To support business users, provide a GUI-based table creation feature that requires no coding. ... View more

OPENROWSET function can read the files from Azure Data Lake storage or Azure Blob storage.  It would be useful to enable OPENROWSET to also read files form Fabric One Lake and access the files in Lakehouse /Files area. ... View more

Fabric Data Warehouse allows reading Parquet and CSV files using the OPENROWSET function. However, it does not support external tables that reference Parquet or CSV files (CREATE EXTERNAL TABLE (Transact-SQL) - SQL Server | Microsoft Learn). This limitation hinders the straightforward migration of Polybase external tables from Synapse dedicated/serverless pools or SQL Server. The only available workarounds for migration involve either converting Parquet/CSV files to Delta Lake format or rewriting the code to use OPENROWSET, which lacks the same security model. As a result, migrating from Synapse or SQL Server to Fabric Data Warehouse requires data or code conversion.   Supporting external tables with Parquet and CSV files is crucial for seamless migration and ensuring compatibility with other SQL Server/Data Warehouse offerings. ... View more

Fabric Warehouse currently supports the OPENROWSET function, which allows you to read CSV files directly. However, when it comes to reading JSON files, you need to use a workaround. This involves treating JSON files as text or CSV files and then using OPENJSON and JSON functions to parse the JSON documents. While this method works, it is considered a hacky workaround and not very elegant. To improve this, Fabric Warehouse should add a native support for JSON in the OPENROWSET function. This would eliminate the need for tricks with OPENJSON and JSON functions, making the process of reading JSON files more straightforward and elegan.   The current workaround for reading json is: SELECT TRY_CAST(JSON_VALUE(jsonContent,'$.orderId') AS INT) AS orderId, JSON_VALUE(jsonContent,'$.orderDate') AS orderDate, JSON_QUERY(jsonContent,'$.orderDetails') AS orderDetailsObject, JSON_QUERY(jsonContent,'$.deliveryAddress') AS deliveryAddressArray FROM OPENROWSET( BULK 'https://<storage>.dfs.core.windows.net/datalakehouseuk/raw/json/orders.json', FORMAT = 'CSV', FIELDTERMINATOR ='0x0b', FIELDQUOTE = '0x0b' ) WITH ( jsonContent VARCHAR(500) ) AS r;   ... View more

Currently the server name for SQLDB (and DW) is a double GUID of {tenant-Id-workspace-id.}database.fabric.microsoft.com   This causes a lot of issues as 1. This is impossible to remember 2. We ahve no way to distinguish test from PRD   We request user frirnedly DNS Names like  Prodata-Sales-DEV.database.fabric.microsoft.com   This would help clearly label an worksapce as DEV or PRD and avoid confusion and make the platform easier to work with., ... View more

Microsoft SQL Server has a very powerful feature by the way of "synonyms." It allows users and DBAs to do all sorts of powerful magic such as rewiring objects under the hood (e.g. run the code against another test database). Synonyms are lightweight and more resilitent than views.   This is more general than lakehouse short cuts. It is completely different from Clone.   At the moment it is not supported, please support. ... View more

Add functionality to Manage (Data Model Settings) Page Currently this just shows a grayed out >Data Source Credentials which cannot be maintained in the Power BI Service. Add functionality to manage gateway like PBIX Reports Add functionality to manage paramaters (see below for more detail) Ability to manage parameters in the Power BI Service for Power BI Report Builder Paginated Reports Currently paginated reports (.rdl) do not have the parameter flexibility like the standard power bi reports (.pbix) where parameters can be changed on the back end (e.g. pointing test to test and dev to dev database via a parameter used for the data source).   It would be awesome for both Internal paginated report parameters AND power query parameters to be managable from the power bi service in the manage screen.   Rename the Manage and Settings menu options in the Power BI Service (elipses options on RDL files) For RDL, Manage option appears to take user to a screen similar to PBIX Data Model 'settings' while Settings appears to take you to Report 'settings'. To make things easier to follow, simply rename Manage to Report Settings and Settings to Data Model Settings.   Workspace level setting Add functionality to set which roles can create, read, update, and/or delete ALL associated subscriptions rather than just the ones created by current user for. ... View more

Currently, the Microsoft Fabric Data Warehouse supports metadata-based backup capabilities, which are effective for recovery in the event of failures. While Disaster Recovery (DR) provides fault tolerance, its high cost makes it challenging for organizations with limited budgets to implement. To address these limitations and provide a more cost-effective solution, we strongly request the addition of a feature to directly export or back up tabular data stored in OneLake in the Delta Parquet format as part of the Data Warehouse functionality. Furthermore, implementing backup options such as "full" and "differential," similar to those available in SQL Server, would improve flexibility and operational efficiency. Introducing these features would significantly enhance Fabric's backup and recovery capabilities, providing users with a robust and cost-effective data protection solution. ... View more

