Merge pull request #10544 from MicrosoftDocs/main

Auto Publish – main to live - 2025-09-23 22:04 UTC

Author: learn-build-service-prod[bot]

docs/fundamentals/direct-lake-overview.md

@@ -73,7 +73,7 @@ The following table compares Direct Lake storage mode to Import and DirectQuery
 | Licensing | Fabric capacity subscription (SKUs) only |Fabric capacity subscription (SKUs) only |  Any Fabric or Power BI license (including Microsoft Fabric Free licenses) | Any Fabric or Power BI license (including Microsoft Fabric Free licenses) |
 | Data source | Tables of any Fabric data source backed by Delta tables | Only lakehouse or warehouse tables (or views) | Any connector | Any connector that supports DirectQuery mode |
 | Connect to SQL analytics endpoint views |No | Yes – but will automatically fall back to DirectQuery mode | Yes | Yes |
-| Composite models | Yes - can combine with import storage mode tables in Power BI web modeling and DirectQuery tables with XMLA tools. |No <sup>1</sup> | Yes – can combine with DirectQuery or Dual storage mode tables | Yes – can combine with Import or Dual storage mode tables |
+| Composite models | Yes - can combine with import storage mode tables in Power BI web modeling and DirectQuery tables with XMLA tools. |No <sup>1</sup> | Yes – can combine with DirectQuery, Dual, and Direct Lake storage mode tables | Yes – can combine with Import, Dual, and Direct Lake storage mode tables |
 | Single sign-on (SSO) | Yes |Yes | Not applicable | Yes |
 | Calculated tables |Yes – but calculations can't refer to columns of tables in Direct Lake mode. |No – except [calculation groups](/power-bi/transform-model/calculation-groups), [what-if parameters](/power-bi/transform-model/desktop-what-if), and [field parameters](/power-bi/create-reports/power-bi-field-parameters), which implicitly create calculated tables | Yes | No – calculated tables use Import storage mode even when they refer to other tables in DirectQuery mode |
 | Calculated columns |No | No  | Yes | Yes |

docs/fundamentals/direct-lake-web-modeling.md

@@ -16,8 +16,8 @@ LocalizationGroup: Admin
 
 | Scenario | Direct Lake on OneLake | Direct Lake on SQL |
 |:-----------|:------------|:  ------------|
-| Creating in the web       | Select **Create** in the left navigation, then **OneLake catalog**. Select **New semantic model** from Lakehouse. Select **OneLake catalog** from web modeling. Select **New item** from a workspace and choosing **Semantic model**, then **OneLake catalog** | Select **New semantic model** from SQL analytics endpoints or Warehouses.       |
-| Editing in the web       | Select **Open data model** from the semantic model details page or context menu. | Select **Open data model** from the semantic model details page or context menu.        |
+| Creating in the web       | <ul><li>Select **Create** in the left navigation, then **OneLake catalog**.</li><li>Select **New semantic model** from Lakehouse.</li> <li>Select **OneLake catalog** from web modeling.</li> <li>Select **New item** from a workspace and choosing **Semantic model**, then **OneLake catalog**</li></ul>    | Select **New semantic model** from SQL analytics endpoints or Warehouses.       |
+| Editing in the web       | Select **Open data model** from the semantic model details page or context menu.    | Select **Open data model** from the semantic model details page or context menu.    |
 
 **Edit in Desktop** is available when web modeling to continue [live editing any Direct Lake semantic model in Power BI Desktop](direct-lake-power-bi-desktop.md). 
 
@@ -28,14 +28,14 @@ To create a semantic model with **Direct Lake on OneLake tables**, take the foll
 1. Select **Create** from the left navigation bar, then select **OneLake catalog** and choose a Fabric item. Alternatively, open the Lakehouse and select **New semantic model**.
 2.	Give your semantic model a name, pick a Fabric workspace for it, and select the tables to include. Then press **OK**.
 
-The semantic model is created and now you are live editing the modeling in the browser.
+The semantic model is created and now you're live editing the modeling in the browser.
 
