Issue Description We are experiencing an issue with Text Input Slicers in Power BI Service where the slicer value does not reset when using bookmarks or the Clear All Slicers option. Observed Behavior A user types a value into a Text Input Slicer (e.g., MRN). Clicking a Reset/Clear bookmark visually clears the slicer. When the user interacts with the slicer again, the previously typed value automatically reappears. This behavior occurs only in Power BI Service. The same report works correctly in Power BI Desktop. Expected Behavior Text Input Slicer values should fully reset when a reset bookmark or Clear All action is applied. After reset, the slicer should not retain or reapply previously typed values. Additional Notes This issue affects all reports using text-based slicers (e.g., MRN, Account Number). Dropdown slicers work as expected and reset correctly. This behavior appears to have started after a recent Microsoft Fabric / Power BI Service update. The issue is not related to report design, bookmark configuration, or persistent filter settings. Business Impact Prevents reliable Reset Filters functionality. Causes user confusion when old values reappear. Forces redesign of production reports to avoid text input slicers. Current Workaround Replace Text Input Slicers with Dropdown slicers, which reset correctly. Request for Clarification Could someone from the Power BI / Fabric product team please confirm whether this behavior is by design or a known issue planned to be addressed? This impacts multiple production reports and is forcing us to replace text input slicers with dropdowns. Understanding whether a fix or configuration option is planned would help us decide whether to wait for a resolution or permanently redesign our reports. ... View more

Regardless of desktop appearance selection (dark, legacy, light, system default) the top three icons on the left nav (Report view, Table view and model view) are now rendered solid black.  Has anyone seen this? The icons are functional, just solid black.  Pretty sure this showed up just today; repaired the install to no effect. - Using November 2025 Release.       ... View more

Hi, I am experiencing inconsistent behavior between Power BI Desktop and Power BI Service when using the new Text Input Slicer together with bookmarks.   The scenario is the following.   I am building a report that uses Translytical Task Flows, where a Text Input Slicer is used as a free text comment field. The value from the slicer is passed to a User Defined Function and written back to a data warehouse.   Because the Text Input Slicer is used as an input to a Translytical Task Flow, it becomes mandatory. This means the action button remains disabled until the slicer contains a value.   To support a proper user flow, I created a bookmark configured with selected slicers only. The purpose of the bookmark is to reset the report state after the task flow has executed. In Power BI Desktop, this works as expected. When activating the bookmark, the Text Input Slicer is reset along with the other slicers.   However, after publishing the report to Power BI Service, the behavior changes. The same bookmark correctly resets all other slicers, but the Text Input Slicer keeps its previous value and is not reset. This happens consistently and is fully reproducible. Steps to reproduce Create a report with a Text Input Slicer Use the Text Input Slicer as an input to a Translytical Task Flow Create a bookmark with selected slicers enabled Enter a value in the Text Input Slicer Activate the bookmark in Power BI Desktop and observe that the slicer is reset Publish the report to Power BI Service Activate the same bookmark in the Service and observe that the Text Input Slicer is not reset Expected behavior The Text Input Slicer should respect bookmark state in Power BI Service in the same way as in Power BI Desktop, and be reset when included in a bookmark with selected slicers. Actual behavior In Power BI Service, the bookmark does not reset the Text Input Slicer, even though it resets all other slicers correctly. Impact This makes it very difficult to build a clean user experience around Translytical Task Flows. Since the Text Input Slicer is mandatory, users are effectively forced to manually clear or re enter text after each execution. A common workaround would be to use a default placeholder value, such as a single space, and treat it as NULL in the UDF. However, because bookmarks do not reset the Text Input Slicer in the Service, this workaround also fails after publishing. The Clear all slicers button does reset the Text Input Slicer, but it clears it completely. This disables the Translytical Task Flow action button again, so it does not solve the problem from a usability perspective. Additional notes This appears to be a Service specific issue, as the same report behaves correctly in Power BI Desktop. ... View more

A couple of days ago we noticed an issue with our Azure maps in reports published via public to web. The map will load like below but without any data points mapped. The only way to get it to load is to switch from using data such as postcodes or street addresses in the 'location' input, to using the 'latitude' and 'longitude' inputs instead.   I tried this on various devices and internet connections and kept getting the same issue.   We think this indicates an issue with publish to web reports not being able to lookup related mapping coordinates to the provided location field? As it always displays as expected in the desktop file and the app.powerbi.com workspace, even when using our location field input.   Some clarification of why this is happening, and whether it can be fixed would be helpful thanks.   I've created a public report that demonstrates the Bing vs Azure map for postcode location (1st page - Azure map doesn't work) and then using latitude/longitude for the Azure map (2nd page - works) https://app.powerbi.com/view?r=eyJrIjoiYmFkNGI3MjItNmU2Yi00OGViLWI3N2MtMGRlMzY5MmQ1MWRiIiwidCI6IjM3YzM1NGIyLTg1YjAtNDdmNS1iMjIyLTA3YjQ4ZDc3NGVlMyJ9   ... View more