When I write a spreadsheet into my lakehouse files folder, I would like to be able to download it without having to install the onelake storage explorer.  Please provide a download link for files in the lakehouse files folder! ... View more

I would like to request an update to the public information attached below in the Japanese version. The English version is up to date, but the Japanese version has not been updated since 2024/11/20.   JP URL: https://learn.microsoft.com/ja-jp/fabric/database/mirrored-database/azure-databricks-tutorial EN URL: https://learn.microsoft.com/en-gb/fabric/database/mirrored-database/azure-databricks-tutorial   Reason for updating: I confirmed in the English version that a new "Prerequisites" was added. The Japanese version does not have this notation, so customers cannot test it following the correct procedure. ... View more

Could Microsoft support SQLMesh project so that they provide good compatibility against Fabric warehouse and lakehouse (via Spark)? Could SqlMesh be added as Fabric native workload to get it as service since this kind of sql modeling functionality with web UI is missing? Add support for Microsoft Fabric · Issue #3374 · TobikoData/sqlmesh ... View more

Since there is a problem with operation when the account of the owner of the created lakehouse content is closed, I would like to add the function to change the function of the owner ... View more

Hi Team,   Is there any plans to add more default collation options when creating a warehouse? eg Latin1_General_100_CI_AS_SC_UTF8   For me collate Latin1_General_100_CI_AS_KS_WS_SC_UTF8 doesn't help with the main issue for case sensitivity on the data select. eg When using SELECT with LIKE %% we still need to change collation which is pain and often gets forgotten leading to incorrect outputs.   eg still need to do this unfortunately  SELECT * FROM [warehouse].[dbo].[MyTable] WHERE [MyColumn] LIKE '%SEARCH%' COLLATE Latin1_General_100_CI_AS_SC_UTF8;   Thanks, Darren ... View more

I was so happy when the support for a case insensitive collation was finally released. However, the fact that you cannot select your collation in the GUI makes it less useful. Currently you have to create a warehouse via an API command to get the case insentive collation. And while this is no issue if all you do is create your Fabric items via code, there are a lot of organizations that still create their items via clickops.   So please make it possible to select collation type in the GUI when you create your warehouse. A checkbox to opt in for the non-default collation is fine, but the option to select from a drop-down menu or similar would also be good. ... View more

Dear Product Group,   Why is the very common function not supported in the Fabric Warehouse?   It is supported on all other flavor of SQL Server.  Yes, I can query the sys.objects table to get the id, but a shortcut function reduces my total amount of code.   Sincerely   John Miner Data Platform MVP         ... View more

A way to get Last Refresh TIme and Last Accessed time of a warehouse table ... View more

Identity columns are a common method for generating surrogate keys in data warehousing scenarios implemented on SQL Server. By adding support for identity columns in Fabric, users could migrate their warehouse workloads from SQL Server to Fabric, without having to make drastic changes to their data model or data load procedures. This would help to streamline the migration process, making Fabric a more attractive option for those looking to move their data warehouse to the new platform. ... View more

Hi Product Team,   In SQL Server, the fastest way to get a row count from a set of tables is shown below.  Unfortunately, the wrong aswer is coming back, zero records.     If we look use a SELECT statement, we get the correct row count.   The statement is supported in TSQL for MS Fabric Warehouse but does not work correctly.   Sincerely   John Miner Data Platform MVP   ... View more

TL;DR: nested WITH statements are supported in all modern data warehouses, lakehouses, and database engines except for Fabric & T-SQL products. This makes the usage of SQL templating engines unnecessarily complex. Fabric supports regular WITH statements like the following: with customers as ( select * from lakehouseone.dbo.customers ), orders as ( select * from lakehousetwo.dbo.orders ) select * from orders o join customers c on o.customer_id = c.id However, the following nested version is currently not supported: with customers_with_addresses as ( customers as ( select * from lakehouseone.dbo.customers ), addresses as ( select * from lakehousetwo.dbo.addresses ), final as ( select c.id, c.name, a.line_1, a.zip, a.country from customers c join addresses a on c.address_id = a.id ) select * from final ), orders as ( select * from lakehousetwo.dbo.orders ) select * from orders o join customers_with_addresses c on o.customer_id = c.id You can see how working with CTEs allows for a very clean approach to SQL and enables the use of modular blocks of SQL. This is especially popular in SQL templating engines like dbt. Nested WITH statements allow for easy query injection. Wrapping a SELECT statement and inserting it into a query as a CTE only works if that inserted statement doesn't contain any CTEs. More users have complained about the lack of nested WITH statements here and here. ... View more