Microsoft says progress made fixing long outage that hit Outlook, Teams and Copilot

Updated 7:38 p.m. ET, Nov. 13

Key points

• Microsoft said it is making progress restoring Outlook, Teams and Copilot after an outage that began Thursday afternoon.
• Microsoft traced the failure to a bug from a network capacity upgrade and ruled out a cyberattack, while Outlook reports peaked near 140,000.
• MSFT closed up 0.17% today and is down 0.39% after hours, while emails in the backlog can reach recipients until Saturday afternoon.

Microsoft said on Friday it was making progress restoring its flagship workplace software, as an outage that began on Thursday afternoon ran into a second day. Outlook, Teams and Copilot were among the products hit, and the company described its recovery as encouraging across a growing part of its infrastructure.

The disruption reached government offices, schools, traders and businesses in many countries. Hospitals, courts and police forces in the UK were among the public bodies that had to switch to phones, fax machines and in-person delivery.

Recovery and the email backlog

Recovery has been uneven. In morning service updates at 6 a.m. and 7 a.m. ET, Microsoft reported progress across Europe and the Asia Pacific region, with Japan described as comparatively more stable. By 7 a.m., services in North America had also improved, though the company warned that some users may still experience intermittent issues as recovery actions continue.

To protect delayed messages, Microsoft extended its email timeout window to 48 hours. Under a standard window, mail servers would bounce undelivered messages back to senders. Under the longer window, emails held in the backlog since Thursday afternoon can still reach recipients until at least Saturday afternoon, once systems recover.

By Friday morning some Outlook users could send and receive email again, while others still could not. Microsoft has also cautioned that services already restored can become unresponsive again, and that customers in any region could be affected.

Cause and what Microsoft said on record

Microsoft said the incident came from a bug introduced during a network capacity upgrade. It also said a number of servers were unexpectedly unavailable, which strained the remaining infrastructure. The company ruled out a cyberattack and said it had found no link to AI agents.

"We are deeply sorry for the disruption this incident has caused our customers and partners." Microsoft said it would publish a full report on the root cause once its internal investigation is complete.

According to a person familiar with the internal response, senior engineers and executives spent the night on a video call weighing fixes and deciding which services to switch off so that restored systems would not be overwhelmed, according to The New York Times.

Who was hit and how

Copilot, the AI assistant inside Microsoft 365, was badly hit, and people could not load conversations, get generated replies or retrieve information from their mail, calendar entries and files. Calendar viewing, meeting invites and event edits were also hit, and external file sharing in SharePoint Online was blocked. Messaging and online meetings were largely functional, according to Microsoft's own description of the outage.

UK health services were hit hard. Senior health sources described disruption as serious and widespread, with hundreds of hospitals, GP surgeries and community services affected, according to The Guardian. Some staff used paper patient records after losing access to email, and some cancellations were made.

Airlines were affected as well. IndiGo and Akasa Air reported problems with customer-support email, and Air India temporarily rerouted its email services. OpenAI and Anthropic reported issues with their Microsoft 365 integration tools on Friday morning, and both said on Friday afternoon that those issues were beginning to subside.

Outlook problems peaked at nearly 140,000 reports on the Downdetector monitoring site around 18:00 GMT on Thursday, according to BBC. Microsoft also reported an outage of its Azure cloud platform last month that affected its productivity suite, making this the second large disruption to its services in recent weeks.

MSFT shares closed up 0.17% today and are down 0.39% after hours as of 7:37 p.m. ET.

What's next: Emails held in the backlog can reach recipients until at least Saturday afternoon, when the extended 48-hour timeout window runs out.

Sources: Axios, BBC, Bloomberg, CNBC, Financial Times, Fox Business, MarketWatch, The Guardian, The New York Times