum.

The difference you’re seeing in the cloud connection types is expected. In Microsoft Fabric, the semantic model from the Data Warehouse shows Azure Data Lake Storage because that reflects how the data is physically stored in OneLake. When you connect to the same model through Power BI, it appears as an Analysis Services endpoint because that’s the layer Power BI uses to query the data. So even though it’s the same data, the connection types do not need to match. For the issue itself, I recommend checking your setup in Power BI Service. Make sure the semantic model is using an Organizational account OAuth2 under Data source credentials, and confirm it is mapped to a cloud connection and not an on-premises gateway, since Fabric sources do not require a gateway. If the issue still continues, you may try re-authenticating or rebinding the dataset to refresh the connection. Also, if you are using the SQL endpoint, keep in mind it may not update instantly, so you may consider using Direct Lake mode if available, as it avoids refresh and sync delays.

Thank you.

[Mikerhode_44]

Thank you very much. Updating the settings worked. I also added the Entra ID / single sign on check and it resolved everything. 

[lbendlin]

this is the root cause for my error and why it's not syncing

I assume you are aware that the SQL Endpoint is not synchronizing from the Datawarehouse in real time, right? It can take many minutes before changes in the DW are reflected in the SQL Endpoint.

[Mikerhode_44]

This has been going on for a while. Where it's puzzling is the semantic model from the Fabric Data Warehouse shows Cloud connection data source as AzureDataLakeStorage, where the Cloud connection to power bi for the same semantic model as Analysis Services server. Shouldn't they be the same?