Connect with us

NEWS

Microsoft 365 Outage Took Copilot and Police Offline

A 13-hour Exchange Online failure took Copilot, Safe Links and police reporting with it, leaving governments on phones, paper and fax.

Published

on

A Microsoft 365 outage that began at 16:17 UTC on November 12 stretched past 13 hours, knocking Exchange Online, Outlook, Teams calendars and Copilot offline in 74 countries. Microsoft said a process that sets up secure connections on backend servers became overwhelmed, then shifted load onto the machines that were still up.

The company said it was not aware of malicious activity. Restored servers went unresponsive again once traffic returned, and a full restore time was not given on Friday, November 13.

The Fix Kept Losing Servers It Had Just Restored

Microsoft opened Exchange Online incident EX1550543 after mailbox access failed in the Americas on Outlook on the web, desktop Outlook (MAPI), Exchange ActiveSync and REST connections. Within two hours it warned that users in any region could be hit. A wider Microsoft 365 incident, MO1550821, followed about 1:26 p.m. Eastern Time as Teams, SharePoint Online, Copilot, Defender XDR, Intune, Power BI and Purview were pulled in.

The first story was capacity. The next one was missing servers. On the public Microsoft 365 service health dashboard, the company wrote that elevated load was not coming from more traffic, but from a large number of servers that had become unexpectedly unavailable. A later update named the failing piece: a process on those servers that establishes secure connections, which then stopped processing requests.

Engineers restarted kit, rolled back a recent network change that lined up with the start of the trouble, and tried an alternate health check that brought a subset of servers back. Some of those machines went dark again under load. Overnight updates said the latest changes were “not taking as expected.” Microsoft also stretched the email timeout window to 48 hours so delayed mail would be held rather than dropped. Downdetector logged more than 130,000 Outlook problem reports at the U.S. peak.

WHAT WE KNOW

  • The start: Exchange Online failed at 16:17 UTC on November 12, 2026, first in the Americas, then worldwide.
  • The scope: Features that talk to Exchange Online were hit; parts that do not depend on mail stayed up.
  • The cause ruled out: Charles Lamanna, Microsoft’s executive vice president of Copilot, Agents, and Platform, said there was no indication of malicious or cyber activity.

WHAT IS UNCONFIRMED

  • Root cause: Microsoft has promised a post-incident report after mitigation and has not published one.
  • Full restore time: Status updates on Friday described phased recovery and capacity still below normal, with no ETA.
  • Lost mail: Delayed delivery is confirmed; claims of deleted inboxes are not.

The public status page existed for a blunt reason. Microsoft said administrators might be unable to reach the admin center, so the same incident notes had to sit in the open.

THE INCIDENT CLOCK

  1. November 12, 2026, 16:17 UTC: EX1550543 opens. Users cannot open mailboxes through web, desktop, mobile or REST.
  2. November 12, about 1:26 p.m. ET: MO1550821 covers Teams, SharePoint, Copilot, Defender XDR and other Exchange-dependent features in every region.
  3. November 12, evening: Servers return to service, then go unresponsive again once they take traffic.
  4. November 13, Friday: Microsoft reports early signs of stability in some regions, extends the mail timeout to 48 hours, and still gives no full-restore time.

That loop, servers coming back and then falling over, is what stretched a mailbox fault into a second working day.

Nineteen Microsoft 365 Services Depend on Exchange Online

Microsoft’s status page listed 19 Microsoft 365 services as affected, and it described one failure, not 19. “Features of the Microsoft 365 services below that interact with Exchange Online and email services are impacted,” the page said. “Elements of those services that do not depend on interaction with Exchange Online remain functional.”

Teams calling and chat mostly held. File copies sitting in SharePoint still opened. Everything that needed a mailbox, a calendar, a Safe Links check or a Copilot read of work mail did not.

WHAT STILL WORKED, AND WHAT DID NOT

Service Still working Broken or delayed
Exchange Online / Outlook On-prem or disconnected apps Web, desktop, ActiveSync and REST mailbox access; delayed mail
Microsoft Teams Calls, meetings, group chat, channel messages Calendar, scheduling, live/virtual event creation
SharePoint Online Files stored directly in SharePoint Features that call Exchange Online
Microsoft 365 Copilot None of the work-data path Chat load, responses, access to email, calendar and files
Defender for Office 365 Broader Defender tooling not tied to mail Safe Links URL checks, quarantine release, Threat Explorer
Microsoft Purview Stores that do not touch Exchange Auto-labeling, DLP, eDiscovery, protected content
Microsoft Graph Calls that avoid mail data Mail, calendar and contacts API timeouts

Safe Links is the quiet one in that table. Microsoft said users who clicked wrapped URLs in Outlook and Teams would fail to navigate, because Safe Links could not process the click. A supplier invoice that had already been rewritten did not open. Messages sitting in quarantine could not be released. Purview’s long status entry covered auto-labeling, data loss prevention, communication compliance and eDiscovery. Graph timeouts meant third-party tools that read calendars inherited the same hole.