Hello, The November update of Power BI Desktop messed up the metrics and visuals in my new card visual. It happened with the image in the callout, and the reference labels had their layout changed, no longer keeping the spacing I wanted. This is really frustrating because the users who view the report daily (including the bosses) are not getting the metrics they need. Could you please provide a fix or workaround for this problem? Thank you for your support. Just a sample image for your reference.   ... View more

Hello everyone, I’m facing an issue with the Azure Maps visual in Power BI after this new update and I’d appreciate any help or insights. I added a reference layer to an Azure Maps visual. In Power BI Desktop, the reference layer behaves as expected: the transparency works correctly, allowing me to clearly see the base Azure Maps content behind it (street names, places, establishments, etc.). However, after publishing the report to Power BI Service (web), the same reference layer appears as if its transparency were 0% (fully opaque). As a result, it completely hides the underlying Azure Maps details, such as street names and other map information. Has anyone experienced a similar issue with Azure Maps visuals behaving differently between Desktop and Service? Is this a known limitation or bug, or is there any workaround to ensure the transparency is respected after publishing? ... View more

Dear MS Power BI Team,   Since ± the beginning of October 2025, I have noticed an issue that the "Grid" type of Arrangement is broken on the service side:   Expected: in case of having 4 columns out of 6, an extra space on the right side left for 2 empty columns     Once published to the service side, the buttons stretch across the entire width, same way as behave "Horizontal" type of arrangement:     Can you please pay attention to this service side issue and fix?   regards, Alena ... View more

Hello, I built a Tranlytical Task Flow based on text input for commentary in Power BI and suddenly slicer now doesn't accept more than 99 characters, truncates the input every time I try to run the function. This is disrupting the application I built.    On PBI Desktop everything works fine, slicer works as expected.    I know this is a preview feauture, but it's been 8 months since original release of Translytical Task Flows and this change is rather unexpected.  ... View more

In a column and line combo chart, custom labels for the line series do not work as expected. As shown in the GIF below, even though a custom label is applied, the visual still displays the actual line value instead of the custom label.   ... View more

I have an issue where the “new card” visual layout becomes misaligned / messy after publishing to Power BI Service, even though everything looks neat and properly aligned in my PBI local (desktop).     PBI Desktop         PBI Services   is there anyone have same issue guys? ... View more

I have a report with a matrix visual, which is powered by a DirectQuery source that gets updates every minute. The report refreshes automatically with a change detection on the column with last updated timestamp. The problem is that the matrix visual scrolls up to the 101st position after each visual refresh. Users have to scroll back down to the position they were looking at before, and after a minute it jumps back up again. The faster I wish to refresh this report, the more prominent this issue becomes. The position of the scroll bar does not reset when the visual displays records in range 1-101 prior to refresh. The root cause seems to be present in the DAX query generated on the matrix refresh event - it contains "TOPN(101)" filter. Each scroll event past that number of rows generates a subsequent query that fetches next 101 records. This degrades the user experience massively, and puts unnecessary strain on my source by sending lots of small queries. In this case, I'm working with around 3000 raw records, which aggregate to 350 rows in the matrix. This could be easily fetched with a single query, which should keep the position of the scroll bar in place between refreshes. I would like to be able to specify the number of records fetched in a single query to my source; 101 is a very conservative number. ... View more

After the latest updates released with the November Update, all our reports that use the New Card Visual have a serious problem with the management of images inserted in the Card. For example. I have this card with three blocks in which I have inserted a small icon (see image). In PowerBI desktop, the rendering is correct and consistent with the values entered in the configuration. However, when I publish the report on PowerBI Service, the card rendering is completely incorrect: (note how the image in the card is significantly larger, as if the “size” 50px parameter were not being taken into account) Could you kindly intervene as soon as possible, as all our reports using the new cards are currently illegible and, in some cases, do not contain any information? Regards -- Gianluca Pinelli ... View more

To connect to our IBM DB2 database, we need to set a custom "packageCollection" value as an additional connection property in a copy activity.   However, when inspecting the pipeline's JSON, we noticed that "packageCollection" is added with a trailing space ("packageCollection "). This causes issues, as the DB2 database doesn’t recognize the property correctly.   Manually removing the space in the JSON makes the pipeline work again, so for anyone facing the same issue, that’s a temporary fix. But ideally, this bug should be resolved at the source.   As it stands, fixing it manually means the property can’t be configured through the UI anymore.   ... View more

