Hi, I added the ADF Managed Identity to the fabric workspace, but the status shows as "User blocked." The same Managed Identity works fine with the Dev workspace. I'm not sure what the issue could be. Could you please help me with this?     ... View more

Power BI currently exports visual data primarily in CSV format. When these files are opened in Excel, text fields containing commas or line breaks may cause column misalignment due to delimiter interpretation. Although the dataset is correct, this behavior leads to structural misinterpretation of data in Excel, especially in reporting scenarios. Suggested improvement: Enable direct export to Excel (.xlsx) format Or enforce proper text encapsulation in CSV exports (e.g., double quotes) This would ensure reliable data export and prevent misalignment issues in Excel-based workflows. ... View more

The Filters pane selection area seems to be broken - you can select the top half of the arrow but not the bottom half, but then you can select in a small white space just below - really annoying when you go to click to expand filters and nothing happens... ... View more

Often times when provisioning Workspace, Lakehouse or OneLake security, it takes time to be applied.  It would be wonderful if there was a status that said (In progress, Success, Failed).  This would allow data security testing to be more effective.  Currently we wonder if the security is still in progress or if our administering of the secuirty was not effective. ... View more

I'm not sure this was even suggested but I searched and somehow this hasn't come up.   I'm just suggesting there should be a way to export a PowerBI data model directly to an ERD Diagram in the form of a Visio .vsdx file.   It just seems so sensible since they're both Microsoft products! Maybe there is a way but I just haven't figured it out yet?   Thanks! ... View more

Scenario: I want to build complex SQL transformations on the data in my evenstream. Currently, I am using Confluent and building Flink SQL on a Kafka topic. I can create views to encapsulate complex logic which is then computed on the fly. This enables me to syntactically split up complex transformations in more managable, multi-step operations (SQL CREATE VIEW Statement in Confluent Cloud for Apache Flink | Confluent Documentation).   Problem: In Fabric Eventstreams, I can't chain SQL operators to achieve the same behaviour. This is specifically listed in limitations here: Process Events Using a SQL Operator - Microsoft Fabric | Microsoft Learn. I am forced to materilize the query result as a derived stream, incurring additional cost and cluttering the UI.   ... View more

I would like to have more control over which queries are sent to a particular custom sql pool in the warehouse experience. As things exist now, only application name can be used to filter queries between custom sql pools. Have the option to create tags to identify queries/workloads would be a nice feature.  I could then prioritize Central-to-whole-business Reporting queries from Power BI versus departmental reporting queries. As things stand now, both would just be classified as application name.  ... View more

Problem Statement Today, on‑premises data gateways must be registered in the Power BI tenant’s home region. In scenarios where the tenant is hosted in one geography and on‑premises data sources are located in another, this forces cross‑region execution for Import refresh and metadata operations. For large Import models, this significantly increases refresh duration because bulk data ingest and control‑plane operations must traverse regions, even when the gateway and data source are co‑located. Requested Capability Allow administrators to explicitly register an on‑premises data gateway in a region other than the home tenant region, or to associate a gateway with a regional execution context closer to the data source. This could be: A selectable region at gateway registration time, or A regional binding model similar to other Fabric connectivity constructs Business Impact Material reduction in Import refresh times for globally distributed organizations Improved predictability and reliability of scheduled refreshes Better alignment with enterprise data‑locality and network‑topology constraints Reduced need for architectural workarounds when on‑premises data cannot be relocated Why This Matters in Fabric As Fabric becomes the unified analytics platform, customers increasingly operate with: Centralized tenants Distributed on‑premises estates Large Import workloads that are sensitive to ingest latency Providing regional flexibility for gateways would materially improve customer experience without requiring changes to data residency policies. ... View more

Currently if you have a lot of elements on a dashboard and you want a slicer to only impact some of them, you have to manually go through and turn off the interaction with everything else. This can be very time consuming. What would be great is if there was an option within the edit interactions to turn it off for everything at once. This would mean that rather than having to turning the interaction off for everything manualy, you could quickly turn it off for everything then select just the ones you want it to interact with.   Just to be clear, I'm not suggesting the default be changed, only to have an additional functionality to make it quicker if you only want a particular slicer to impact a small number of elements on the page ... View more

