Worldwide Microsoft 365 outage: Outlook, Teams and Copilot fail for almost a day after a network change

Technology

A Microsoft 365 outage that began on Thursday, November 12, left users around the world without access to Outlook, Teams, SharePoint and Copilot for almost a day. Microsoft declared the service restored on November 13, rules out a cyberattack and blames an issue that surfaced after a capacity upgrade to its network.

The incident coincided with one of the busiest news days of the year in tech and hit businesses, public bodies and schools that rely on Microsoft's cloud email and meetings. The company has not said that Azure was affected.

What happened

Times are US Eastern (add five hours for UTC):

  • November 12, 11:17 AM: Microsoft opens incident EX1550543 because some users cannot connect to their Exchange Online mailboxes.
  • 12:54 PM: first public notice, limited at that point to the Americas. Shortly afterwards the company says any region may be affected.
  • 1:26 PM: a second, broader incident, MO1550821, is opened for connection failures across several Microsoft 365 services.
  • 8:06 PM: Microsoft starts reverting a recent network configuration change.
  • November 13, 11:10 AM: user impact ends. That night, at 11:17 PM, the company marks both incidents as resolved.

Which services failed

Microsoft named Exchange Online and Outlook, Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR. Sources that tracked the official advisory add Intune, Purview, Planner, Power BI and the admin centers. In Teams, chat and ongoing meetings mostly kept working, but calendars and invitations failed.

On Downdetector, user reports about Outlook passed 130,000 at the peak; sources give figures between 130,000 and 138,000. These are reports submitted by users, not a count of affected accounts.

The cause, according to Microsoft

The explanation shifted during the day. At first the company said the problem was not caused by excess traffic but by a number of servers becoming unexpectedly unavailable, and that no recent change matched the start of the incident. Hours later it explained that the process that establishes secure connections had been overwhelmed, and it began rolling back a network change.

In its final statement, Microsoft says the fault "was introduced by an issue during an earlier network capacity upgrade" and surfaced following a recent network change. It also stresses that it was "a technical issue not related to cyber or AI / agents". It has promised a preliminary report within two business days.

What to do if you were affected

  1. Don't change passwords, reinstall Office or delete Outlook profiles because of this: the problem was on Microsoft's servers.
  2. Check for late email: messages were delivered with delays, but nothing suggests they were lost. Microsoft extended the delivery retry window to 48 hours.
  3. Be wary of emails that exploit the outage to ask you to "verify" your account: it is a common phishing hook.
  4. Agree on a backup channel (phone or another messaging app) for next time.

What to watch next

  • Microsoft's post-incident report, which should explain why one network change took down so many services at once.
  • The credits provided for in service level agreements for business customers.

Conclusion: there was no attack, just a network change that went wrong and took almost a day to undo. For any organization that runs all its daily work on a single provider, it is a useful reminder of the value of a simple plan B.

Sources and documentation

Share