Why did Microsoft's servers suddenly crash?

Last night around 21.30 Türkiye time, most of the users who could not log in to Outlook or send messages in Teams asked the same question: Is the problem mine? It wasn't. Many services of Microsoft 365, including Outlook, Teams, OneDrive and Copilot, were disrupted for hours around the world. What's interesting is that the problem is not caused by a traffic explosion, but by Microsoft's own servers going down one by one.

Microsoft announced the outage in the Microsoft 365 admin center with the codes MO1550821 and EX1550543. According to the company, users may be unable to connect to Exchange Online (Outlook), Microsoft Teams, SharePoint Online, Microsoft 365 Copilot, and Microsoft Defender XDR. Users also reported that they were unable to access the internet version of OneDrive and Office.

The problem is not in the traffic, but in the servers

In its first statement, Microsoft said that infrastructure load has increased in more than one region. Then he clarified the picture: The load was not due to increased traffic, but to the unexpected downtime of a large number of servers. The burden of the shut down servers was placed on the remaining infrastructure, which further strained the services.

The company's statement from the Microsoft 365 Status account in X is as follows:

https://x.com/MSFT365Status/status/2120962673911181437

According to Microsoft, the servers are being taken out of service because they failed automatic health checks. The company explained that the failed checks may be caused by a specific component that processes health checks on the servers. He added that he could not find any recent changes that coincided with the start of the outage. Stopped all deployments and configuration changes as a precaution.

Over 100 thousand notifications on Downdetector

Complaints quickly reached their peak on Downdetector, which collects user notifications. According to Tom's Guide, at its peak, more than 100 thousand users reported problems. The impact of the outage was not limited to Microsoft; Notifications for Gmail, Zoom and Yahoo Mail also increased due to users looking for alternatives. The direct connection of these notifications to the problem at Microsoft has not been confirmed.

Microsoft confirmed that the issue initially affected Exchange Online users in the Americas, after which users in every region could be affected. This meant that users in Türkiye could also be affected.

Second Teams outage on the same day

This wasn't the first problem Microsoft experienced on the same day. Teams experienced a separate outage for users in North and South America between 17.45 and 18.53 Türkiye time. In that problem, messages were delayed or could not be sent. Microsoft attributed that outage to a resource crunch in the infrastructure and announced that it was completely resolved. The company did not say whether the two incidents were related.

When will services be restored?

Microsoft is restarting the affected servers. The company announced that an alternative health check method returned a group of servers to service and implemented the change on a larger scale. An automatic recovery system that will constantly detect and restart unresponsive servers is also being prepared.

https://x.com/MSFT365Status/status/2120985048337830246

Notification numbers started to drop after midnight, but Microsoft has not yet announced that the problem has been completely resolved. If you still cannot access Outlook or Teams, the problem is not with your device; Services are expected to improve gradually. You can follow the current status on Microsoft 365's status page.