When using visuals such as clustered column charts or column charts, adding a legend automatically assigns a different color to each item, but there are cases where I want all the colors to be the same. Currently, it is not possible to change all colors at once, and I have to select a color for each item individually. I would like to be able to change the colors of multiple items at the same time. ... View more

Introduce tenant setting capabilities that allow controlled exceptions to centrally enforced tenant‑level policies (such as Allow non‑Entra ID authentication in Eventstream or Block Public Internet Access), enabling governance‑approved deviations by domain or workspace without requiring full tenant‑wide enablement. ... View more

Stop releasing garbage software that doesn't work and breaks constantly, just an idea ... View more

Purview + Fabric Warehouse ≠ enterprise‑ready governance. In our environment, the Fabric Warehouse is the system of record, but Purview can only catalog the warehouse container—not the tables or columns that actually matter for governance. That makes Purview unusable for core enterprise requirements: business glossary mapping, data products, stewardship, discoverability, and trustworthy AI/Copilot grounding. Microsoft’s message is effectively “go redesign your architecture around Lakehouse,” which is unacceptable: we will not duplicate, mirror, or re‑platform our data just to make governance work. Until Purview supports sub‑item metadata scanning for Fabric Warehouses (tables + columns), Purview is not fit for purpose in a Fabric‑first enterprise—and we will drop it. This gap directly contradicts the “single, integrated SaaS framework” promise of Unified Catalog for governance in the AI era. ... View more

If would be nice to disable drill capabilities for a page; either in the canvas properties or URL parameter (similar to &pageName or &authAuth) ... View more

Currently, I am not able to use STRING_AGG and/or JSON_AGG in Scalar UDFs in Fabric Warehouse and SQL anlalytics endpoint. I would like to be able to do that w/o needing to do a code change. We need more votes to understand importance of this requirement. Please describe your exact scenario(s). ... View more

I suggest allowing users to filter by columns that are present in the underlying source table used for configuring 'Verified Answers' within the 'Prepare data for AI' section of semantic models. For example: the source table have collumns: - country - sales_representative - fit_score (business metric)   If I set an verified answer like: 'Please rank the top performers sales representatives for last quarter' it works just fine. But if I add one aditional when I am using copilot to verify the outputs it doesnt work 'Please rank the top performers sales representatives for last quarter for Finland' We added the for Finland to filter specific country.   Please investigate how to allow end users to use it, right now the viable solution would be seting up one question per country which is very demanding. ... View more

Hi Team,   The code below does not find any files.  However, if I replace 'S&P-2013' with 'test' and move over a single test file, it works fine. I am pretty sure this is a bug since it is a allowable path name in ADLS.   Sincerely,   John Miner   PS:  The new statement aligns with other vendors such as SNOWFLAKE.  I wish there was a way to place the secret into a vault and reference the vault instead of hard coding a SAS key that could be an attack vector.    https://docs.snowflake.com/en/sql-reference/sql/copy-into-location    -- -- Load table - all years --   COPY INTO [stocks].[snp500] FROM   "https://sa4adls2030.dfs.core.windows.net/raw/stocks/S&P-2013/*.CSV" , WITH (     FILE_TYPE = 'CSV',     CREDENTIAL=(IDENTITY= 'Storage Account Key', SECRET='<skipped>'),     FIELDQUOTE = '"',     FIELDTERMINATOR=',',     ROWTERMINATOR='0x0A',     ENCODING = 'UTF8',     FIRSTROW = 2 ) ... View more

There doesn't appear to be an easy way to set a default value for Slicers following a data refresh, which is possible in Reporting Services report parameters.   For example, if I have a Slicer that lists Dates in descending order (with the latest date at the top), when I refresh the data I would like the report to always select the latest date, regardless of what that is. Without this, the report always defaults to whatever value was selected when the report was published and users have to change it to the most recent date themselves, which is not a great user experience.    This could be done via a variable Visualisation option (e.g. provide a default value via a DAX measure), or simply an option to 'Select top value after refresh' ... View more

when you have a slicer as a drop down, the list is set to the width of the slicer, if the drop down list could expand or text wrap the values (or an option for both) as you interact with it, it would make selections easier for then there are wide lists of values. ... View more