Managing multiple roles and members for multiple models is a tedious task. It would be more helpful if we had a relevant REST API. ... View more

It would be extremely valuable if Microsoft Fabric allowed developers to invoke Fabric Data Agents directly from Fabric Notebooks (PySpark/Python) through an official API or SDK. Current limitation: Data Agents are primarily accessible through the Fabric UI/Copilot experience, but there is no clear notebook-native integration for programmatic usage. Suggested capability: Allow notebooks to:  - Send prompts or questions to a Data Agent  - Receive grounded responses programmatically  - Maintain conversational context/session state  - Integrate agent workflows into data engineering and AI pipelines Example use cases:  - Automated business insight generation  - AI-driven anomaly detection workflows  - Agentic analytics pipelines  - Intelligent report generation  - Integration with Spark/PySpark transformations  - Enterprise AI orchestration inside Fabric Example ideal workflow: response = fabric_data_agent.ask(            agent="RetailInsightsAgent",            prompt="Summarize customer churn trends for Q4" ) Why this matters: Currently, many teams must rebuild semantic grounding, orchestration, and retrieval pipelines manually inside notebooks using external LLM frameworks. Native notebook integration would significantly accelerate enterprise AI adoption within Microsoft Fabric. This feature would make Fabric Data Agents much more production-ready for advanced AI engineering and analytics workflows. ... View more

Please add workspace settings >> Data Warehouse >> Collations to the Fabric rest API config.   This is the only way make Lakehouse SQL Endpoint Case insensitive, it is a must for customers coming from a SQL background.  Currently I deploy workspaces by automation, then manually change the collation then deploy Lakehouses & Warehouses which is time consuming and annoying....   This is pre-req before the terraform provider could also be updated to set.   Thanks, Darren ... View more

I am working on a Workspace that is Power BI PRO. The client is NOT going with Fabric. I am building a Semantic Model in the workspace. I see under the Refresh drop-down, an option for "Create advanced refresh (preview)" and figured I would give it a shot. What happened when I selected that option was that it IMMEDIATELY signed me up for a Fabric Free Trial and moved my workspace over. (Plus it hosed the in-proecess refresh of my semantic model). There was no option to CONFIRM OR DENY the upgrade to Fabric Free Trial. Please put in some type of confirmation so the user must ACCEPT that A) They will be signed up for a Fabric Free Trial and B) Their workspace will be move over to that trial.   Horribly experience! ... View more

I would suggest to introduce "folders" to have new connections organized in a easier and segregated way. In a Enterprise environment, wheer all teh connecitons to resources are managed ina  centralized way, having the possibility of creating folders, as in the Workspaces, would allow better organized resources (e.g. Like in the Deployment pipelines). Thank you for the opportunity, Alfredo Lourido Santos ... View more

Current behavior: The Ribbon Chart visual has two distinct hover zones: the column/bar segments and the ribbon transitions (the lighter connective areas between time periods). Custom report page tooltips only apply to the column/bar segments. The ribbon transition areas always display the built-in default tooltip showing % change and rank, with no option to override or customize. Desired behavior: Allow custom report page tooltips to apply to the ribbon transition areas as well, or at minimum provide a formatting option to choose which tooltip is displayed on each zone. This would give report authors full control over the tooltip experience across the entire visual. Context: This limitation was confirmed by a Super User in the Power BI Community in April 2020 and remains unresolved as of May 2026: https://community.fabric.microsoft.com/t5/Desktop/Ribbon-chart-custom-tooltip/td-p/1040519 ... View more

