This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
Hello,
I have a composite model with heriachical RLS applied. The RLS works perfectly fine except when it comes to using data from the direct query table. I have mutliple tables that are all import and 1 single direct query table.
When using data from the direct query model in a table of data, it totals all values instead of just the values specific to the user viewing per the RLS.
For example, the user will be viewing data for their team. Each user within their team shows the correct number of items however the table total shows the total for all values, not just the values relevant to the user viewing.
In this example, I am viewing as a particular user and you can see the count for each user does not equal the total sum at the bottom of the table.
Any ideas as to what is going on here?
Thanks,
Monique
Hi @moniqueg ,
Can you provide the specific formula corresponding to the work item created? As per your description, it is possible that the formula uses something like the all function, which clears the filter criteria for the user column.
Best Regards,
Adamk Kong
Edit: User field comes from an import table while work items come from direct query. There is a user name field I can use in the direct query however the RLS does not apply to this at all - just shows all users irrespective of who is viewing. I tried created an inactive relationship between direct query table and RLS user table in conjunction with a measure using use relationship. This did not change the outcome.
Check out the April 2026 Power BI update to learn about new features.
Sign up to receive a private message when registration opens and key events begin.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
| User | Count |
|---|---|
| 35 | |
| 32 | |
| 25 | |
| 23 | |
| 16 |
| User | Count |
|---|---|
| 65 | |
| 50 | |
| 30 | |
| 24 | |
| 23 |