We were able to achieve this type of presentation with the new card visual while in preview (reference label only for certain sub cards, centered callout value in the other ones). Last november update messed up a whole report built with the preview version of the visual, displaying empty spaces for all the KPI without reference label data. We couldn't recreate initial visuals, neither with the preview version nor the general availability one.   ... View more

Prior to the September 2025 update, selecting Refresh in Power BI Desktop would refresh both schema and data. For the duration of the refresh, you'd see the number of rows loaded in each query. This was nice, because if you knew how many rows were in the underlying object, you could get a rough sense of % completion (i.e. I know the finished query should have 25 million rows and I see it's at 12 million loaded, so I'm approaching halfway done).   With the September 2025 update, the ability to refresh only schema or refresh only data was added, which is great. But if you choose to refresh only data, there is no communication of how many rows have been loaded so far. All you get is the ever-annoying "Working on it" modal.    Request: please surface # of rows loaded when refreshing only data (no schema). No one likes being presented with a non-dismissible modal which gives absolutely no indication of progress completed.  ... View more

Hello, Starting today, Power BI Desktop freezes every time I try to save.  This happend to me and at least one other colleague. The freezing does not prevent the saving itself so far, but I need to restart Power BI Desktop after each time I save my Report.   Has anyone had a similar issue so far or any suggestion to resolve it? ... View more

Good Day,   Occassionally I filter tables using a Table View to copy filtered results to use outside of Power BI - I.E. in Excel.   As of today it appears that when copying the Table View that has been filtered, it ignores the filters and copies the whole table. This is an issue for me as when I have a table in excess of 700,000 rows it makes this process extremely slow to copy from Power BI, and near impossible to paste it into Excel.   I have already tried uninstalling and reinstalling the program.   Is anyone else experiencing this issue?   As an example, we are a service company and we have a geographic alignment for our field service team. Simply, filtering a table by the "East" Region and trying to copy the table when filtered to the East is not working. Instead, the entire table is copied.   ... View more

Hi, Since the update July 2025, Matrix visuals with fields parameter hierarchies show a confusing behavior when one field is on different hierarchy levels. To illustrate the issue I created a test report with the following field parameter:   FP_Test = { // Group1 ("Category", NAMEOF('Product'[Category]), 0, "Group1"), ("Business Type", NAMEOF('Reseller'[Business Type]), 1, "Group1"), // Group2 ("Country", NAMEOF('Sales Territory'[Country]), 10, "Group2"), ("Category", NAMEOF('Product'[Category]), 11, "Group2"), // Group3 ("Business Type", NAMEOF('Reseller'[Business Type]), 20, "Group3"), ("Color", NAMEOF('Product'[Color]), 21, "Group3") }   The fields "Category" and "Business Type" are part of 2 hierarchies but on different levels. The visual looks like this:   If I now switch to Group2, I get the following display:   For the users, it looks as if the button did not work, as the category is still displayed. They can't even recognize that they are at a drilled down level because no +/- signs and no elements of the next higher hierarchy are displayed.  Switching now to Group3 it gets completely weired because only the lowes level ("Color") is visible whereas the user would expect to start with "Business Type". The users have to be very experienced to notice that they are looking at a drilled down level and need to drill up  to see the full hierarchy for this group:   Starting with an expanded hierarchy in Group1, it is not apparent in which case the hierarchy remains expanded, when it is collapsed and when a drill down takes place without a top level hierarchy. Anyway, a hierarchy that is expanded by default is not recommended or possible for a large number of hierarchy elements.   The same behavior applies to bar charts and column charts, where it is even more difficult to recognize that it is a next lower level and you have to drill up.   Please add a formatting option to all visuals that works with the drill-up/drill-down functionality, where the developer can choose whether the matrix/chart should collapse when switching the dimensions or retains its expanded state.   And if a matrix/chart retain the expanded state, please add the upper levels as well! I'm working as BI Consultant for customers and they complain about this new behavoir, because it appears as if the dynamic doesn't work anymore. I don't want to go back to bookmarks switching between various matrix visuals.   Thanks and best regards! ... View more

Hi, Note: Issue noticed in unreleased July 2025 edition of PBI Desktop.   I m using Field Parameters on Rows of the Matrix and trying to provide dynamic selection for the users. However, the functionality stops after making slicer selection a couple of times and expand/ drill down buttons no longer works.   For the selection as shown in snapshot below, I would expect the user to expand and see Col2 and Col3 in the matrix visual. However.  no matter how many times you click, its not possible to expand. It is not dependent on the order of selection of columns or layout configuration in use. Request you to address the issue ASAP.     Sample PBIX is available at the below link: Test/Test Matrix.pbix at main · modi123p/Test   Thanks & Regards, Madan Prabhakara ... View more