Microsoft investigates server failure as global outage spreads to Asia-Pacific

Microsoft was investigating a possible internal server problem after a major outage disrupted access to its services worldwide on Thursday. Thousands of customers reported connection failures involving Outlook, Microsoft 365 and Teams. By Friday, November 13, disruptions had also been reported in the Asia-Pacific region, and the company had not issued a final resolution.

Microsoft's initial statements pointed to an error that disconnected accounts from servers in North America. Those accounts were then redirected to servers elsewhere in the world, potentially spreading the disruption. Preliminary estimates suggested the additional traffic placed pressure on the receiving servers, causing delays and further service failures. Persistent connection problems led some companies to send employees home early.

The early technical assessment indicated that some servers had failed a health check, triggering the disconnections and the redistribution of users. Experts said Microsoft would likely need to pinpoint the server whose failed check set off the wider outage. The precise cause remained unconfirmed, pending the company's internal investigation and a public status report.

Most experts discounted a cyberattack as the explanation, although the investigation was not complete. Microsoft's description of its response focused on problems within its own server infrastructure. Reports also generally showed normal user levels, without a surge suggesting a Distributed Denial-of-Service attack. No hacker group had claimed responsibility despite several hours of disruption and thousands of reports from affected users.

Israel was among the countries experiencing problems. Downdetector recorded disruptions in Tel Aviv, Jerusalem and Haifa. Alongside Outlook and Microsoft 365, Microsoft's artificial intelligence assistant Copilot also experienced service difficulties. The worldwide incident had lasted more than 12 hours, with the latest reports on Friday showing that users in Asia-Pacific were encountering problems as well.

A smaller Microsoft outage in January lasted nearly nine hours and generated about 3,000 reports on Downdetector. The public assessment in that case linked the disruption to maintenance that took a data center in Cheyenne, Wyoming, offline. That incident also produced a cascading effect across services.

The Microsoft problems followed another major disruption involving Amazon Web Services in October 2025. That outage affected thousands of websites, popular applications including Snapchat and Reddit, and businesses around the world. It was similarly attributed to an internal error rather than a cyberattack.

During the AWS incident, software engineer and cyber expert Junade Ali, a Fellow at the Institution of Engineering and Technology, told Reuters that the problem appeared to involve a networking system used to control a database product. He said a centrally managed fix could mitigate the issue within hours, provided no further problems emerged.