The Microsoft 365 outage that knocked out Outlook and Teams was still not fixed early Friday. It began more than 18 hours earlier. In its 5:22 a.m. ET update, Microsoft said service health “remains variable” and capacity had not fully returned to normal, NBC News reported.
The cause, so far, is Microsoft’s own infrastructure. The company says many of its servers dropped out and overloaded the rest. It has found no sign of a cyberattack, and it hasn’t named a root cause.
Mailboxes failed first, then the rest of Microsoft 365
Microsoft logged the first problem as Exchange Online incident EX1550543, starting at 11:17 a.m. ET on Thursday, Nov. 12. Users couldn’t reach their mailboxes through Outlook on the web and other connection methods.
Earlier that morning, a separate Teams messaging fault in the Americas, TM1550488, ran for about an hour before Microsoft fixed it, BleepingComputer reported. Microsoft hasn’t linked the two.
At 1:26 p.m. ET, Microsoft opened a wider incident, MO1550821. It covered Exchange Online, Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR. Through the evening Microsoft added Planner, To Do, Intune, Purview, Power BI and Windows 365, according to ENow Software’s log of Microsoft’s service health posts.
Microsoft first described an Americas problem, then said users in any region could see it. Downdetector logged reports from 74 countries and territories within about an hour, NBC reported. At the peak it took 19,650 reports a minute, against fewer than 15 on a normal Thursday.
Microsoft’s stated cause: servers dropped out and dragged others down
Microsoft’s explanation has shifted as it learned more, so the order matters. It first blamed heavy load across several regions. Then it said the load came not from extra traffic but from “a large number of servers being unexpectedly unavailable,” ITdaily reported.
Those servers were failing automated health checks, which pulls them out of service. Microsoft paused all deployments, moved traffic off the failing machines and restarted them. That didn’t hold. “Our actions taken thus far have not provided the anticipated relief,” the Microsoft 365 Status account posted on X. “We found a core process on affected servers is becoming overwhelmed, causing capacity loss.”
Servers it brought back stopped responding again once traffic returned, the ENow log shows. Microsoft also began reverting a recent network change. ITdaily reported that Microsoft says the change coincided with the start of the problem. Microsoft hasn’t said it caused the outage.
Overnight, Microsoft patched an upstream authentication and routing component on part of its systems. By its 2:30 a.m. ET update it was “seeing positive results for server stability,” NBC reported. Microsoft said it was “not aware of any malicious activity that is causing this incident.” Charles Lamanna, an executive vice president at Microsoft, wrote on X that there was “no indication this is the result of malicious or cyber activity,” ABC News reported.
Governments, banks and airlines lost email, not flights
The damage was mostly to communication. Japan’s Chief Cabinet Secretary Minoru Kihara said “multiple services, including the Outlook email software, have become unavailable,” NBC reported. He said medical care and patient records had no problems. Japan’s ministries went back to paper for much of Friday, The Next Web reported.
Bankers at Morgan Stanley and Goldman Sachs in Hong Kong told CNN they couldn’t use email or Teams all day, The Next Web reported. Standard Chartered and ING Australia warned customers about scammers posing as tech support. The Philippine Supreme Court warned of delays.
In the US, staff in Austin and Baltimore County couldn’t send or receive email, but 911 lines kept working, NBC reported. Miami Beach asked residents to phone instead. A Colorado sheriff’s office reported delays to prisoners’ bond payments, according to The Next Web.
India’s IndiGo, Akasa Air and Air India lost email support channels, NBC reported. We found no report of flights cancelled because of this outage. Workhorse postponed its earnings release, and Maersk warned of email delays. Microsoft 365 connectors in ChatGPT and Anthropic’s Claude stopped working too.
What users and IT admins can do now
Microsoft extended its email retry window to 48 hours to protect delayed messages, NBC reported. Mail sent during the outage may still arrive, so check before you resend.
Admins have four things to do, based on Microsoft’s own documents.
- Track the incident IDs. In the Microsoft 365 admin center, open Health, then Service health, and find MO1550821 and EX1550543. You can sign up there for email updates on one incident, Microsoft’s guide says.
- Use the public page if you can’t sign in. Microsoft points locked-out admins to status.cloud.microsoft and the @MSFT365Status account on X.
- Watch for the root-cause report. After incidents with broad impact, Microsoft posts a preliminary review within 48 hours of the fix. A final one follows within five business days, its service description says.
- Read the SLA before you claim credit. Microsoft’s Service Level Agreement for Online Services covers Office 365. It sets the uptime terms and the claim process, and it decides whether this outage qualifies.
For everyone else, the banks’ advice applies. Treat any “Microsoft support” email, text or call offering to fix your account with suspicion. If your office runs AI tools at work through Microsoft 365 connectors, check them again once mail is back.
The next marker is a “service restored” status
Microsoft hadn’t marked either incident resolved as of early Friday US time. It has promised full details of the root cause once the problem is fixed. Its preliminary report should then reach admins’ Service health pages within 48 hours.
