Microsoft 365 outage hits Outlook, Teams and Copilot worldwide as servers drop out

Microsoft 365 suffered a worldwide outage affecting Outlook, Teams and Copilot, with the company saying a large number of its servers had become unexpectedly unavailable. Downdetector showed nearly 130,000 reports of problems, according to the Guardian, covering Word, Excel, PowerPoint, Outlook and Teams as well as SharePoint and OneDrive. The BBC reported more than 131,000 Outlook reports in the US at 18:15 GMT on Thursday.

Microsoft first acknowledged problems with Teams for users in North America, then widened the warning to users in any region, according to SBS News. It said a portion of its infrastructure was under elevated load, and that the load was not caused by a rise in traffic but by servers dropping out and shifting work onto the remaining infrastructure. As a precaution, the company paused all deployments and configuration changes while it reviewed recent changes.

Microsoft said the affected servers were failing automated health checks and becoming unavailable, and it began restarting them. It later said it had validated an alternative health check method on a subset of servers, which allowed them to return to service, and that it was applying the change more broadly. The company also said a review had identified a small number of configuration updates that it was investigating as possible contributors, after first saying it had found no change correlated with the start of the impact.

The fix did not hold. Microsoft said its actions had "not provided the anticipated relief" and that a core process on the affected servers was becoming overwhelmed, causing them to stop responding. It said it was isolating a portion of its infrastructure, restarting components and gradually reintroducing traffic, and was assessing whether isolating traffic within each region could let regions recover independently. The ABC later reported Microsoft had begun returning a portion of affected servers to service and that they were remaining stable so far. Microsoft said it was applying protective traffic management at one location and warned users might see gradual recovery as clients reconnected.

The ABC reported that the list of potentially impacted services grew to include Exchange Online, SharePoint Online, Defender XDR, Planner, To Do, Intune and several Purview services. OpenAI and Anthropic said the issue affected the Microsoft 365 connector linking ChatGPT and Claude to Microsoft services.

In Australia, the ABC reported that the home affairs minister's office said the National Cyber Security Coordinator, Lieutenant General Michelle McGuinness, knew of the outage and was examining it. Western Meat Exporters director Campbell McPhee said staff went "old school" and phoned to confirm orders after email went down. The School of the Air in Longreach cancelled online classes, and the Capricornia School of Distance Education reported students unable to reach lessons.

RMIT associate professor Mark Gregory said the outage appeared to affect Microsoft's network layer and was potentially similar to the configuration fault that brought down Optus's network in 2023. He said losing part of a server pool overloads the rest, creating "a cascading problem". Gregory said companies should invest more in testing environments for configuration changes, and that businesses reliant on cloud services should have backup options.