 To create a semantic model with **Direct Lake on SQL tables**, take the following steps.
 
 1.	Open the SQL analytics endpoint or warehouse, go to **Reporting** and then select **New semantic model**.
 2.	Give your semantic model a name, pick a Fabric workspace for it, and select the tables to include. Then press **OK**.
 
-The semantic model is created and now you are live editing the modeling in the browser.
+The semantic model is created and now you're live editing the modeling in the browser.
 
 > [!NOTE]
 > Check your pop-up blocker if web modeling doesn’t appear after clicking OK.
@@ -46,10 +46,10 @@ The differences between Direct Lake on OneLake and Direct Lake on SQL are explai
 
 To edit a semantic model with Direct Lake tables later, take the following steps.
 
-1.	Navigate to the semantic model in the Fabric Portal. **Home**, **Onelake catalog**, and **search** at the top of the page are available to help you find it.
+1.	Navigate to the semantic model in the Fabric Portal. **Home**, **OneLake catalog**, and **search** at the top of the page are available to help you find it.
 2.	Select the semantic model to open the details page, or use the context-menu, then select **Open data model**.
 
-Now you are live editing the semantic model in the web. The model opens by default in **viewing mode** to avoid accidental edits. Change to **editing mode** in the top right-hand corner of the window. **Edit in Desktop** is also available to change to [live edit in Power BI Desktop](direct-lake-power-bi-desktop.md).
+Now you're live editing the semantic model in the web. The model opens by default in **viewing mode** to avoid accidental edits. Change to **editing mode** in the top right-hand corner of the window. **Edit in Desktop** is also available to change to [live edit in Power BI Desktop](direct-lake-power-bi-desktop.md).
 
 :::image type="content" source="media/direct-lake-web-modeling/web-modeling-mode.png" alt-text="Screenshot of edit in desktop from web modeling option." lightbox="media/direct-lake-web-modeling/web-modeling-mode.png":::
 

docs/real-time-intelligence/.openpublishing.redirection.real-time-intelligence.json

@@ -1,5 +1,10 @@
 {
   "redirections": [   
+    {
+      "source_path_from_root": "/docs/real-time-intelligence/event-streams/whats-new.md",
+      "redirect_url": "/fabric/fundamentals/whats-new#real-time-intelligence-in-microsoft-fabric",
+      "redirect_document_id": false
+    },        
     {
       "source_path_from_root": "/docs/real-time-intelligence/data-activator/activator-basic-concepts.md",
       "redirect_url": "/fabric/real-time-intelligence/data-activator/activator-introduction",

docs/real-time-intelligence/event-streams/whats-new.md

@@ -87,8 +87,6 @@ items:
     items:
     - name: Eventstream Overview
       href: event-streams/overview.md
-    - name: What's new in Fabric event streams?
-      href: event-streams/whats-new.md
     - name: Eventstreams capacity consumption
       href: event-streams/monitor-capacity-consumption.md
     - name: Create eventstreams

docs/real-time-intelligence/toc.yml

@@ -6,7 +6,7 @@ ms.author: mimart
 ms.reviewer: danzhang
 ms.topic: how-to
 ms.custom:
-ms.date: 08/18/2025
+ms.date: 09/22/2025
 
 #customer intent: As a workspace admin, I want to configure workspace-level Private Link on my workspace to prevent access to the workspace from the public internet.
 
@@ -162,11 +162,24 @@ Next, access Fabric privately from the virtual machine you created in the previo
 
 ## Step 8. Deny public access to the workspace
 
-You can optionally deny public access to the workspace. When the workspace is set to deny public access, it means the workspace can only be accessed via workspace-level private link. You create the access rule to deny public access using an API. Currently, the Fabric portal UI doesn't contain a setting for creating the deny public access rule. You must use the API to set the rule.
+You can optionally deny public access to the workspace. When the workspace is set to deny public access, it means the workspace can only be accessed via workspace-level private link. You can use either the Fabric portal or the Microsoft Graph API to create the access rule for denying public access.
 
