Databricks AWS and Power BI: Power BI Identity Passthrough

Question

In an attempt to simplify access control, I want to use Databrick's Unity Catalog for central governance. Essentially, I want Power BI to passthrough the user's identity and for Databricks to return the appropriate results for the user (e.g. user has access to a dynamic view that filters the results based on their identity `[email protected]`). I am referencing this Databricks blog where there is a screenshot of a checkbox that reads, "Report viewers can only access this data source with their own Power BI identities using DirectQuery." When setting up a connection for a semantic model that uses DirectQuery, I don't see this checkbox. Is there a setting in the Power BI service that needs to be enabled in order to see this in the UI? Or perhaps some other path needs to be followed to see this option

anonymous · Answer

Thanks&nbsp;GilbertQ&nbsp;. DirectQuery is a requirement and I have already published a dataset that uses DirectQuery. The problem I am trying to solve is once the semantic model is in the Power BI service that the identity of the user using the model is passed to Databricks (identity passthrough). And I have seen a few pieces of documentation where the checkbox seen in the image below is referenced. And I am trying to figure out how to get that checkbox to appear in the Power BI service user interface.&nbsp;&nbsp;

gilbertq · Answer

Hi&nbsp;data_chef&nbsp;
&nbsp;
As far as I'm aware, in order to enable direct query, you would have to create this using power BI desktop

gilbertq · Answer

Hi&nbsp;Anonymous&nbsp;
&nbsp;
As far as I am aware that is currently not possible in the service as far as I am aware. You can create or vote for the idea here:&nbsp;https://aka.ms/FabricIdeas&nbsp;

anonymous · Answer

Thank you&nbsp;GilbertQ&nbsp;
Hi,&nbsp;Anonymous&nbsp;
Based on the screenshots you provided, it looks like you're setting the privacy level of your data source, which you can find in a few places:
When adding a gateway connection:

When creating Dataflows using databricks:

In short, the privacy level is set by the report author, not by the report reader.&nbsp;As described in the article you provided, he's editing the data source credentials:

You can edit the credentials for a semantic model if you select it:

&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
Best Regards
Jianpeng Li
If this post helps, then please consider Accept it as the solution&nbsp;to help the other members find it more quickly.
&nbsp;
&nbsp;
&nbsp;