This is the second-order hit. The outage was filed as mail. The blast radius was every product that treats the inbox as shared plumbing, including the security controls that sit on that plumbing.

Police Told the Public to Call 101 Instead

NHS England told patients to attend appointments unless they heard otherwise. Local NHS teams moved to other IT systems or paper records. The GMB union’s national secretary, Rachel Harrison, said the outage was having a “serious impact” across ambulance and NHS trusts. London Ambulance Service said its critical platforms do not use Microsoft and were unaffected.

In London, the Metropolitan Police said a lack of email was hitting its ability to take some crime reports online, though call response was not affected. Greater Manchester Police, North Yorkshire Police, South Wales Police, Dyfed-Powys Police and Police Scotland issued similar notes: skip the web form, call 101 for time-critical but non-emergency reports.

Australia woke up inside the same incident. Down Detector logged issues just before 5 a.m. AEST, with more than 5,500 reports peaking at 9:22 a.m. Home Affairs Minister Tony Burke said Microsoft did not suspect a cyber attack and that the impact was “very much on e-mail systems.” He compared the event to the 2024 CrowdStrike update failure. ASIO director-general Mike Burgess and Australian Federal Police commissioner Krissy Barrett said their agencies were affected. Burgess put it on the email system. Lieutenant General Michelle McGuinness, the national cyber security co-ordinator, was monitoring. Burke said Microsoft believed it might be “an update that has gone awry.”

WHERE WORK CONTINUED WITHOUT EMAIL

  • UK police: Forces directed the public to 101 after online crime forms and live chat failed.
  • NHS clinics: Staff used other systems or paper; patients were told to keep appointments.
  • Euronext: Listing teams told companies to use the phone for corporate information.
  • Maersk: The carrier said some emails and updates may be delayed and pointed customers to local office phone numbers.
  • Taiwan desks: One securities firm cancelled a bond repurchase; another confirmed trades by fax, according to people familiar with those desks.

A.P. Moller-Maersk stressed that the fault was Microsoft’s, not its own network, and asked customers to verify odd messages with a local representative. Aviva activated backups for earnings calls. Workhorse Group moved its results from November 12 to November 16, citing the disruption. Richemont briefly delayed its first-half figures on Friday. IndiGo, Akasa Air and Air India said email support was hit. OpenAI said new Microsoft 365 connections from ChatGPT might be broken, and Anthropic reported connector trouble.

Major UK airports and UK intelligence services were described as unaffected. That split is the point. The organisations that could still function had a path that did not run through Exchange Online. The ones that could not fell back to 101, a desk phone, paper notes or a fax machine.

The Assistant Could Not Open Work Mail

Copilot’s status line was a double miss. Users might be unable to load Copilot Chat or receive responses in the web and desktop apps, Microsoft said, and Copilot experiences might be unable to access work data such as email, calendars and files. The assistant Microsoft has been selling as a layer on top of work mail went dark with that mail.

Satya Nadella, the chief executive, told analysts on July 29, 2026 that Microsoft had over 30 million paid Microsoft 365 Copilot seats at the end of June. The company last counted more than 450 million paid commercial Microsoft 365 seats in its Q2 fiscal 2026 report, covering the period to December 2025. Thirty million is about 6.7% of that base. Those Copilot seats still need Exchange if the pitch is that the bot reads your inbox.

Charles Lamanna posted the company’s security line while recovery was still failing to stick.

“We regret that today’s outages caused disruptions for our customers, and we are working to address the issue,” he wrote. “For now I can confirm that there is no indication this is the result of malicious or cyber activity.”

Enterprises have priced Microsoft 365 as a licence and never as a single point of failure.

Sanchit Vir Gogia, chief analyst, Greyhound Research

Gogia’s follow-on was practical. The answer, he said, is not a second suite. It is a rehearsed fallback for the few things that cannot stop: an out-of-band channel for incident command, mail continuity for critical roles, and a written decision on whether each security control fails open or closed. Safe Links failing closed, links that simply do not open, is that decision made by accident.