Summary The recently introduced Recycle Bin in Microsoft Fabric is a very useful addition; however, it currently lacks support for multi‑select and bulk delete operations. Although the UI displays a checkbox (square box) suggesting multi‑selection is possible, selecting multiple items does not allow deleting or restoring them in one action. Current Limitation Users can only delete or restore one object at a time. The UI visually implies multi‑select capability, but the functionality is not implemented. This becomes inefficient and time‑consuming when cleaning up: Abandoned pipelines Temporary lakehouses Test notebooks CI/CD artifacts Failed or duplicate objects Expected / Requested Enhancement Enable the following capabilities in the Fabric Recycle Bin: True multi‑select support Allow selecting multiple objects using checkboxes. Bulk actions Delete selected items in a single operation. Restore selected items in a single operation. “Delete All” or “Empty Recycle Bin” option Similar to the Windows Recycle Bin experience. Clear visual feedback Enable/disable action buttons based on selection state. Why This Is Important Improves productivity and usability, especially in development and test environments. Aligns user expectations with standard operating‑system behavior (Windows Recycle Bin). Reduces repetitive manual actions and accidental errors. Supports better workspace hygiene, especially for teams using: CI/CD Frequent branching Automated pipeline generation Business Impact Faster environment cleanup Better developer experience Lower operational friction More intuitive and consistent UI behavior Conclusion Since the UI already indicates multi‑select capability, enabling bulk delete and restore would be a natural and highly valuable enhancement. Making the Fabric Recycle Bin function similarly to the Windows Recycle Bin would significantly improve usability and efficiency. ... View more

I have a set of scheduled object that I am trying to trace down the failures for it. I use the Monitor Hub amnd filter for Failed items and "Last 24 hours". My system is showing 8:04 AM (Eastern Daylight time) on 5/7/2026. I see start times as far back as 1:30 AM on 5/06/2026, which by my count would be about 30 hours ago. Please fix this. This is just BASIC stuff! ... View more

When using Shortcut Transform in Microsoft Fabric with the Japanese UI, the item may be displayed as “正体不明” (meaning "unknown identity") while the shortcut transformation is still running. This wording is confusing for Japanese users because “正体不明” sounds like something is broken, invalid, or in an error state. However, in some cases, the transformation is simply still in progress, and the table appears correctly after refreshing or waiting for a while.     Please improve the Japanese UI wording for this state. For example, instead of “正体不明”, it would be clearer to show one of the following: - Transforming - Processing - Table Recognition - Shortcut Transforming - Checking Status This would help users understand that the shortcut transformation is still in progress, rather than assuming that an error has occurred. This is especially important for new Fabric users because the current wording can cause unnecessary confusion during Shortcut Transform. ... View more

🧠 Imagine you've written a brilliant DAX UDF in model view. Finally you decide to add a comment, and suddenly the whole function disappears. 'Undo' is not working. 🚨This happened to me a few days ago. It was my fault for trying to add a comment with "//" above the function header. However, I hadn't expected this to delete the entire function without being able to do an undo. 💡As I discovered, the problem isn't the comment itself, but characters that aren't allowed in a function name. After(!) deleting the function Power BI displayed a warning : "Special characters are not allowed in a function name..." That put me on the right track. Step-by-step guide to follow: ➡️ Write a DAX UDF in model view (or open an existing one) (don't spend to much time on it, as your work will get lost) ➡️ Go to the function name ➡️ Add an invalid character such as "/" to the function name (or add a new first line an enter an invalid character there) ➡️ Press Enter or click ✅ trying to save your function ➡️ Is your function still there? ➡️ If not, check wether undoing (Ctrl-Z) works ❓ Is this a bug or a hidden feature? If it is a feature what is it good for?  ... View more

Problem: Currently, there's no way to filter the "CU% over time" visual by workspace.   This means we're unable to track interactive CU% by workspace. We have 20+ workspaces in a capacity, and we wish to filter the "CU% over time" visual by workspace.   Proposed solution: Make it possible to filter the CU% over time visual by workspace. ... View more

PROBLÈME CONSTATÉ Les signets Power BI permettent de contrôler la visibilité des visuels sur la toile, l'état des slicers et les filtres au niveau de la page, mais ils n'ont aucune influence sur le volet filtre natif . Les champs ajoutés au volet filtre restent toujours visibles, peu importe le signet actif ou le bouton sur lequel l'utilisateur clique. "Je ne peux pas masquer la colonne française « Catégorie » quand l'utilisateur bascule en anglais, ni afficher uniquement « Category » , les deux apparaissent toujours simultanément dans le volet filtre, ce qui brise complètement l'expérience multilingue." SITUATION ACTUELLE VS. SOUHAITÉE ❌ AUJOURD'HUI Tous les champs ajoutés au volet filtre sont visibles en permanence. Les signets ne peuvent pas masquer ou afficher des champs individuellement. Les développeurs doivent abandonner le volet filtre natif et le reconstruire entièrement avec des slicers sur la toile. ✅ SOUHAITÉ Chaque champ du filtre dispose d'un paramètre de visibilité sensible aux signets . Lorsque le signet BK_EN est actif, les champs anglais sont affichés et les champs français masqués, et inversement pour BK_FR . SOLUTION PROPOSÉE Ajouter un bouton de visibilité par champ dans le filtre du volet (icône œil, similaire aux visuels du canevas) Rendre cet état de visibilité capturé par les signets , comme le masquage/affichage des visuels sur la toile En option : permettre à une mesure DAX ou un paramètre de piloter la visibilité de façon dynamique. Périmètre : applicable aux sections « Filtres dans cette page » et « Filtres dans toutes les pages ». CAS D'USAGE BÉTONS * Rapports multilingues (FR / EN) * Filtres par rôle utilisateur *Vue débutant vs. avancé ENTREPRISE À IMPACT Les rapports Power BI multilingues sont de plus en plus exigés par les entreprises internationales, les marchés bilingues (Canada, Belgique, Suisse) et les équipes BI mondiales. Aujourd'hui, le seul contournement est de remplacer entièrement le volet filtre natif par des slicers personnalisés via des signets, une solution coûteuse et complexe qui perd l'expérience native et augmente significativement la maintenance des rapports. Cette seule fonctionnalité éliminerait le problème n°1 dans la conception de rapports Power BI multilingues de qualité professionnelle. ... View more

We had a situation where a rogue report used a ridiculous amount of CUs in a Fabric Capacity, causing the Capacity to lock up. Anyway, while reviewing Fabric Capacity Metrics dashboard the same day, the huge spike reduced our normally populated CU % Over Time graph to compress the bars so badly that they are no longer visible. Also, even if we adjust the date/time slider to omit the time of the problem spike, the Y scale stays the same for the remaining data and we still can't display it.  See the attached screenshots - one with the spike showing, and one without. See the Y Axis. In the graph without the spike, there should be normal 20%-100% traffic showing. Could the Y axis have an option to use a logarithmic scale if required, and if not, can it just top out at 100%, but the tooltip and any data exports show the actual number? ... View more

It would be very useful to right-click on a column header in Power Query and find out which step defined the column in its current state.  It would save a lot of clicking through applied steps or searching through M code in the Advanced Editor, and it would be very helpful in cleaning up queries. ... View more

After I create a new step, I want to be able to rename the step to document the purpose of the step.  However, I have to wait for the step to completely execute first. This can cause a delay of several seconds for a complicated query, and if I'm adding several steps in sequence, this adds up.  Other commands on the query step are available without waiting, but I'd like Rename to be included in those commands.   I know I can skip the wait by using Advanced Editor, but then I have to edit the name in multiple locations to ensure that I don't break the query. ... View more

Customers using the Power BI Output Tool in Alteryx to publish datasets to the Power BI Service are currently limited by the backend Push Dataset API restrictions. Since this tool relies on the Push Dataset API, the same limitations apply during publish, and datasets with 10M+ rows cannot be pushed successfully. We would like an enhancement to this limitation so customers can publish large datasets/models from Alteryx flow to the Power BI Service using Power BI output tool. This is important for enterprise scenarios where Alteryx is used as the primary data preparation and publishing tool, and the current API-based limit prevents large-scale reporting workflows from being implemented efficiently. ... View more

Customer is seeking a solution to retrieve gateway roles assigned to all users across all gateway connections within their tenant. However, based on our current assessment, there is no out-of-the-box method or available API that supports retrieving this information at a tenant-wide level. Could you please confirm if there are any alternative approaches or solutions to achieve this requirement? Raising this on behalf of our customer. ... View more

Server-Sent Events (SSE) is a highly popular and rapidly growing protocol particularly for applications requiring one-way, real-time data streaming from server to client.   This is a nice addition to the HTTP connection source as offered at this moment. The SSE protocol built-in auto-reconnection with a 'last received message id' makes it easy to offer disconnects without message loss.   We see SSE also being offered by Azure IoT Operations so there is knowledge with Microsoft on how to set this up efficiently. ... View more

I know there's a hacky roundabout way of achieving this using the line and clustered bar chart, but the functionality to add vertical lines to show events really should be a native feature of the line chart. As data storytellers, this feature would enable us to provide important context for our stakeholders/end users. ... View more