 > [!NOTE]
 > The workspace-level setting to deny public access can take up to 30 minutes to take effect.
 
+### [Fabric portal](#tab/fabric-portal)
+
+1. In the Fabric portal, select the workspace you want to configure.
+1. Select **Workspace settings**.
+1. Select **Inbound networking**.
+1. Under **Workspace connection settings**, select **Allow connections only from workspace level private links**.
+
+   :::image type="content" source="media/security-workspace-level-private-links-set-up/workspace-connection-settings.png" alt-text="Screenshot showing the Workspace connection settings with the radio button selected for the option Allow connections only from workspace level private links." lightbox="media/security-workspace-level-private-links-set-up/workspace-connection-settings.png":::
+
+1. Select **Apply**.
+
+### [API](#tab/api)
+
 Use the [Workspaces - Set Network Communication Policy API](/rest/api/fabric/core/workspaces/set-network-communication-policy) to set the workspace public access rule:  
 
 `PUT https://api.fabric.microsoft.com/v1/workspaces/{workspaceID}/networking/communicationPolicy` 

docs/security/media/security-workspace-level-private-links-set-up/workspace-connection-settings.png

@@ -6,7 +6,7 @@ ms.author: mimart
 ms.reviewer: danzhang
 ms.topic: overview
 ms.custom:
-ms.date: 08/21/2025
+ms.date: 09/22/2025
 
 #customer intent: As a workspace admin, I want to get more information about how to use workspace-level private link in supported and unsupported scenarios.
 
@@ -255,9 +255,9 @@ A virtual network data gateway must be used for every dataflow connector. The vi
 
 ## Supported and unsupported tools
 
-- Workspace-level private link connections are supported via REST API.
-- The Fabric portal doesn't currently support workspace-level private links. If a workspace allows public access, the Fabric portal continues to function using public connectivity. If a workspace is configured to deny inbound public access, the Fabric portal displays an **Access restricted** page.
-- SQL Server Management Studio is supported for connecting to warehouses via private link.
+- You can use either the Fabric portal or the REST API to manage all [supported item types](#supported-item-types-for-workspace-level-private-link) in workspaces with private links enabled. If a workspace allows public access, the Fabric portal continues to function using public connectivity. If a workspace is configured to deny inbound public access, the Fabric portal displays an **Access restricted** page.
+- Direct deeplinks to a Monitoring hub Level 2 (L2) page might not work as expected when using workspace-level private links. You can access the L2 page by first navigating to the Monitoring hub's Level 1 (L1) page in the Fabric portal.
+- SQL Server Management Studio (SSMS) is supported for connecting to warehouses via workspace-level private link.
 - Storage Explorer can be used with workspace-level private links.
 - Azure Storage Explorer, PowerShell, AzCopy, and other Azure Storage tools can connect to OneLake via a private link.
 - To use OneLake File Explorer, you must have access to your tenant, either via public access or a tenant private link.  
@@ -274,6 +274,12 @@ A virtual network data gateway must be used for every dataflow connector. The vi
    - To query Lakehouse files or tables from a workspace that has workspace-level private link enabled, you must create a cross-workspace managed private endpoint connection to access resources in the other workspace. <!--For instructions, see [Cross workspace communication](security-cross-workspace-communication.md).-->
    - You can use either relative or full paths to query files or tables within the same workspace, or use a cross-workspace managed private endpoint connection to access them from another workspace.
 - You could run into Spark issues in the following regions when outbound access protection is enabled for the workspace: Mexico Central, Israel Central, and Spain Central.
+- Dataflows currently don't support using a Fabric Warehouse or Fabric Lakehouse in the same workspace as either a data source or an output destination.
+- Current limitations for Private Link with an eventhouse:
+   - Copilot features: Machine learning workloads might experience limited functionality due to a known regression.
+   - Eventstream pull: Eventstream workloads don't currently support full polling functionality.
+   - Fabric doesn't currently support Event Hub integration.
+   - Queued ingestion via OneLake isn't currently available.
 
 ## Common errors and troubleshooting