Microsoft Fabric outage history

Users across the globe experienced refresh issues for Gen1 Dataflows in Power BI on Microsoft Fabric. Refresh operations could remain stuck in an In Progress state, impacting dashboards and data workflows across Americas, Europe, Asia Pacific, Africa, and the Middle East. The disruption began at 7:39 AM UTC as engineers started investigating and applying mitigations.

An update at 10:29 AM UTC indicated that the root cause had been identified and mitigation work was underway. By 3:19 PM UTC, mitigation had been applied to several affected clusters while work continued on remaining clusters. By 3:29 AM UTC, all regional Power BI clusters had recovered and Gen1 Dataflows refresh was returning to normal operation.

The Central US region experienced delays when interacting with Microsoft Fabric services, impacting the Americas / Fabric and Americas / Power BI components. Users likely saw slower responses or intermittent timeouts while performing tasks in Fabric and Power BI. The disruption began for Central US tenants at Aug 6, 2026 8:14 PM UTC and persisted for roughly four and a half hours, with services returning to normal by Aug 7, 2026 12:39 AM UTC. Engineers indicated they were investigating the issue and would share updates as they learned more.

The outage began at 1:44 AM UTC when Fabric in the Americas entered a degraded state, affecting Brazil South users attempting to submit Spark jobs, run notebook operations, or render Python and R visuals. Over the next several hours, the disruption expanded to additional regions including UK West, Australia Southeast, and South India, with similar issues impacting Spark submissions, lakehouse operations, notebook tasks, and visuals. Engineers cited an Azure outage as the root cause as updates continued, with the issue evolving across regions. By 4:19 AM UTC, Fabric components in the Americas and Asia Pacific had moved back to up, signaling partial recovery for those regions.

Users in Central India experienced issues accessing artifacts in Microsoft Fabric and faced problems submitting Spark jobs, executing notebook operations, and rendering Python and R visuals. The incident began at 2:09 PM UTC when the Asia Pacific / Fabric component entered a degraded state, with engineers actively investigating and providing updates.

Over the next few hours, mitigation efforts were underway as updates confirmed ongoing impact in Central India. The issue remained focused on the Asia Pacific / Fabric region and was resolved when the component returned to an up state at 5:34 PM UTC. The outage lasted about three and a half hours before services recovered.

In West Europe, Microsoft Fabric services experienced a degraded interaction affecting the Europe Fabric and Europe Power BI components. The degraded state began at 1:04 AM UTC, with a status-page note at 1:09 AM UTC that West Europe users may be affected while engineers investigated. The degraded experience persisted for about six and a half hours as teams worked to resolve the issue and provide updates. The affected components returned to an up status at 7:29 AM UTC.

Users in the Americas region experienced degraded interactions with Microsoft Fabric beginning at 10:59 AM UTC in the WABI-US-CENTRAL-B-Primary area, affecting the Americas / Fabric and Americas / Power BI components. Access and performance during typical tasks were impacted while engineers investigated the issue and provided updates. The problem appeared regional, with no broad impact reported outside the specified region.

By 12:19 PM UTC, the issues were resolved and services returned to normal in the region, marking a duration of roughly 1 hour and 20 minutes. Fabric and Power BI components returned to regular operation, restoring normal user experience in the affected area.

Impact began at 3:39 PM UTC as West US tenants reported issues accessing Publish to Web reports, with intermittent delays affecting Power BI and Fabric surfaces. The disruption broadened to other regions, with intermittent access problems noted for artifacts and services in Europe and Asia Pacific. Tenants in Spain Central, New Zealand North and Australia SouthEast were specifically highlighted for delayed access. The outage occurred in the context of an Azure service disruption, and engineers were actively investigating the root cause and coordinating mitigations.

Over the afternoon, regional updates reflected gradual recovery in some areas and continued impact in others. By 6:19 PM UTC, Fabric components in the Americas had recovered to up, and by 7:04 PM UTC Power BI services in the Americas had returned to up. Nevertheless, the Publish to Web scenario remained actively impacted in the West US, and ongoing Azure dependencies meant some cross-region artifact access remained temporarily degraded. Other regions saw mitigations with Fabric and Power BI artifacts recovering in Europe and APAC as engineers continued to restore full service.

Power BI users in East Asia experienced latency in scheduled refreshes, delaying data updates for dashboards and reports. The issue affected the Asia Pacific / Power BI component, with latency observed in the East Asia region and specifically in the WABI-EAST-ASIA-A-PRIMARY region. Status updates indicated engineers were investigating the impact and would provide updates as they worked on a fix. The incident began at 3:59 AM UTC and was resolved by 7:04 AM UTC as the Asia Pacific / Power BI service returned to normal.

Across Africa, the Americas, Europe, Asia Pacific, and the Middle East, Microsoft Fabric users encountered replication issues in mirrored databases when any column contained invalid characters. The Fabric status page indicated a degraded condition while engineers worked to mitigate the problem, with an update promised soon. The incident began at Jul 19, 2026 12:04 AM UTC and impacted data replication workflows for those scenarios.

Fabric customers in the Northcentral US region may have experienced issues connecting to SQL endpoints and executing SQL queries for warehouse artifacts. This regional degradation affected the Americas / Fabric component and could slow data workflows that rely on warehouse queries. The incident was acknowledged by Microsoft Fabric and engineers were actively investigating, with updates provided as the situation evolved. Impact began at 3:09 AM UTC and was resolved by 6:04 AM UTC.

The service returned to normal after the resolution, indicating the issue was regional to Northcentral US and affecting SQL endpoints and query execution for warehouse artifacts as reported on the status page.