Article · AI translation
Thanks to IT Home readers 刺客, 软媒新友2131549, and 谁也不知道 for the tips!
IT Home reported on November 13 that, on November 12 local time, several core Microsoft 365 cloud services suffered a widespread outage, affecting products including Outlook, Exchange Online, Microsoft Teams, OneDrive, SharePoint, and Microsoft 365 Copilot. Users in multiple regions worldwide reported being unable to access the services normally; some could not send or receive email, load their mailboxes, or access cloud files.
After the outage began, monitoring site Downdetector received a large number of user reports, with reports of Outlook problems at one point approaching 130,000. In addition to email services, Microsoft Teams, Copilot, and other Microsoft 365 services experienced connection problems to varying degrees.
Microsoft initially said it was investigating problems encountered by users in the Americas when accessing Exchange Online mailboxes through various connection methods. The company later broadened its description of the impact, confirming that users anywhere in the world could be affected.
According to service status information published by Microsoft, affected services included Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot, Microsoft Defender XDR, Microsoft Planner, and Microsoft To Do. Some OneDrive users also reported being unable to access cloud files normally.
Regarding the cause of the outage, Microsoft said some infrastructure had experienced abnormally high load, causing connection errors for users. Further investigation found that the problem was not a sudden increase in overall traffic; rather, a large number of servers unexpectedly became unavailable, placing greater load on the remaining servers.
Microsoft said some servers failed automatically run health checks and were therefore deemed unavailable. The company is investigating the specific cause of the health check failures while reviewing recent configuration changes, and said some configuration updates may be related to the outage.
To mitigate the impact, Microsoft has paused all deployments and configuration changes, adjusted the distribution of connection traffic across its global infrastructure, and begun restarting affected servers to gradually restore service capacity.
As of 5:00 a.m. Beijing time on November 13, Microsoft had verified an alternative health check method on some affected servers, which were successfully restored to operation, and then began rolling out the approach more widely. IT Home noted that Microsoft was also preparing a mechanism to automatically detect and restart unresponsive servers.
The Microsoft outage also affected some administrative functions, and administrators may be unable to access the Microsoft 365 admin center normally. Microsoft advised administrators who cannot access the admin center to get the latest updates from its public service status page.