Microsoft Fabric Outage History

Microsoft Fabric experienced a regional issue affecting Reflex artifacts for tenants in Europe and the Americas. The incident began at 1:54 PM UTC when the Europe region was listed as a warning, and by 2:34 PM UTC the issue extended to Europe and US regions. Access to Reflex artifacts may have been intermittently affected during this window, with degraded performance in the affected regions. The problem was resolved by 6:34 PM UTC, with Europe and Americas components returning to normal and service restored. The outage lasted approximately 4 hours and 40 minutes. The provider issued 3 status updates during the incident.

Microsoft Fabric experienced a regional outage in the Canada Central (canadacentral) region, impacting Spark Jobs and Notebook operations for tenants located there. The Americas / Fabric component showed degraded service during the incident. The outage began at Sep 18, 2025 5:09 AM UTC when engineers started investigating and was resolved by Sep 18, 2025 6:24 AM UTC, lasting about 1 hour and 15 minutes. The provider issued 2 status updates during the incident, describing ongoing investigation and restoration of service. The impact was geographically limited to Canadacentral; other regions were not indicated as affected.

Microsoft Fabric experienced a partial outage affecting Power BI dataset refreshes for capacities in West US (Americas) and UK South (Europe) when using the Snowflake connector as a data source. Users attempting refreshes encountered errors such as 'The key didn’t match any rows in the table,' leading to failed or delayed data in dashboards and reports. The issue was limited to these regional Power BI capacities and did not indicate a broader service disruption across all Microsoft Fabric components.

The incident began at Sep 17, 2025 8:14 PM UTC with the Power BI capacities entering a degraded state. Mitigation was expected by Sep 17, 2025 10:00 PM UTC (15:00 PDT). The provider issued 2 status updates during the incident and service was restored by Sep 17, 2025 11:09 PM UTC, for a total duration of roughly 2 hours and 55 minutes. A regional pattern affected West US and UK South.

The outage began on Sep 10, 2025 12:49 AM UTC and was resolved by Sep 16, 2025 5:14 PM UTC. Microsoft Fabric and Power BI users across Asia Pacific, Americas, Europe, Africa, and the Middle East experienced replication failures and degraded service as Fabric attempted to replicate data between mirrored SQL Server databases after deletions. This led to disrupted data replication, slow or unavailable Dataflow refreshes, and issues with Spark jobs, notebook rendering, and data gateway connections. The incident affected multiple Fabric and Power BI regions and components, with a series of status updates indicating the ongoing nature of the issue over several days.

Microsoft Fabric in the Europe region experienced degraded access to notebook artifacts, with users reporting connectivity issues and slow notebook displays. The first user reports appeared around 9:45 AM UTC, and the Fabric team acknowledged the issue at Sep 9, 2025 10:44 AM UTC, listing Europe / Fabric as affected. The incident lasted until Sep 9, 2025 2:34 PM UTC when the service returned to normal.

Two official status updates were posted during the incident. The impact was focused on notebook artifacts access within the Europe region, affecting the Europe / Fabric component. User reports described connectivity problems and slow notebook displays, consistent with a notebook artifacts access issue.

Microsoft Fabric in the North Europe region experienced a brief degradation that affected notebook and Spark operations for tenants in North Europe. Users reported notebook session connection problems, including private-link-check failures observed in the UK (Manchester and Harrow). The incident began at 8:39 AM UTC and lasted about 40 minutes, with initial user reports appearing around 8:15 AM UTC. The impact was centered on the Europe / Fabric component and appeared regionally.

Microsoft Fabric experienced a regional disruption in Europe affecting access to workspaces and notebooks for tenants in the WABI-EUROPE-NORTH-B region. The issue began when the Europe / Fabric component transitioned to a degraded state at 9:29 AM UTC, with engineers initiating investigation and promising an update. The outage lasted about 50 minutes and was resolved when the Europe / Fabric component returned to normal at 10:19 AM UTC.

The provider issued two status updates during the incident, describing the investigation progress and restoration steps. User reports from Europe indicated users could not access workspaces and notebooks during the downtime. The disruption was regional to Europe, with no indication of a broader global impact.

Microsoft Fabric experienced latency and access issues when opening or accessing Dataflows artifacts in the Americas region. Users attempting to work with Dataflows artifacts may have faced delays or intermittent access problems. The incident began on Sep 4, 2025 5:44 PM UTC and the Fabric team indicated they were investigating, with an advisory noting latency. The impact was regional, affecting the Americas / Fabric component.

Microsoft Fabric reported intermittent issues in the Europe region affecting users opening notebook artifacts. The problem could cause the Power BI UI to stall during data loading. Engineers began investigating and promised an update. The incident started at Sep 4, 2025 9:04 AM UTC and lasted about 1 hour.

The outage affected Microsoft Fabric dataflows and semantic model refreshes, causing failures across Power BI and Fabric refresh jobs. Users experienced failed data refreshes, stale dashboards, and disrupted analytics workflows. The outage began around Sep 3, 2025 8:15 PM UTC as user reports accumulated, and the service remained in a degraded state until Sep 3, 2025 11:09 PM UTC.

Two official status updates were issued during the incident. The provider indicated issues with dataflows and semantic model refreshes and that engineers were investigating, and later confirmed that services had returned to normal by Sep 3, 2025 11:09 PM UTC. The incident affected 10 components spanning Fabric and Power BI across multiple regions.