Microsoft shares fell 1.7% in premarket trading on Friday, lagging the wider market, after a gain of about 8% in 2026. Josh Gilbert, a market analyst at eToro, said a repeat reliability problem “hands every IT department a reason to think twice before putting even more of their business in one provider’s hands.” Thinking twice is cheaper than moving. Cledara’s October 2026 payment data found that 67.5% of companies that also pay for Microsoft 365 while paying for Google Workspace already hold both. Dual bills did not give those firms a live mail path on November 12 unless someone had actually wired the second suite for failover.

No Second Mail Path After the 2025 Cutoff

Microsoft has spent years telling customers that the best move is to leave their own Exchange servers. Its own documentation still lists larger mailboxes, anti-spam, compliance tools and “you don’t have to worry about purchasing and maintaining hardware” as reasons to migrate. Support for Exchange Server 2019 and 2016 ended support on October 14, 2025. After that date Microsoft stopped security fixes, bug fixes and technical support on those builds. A six-month extended security update program ran only to April 14, 2026. The remaining on-prem road is Exchange Server Subscription Edition, shipped July 1, 2025, or a hybrid stub kept so Active Directory can still manage cloud recipients.

Shops that still host mail on their own servers were somewhat spared on November 12, which is the control group this outage handed every board. They had a path that did not share fate with EX1550543. Everyone who completed the migration Microsoft recommends did share that fate, including governments that also buy the security add-ons that sit on the same mail layer.

This was not Microsoft’s first long night in 2026. StatusGator timed a January 22 Microsoft 365 incident at 10 hours and 50 minutes, with Outlook, Defender and Purview in the mix. On October 29, 2025, an Azure Front Door configuration fault ran from 15:41 UTC to 00:05 UTC on October 30, 8 hours and 24 minutes, and took Microsoft 365 with it. Tony Burke called Friday’s event the third Microsoft global disruption of 2026. The January clock already showed that “resolved” on a status page can sit many hours after mail actually moves.

Some office workers treated the blackout as an unplanned quiet day. That reading only works if your job is not crime reports, clinic bookings, ship notifications or a bond desk. For those rooms, the continuity plan was analog. Fax confirming a trade in Taipei is not a colorful aside. It is what dual-cloud slides look like when nobody rehearsed the failover.

What a 13-Hour Outage Is Worth in Credits

Microsoft’s Service Level Agreement for Online Services, in the edition dated November 1, 2026, still promises 99.9% monthly uptime for Exchange Online, Teams, SharePoint and OneDrive, measured in user-minutes. Downtime for Exchange is defined as users being unable to send or receive mail with Outlook on the web. Credits are a cut on a later bill, only if a customer files, and they are the only contractual remedy. Lost sales, overtime and missed filings are outside the document. Microsoft does not apply the credit on its own. The claim window runs to the end of the calendar month after the incident, so December 31, 2026 for this one. Incident IDs to quote are EX1550543 and MO1550821.

MICROSOFT’S MONTHLY UPTIME CREDITS

Monthly uptime Service credit Downtime in a 30-day month
Below 99.9% 25% More than 43.2 minutes
Below 99% 50% More than 7 hours 12 minutes
Below 95% 100% More than 36 hours

A tenant that was down from Thursday afternoon through Friday morning is already past the 50% tier. It is not past 36 hours. The cheque, if it comes, is a slice of the Exchange Online fee, not a make-whole for a cancelled repurchase or a delayed earnings release.

Microsoft said on Friday it was seeing positive results from traffic changes in some regions and would keep applying them. Capacity had not returned to normal. Copilot still could not read the mail it is paid to summarise. Police still wanted the public on the phone. The post-incident report, when it lands, will have to explain why a secure-connection process on Exchange servers was allowed to take 19 products, and the agencies that bought them, with it.

Harry is the editor and publisher of MY WORLD NEWS 24, an independent title under his own ownership. Ten years of reporting and then editing taught him that a global readership is not served by assuming everyone lives in the same country. Stories here state currencies, units and time zones explicitly, name the country a law or a company belongs to, and explain local context rather than treating it as known. That care extends to sourcing: a claim is anchored to the filing, statement, transcript or dataset that made it, wherever in the world it was issued, and each figure is checked against that source before publication. The site reports news, business and technology, science and sports, entertainment, lifestyle and travel, and auto and gaming, all with the same standard of evidence. Mistakes are fixed under a corrections policy anyone can read, and the page carries a note saying what was changed. Harry reads every message sent by readers and replies from support@myworldnews24.com.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending