 |
| AI-generated illustration of a dark office skyline at dusk with a giant email inbox icon going dark, red warning lights, and a lone laptop showing an error screen |
On Thursday, November 12, 2026, email stopped working for a large slice of the working world. Shortly after noon Eastern time in the United States, Outlook inboxes stopped loading, messages would not send, and meetings would not open. Microsoft confirmed within the hour that its Exchange Online service — the cloud engine behind business Outlook — was failing across the Americas, and the problem soon spread to Teams, SharePoint, Copilot and Microsoft Defender. By 1:27 p.m. Eastern, more than 130,000 people had reported Outlook problems to the outage tracker Downdetector, with another 46,000 reporting Microsoft 365 issues. Microsoft spent the rest of the day restarting servers and moving traffic, and by late Thursday evening service was returning for most users — but the company had still not published a full explanation of what took its servers offline in the first place. This is what happened, what Microsoft has actually said, and what to do if your inbox is still acting up.
The outage matters because of how much of daily work now runs through a single cloud service. Outlook is not just an email app for most companies: it is the calendar, the meeting room, the address book and the paper trail for contracts, invoices and support tickets. When Exchange Online stops, the damage is not a missing app icon — it is missed deadlines, meetings nobody can join, and support queues that go silent. The timing made it worse. The outage arrived in the same week Microsoft was already patching an actively exploited Outlook security flaw, which made the rumours louder and the confusion deeper. The honest position, and the one Microsoft's own updates support, is narrower than the rumours: a large number of Microsoft's own servers became unexpectedly unavailable, nobody has shown it was an attack, and the full cause is still unknown. That is less dramatic than a hacking story, and it is also more worrying in its own way, because it means the failure came from inside the system that is supposed to keep email running.
As an Amazon Associate, we earn from qualifying purchases.
Prices and savings can change. Check the retailer for the current price.
What went down, and when
Microsoft's own incident log puts the start of the Exchange Online failure at 11:17 a.m. Eastern on Thursday, November 12. User complaints began rising sharply just after 12:30 p.m. Eastern, and Microsoft posted its first public acknowledgement at 12:54 p.m. Eastern, saying users in the Americas were having trouble reaching their Exchange Online mailboxes through multiple connection methods. The reports kept climbing through the early afternoon. Downdetector, which counts user-submitted problem reports, recorded more than 130,000 Outlook reports at 1:27 p.m. Eastern, more than 46,000 reports for Microsoft 365, and about 3,000 for Teams. Those numbers are reports, not a count of affected users — one person can report more than once, and most affected people never report at all — but the shape of the spike tells the story: a sudden, steep failure in the middle of the North American working day. Microsoft later widened the warning, saying the issue could affect Microsoft 365 service connectivity for users in any region, not just the Americas. By about 2:12 p.m. Eastern, Microsoft said it had paused all deployments and configuration changes, was moving traffic away from the affected infrastructure and was restarting servers. At 3:31 p.m. Eastern it said traffic had been redistributed across its worldwide infrastructure and the failed servers were being restarted. Through the evening, user reports fell and service came back for most people, although Microsoft had not declared the incident fully resolved in the updates reviewed at the time of writing, and its service health page still listed a service degradation late in the day. A separate, smaller Teams incident in North America earlier the same day was declared fully resolved at 12:08 p.m. Eastern, with Microsoft giving resource contention in underlying infrastructure as the preliminary cause. Microsoft has not said the two incidents share a cause, so they should be treated as separate until it does.
What Microsoft says caused it
Microsoft's explanation changed as the afternoon went on, and the revisions are worth reading carefully because they rule out the simplest theories. In its first updates, the company blamed heavy load on part of its infrastructure. Then it corrected itself: the elevated load was not driven by an overall increase in traffic, but by a large number of servers being unexpectedly unavailable, which was shifting load onto the remaining infrastructure. In other words, the system was not overwhelmed by too many users — it lost too many of its own machines, and the survivors could not carry the weight. Microsoft also said the failed health checks might originate from a specific component responsible for processing server health, and it tested a configuration change aimed at servers that were failing those automated checks. The practical response was the standard one for a cloud failure of this kind: pause deployments, move traffic off the affected systems, and restart the servers that had dropped out. What Microsoft has not done is name the underlying fault — why the servers became unavailable, whether it was hardware, software, a bad configuration or something else. That is the question its post-incident review will have to answer. Microsoft's own service-health documentation commits to a preliminary post-incident review within 48 hours of an incident being resolved, and a final review within five business days. Until that review appears, any claim about the precise trigger is speculation, including claims made by people who were not in the room.
Who was affected, and how bad it was
The official list of potentially affected services was long: Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender XDR. In practice, Outlook email was the centre of the outage — inboxes would not load, mail could not be sent or received, and calendar meetings would not open — while Teams, SharePoint and Copilot were listed as potentially affected rather than confirmed down for everyone. The impact fell hardest on businesses, schools and government offices in the Americas, where the failure landed in the middle of the working day, but reports also came in from Europe and from users elsewhere, and Microsoft itself warned that connectivity could be affected in any region. The knock-on effects reached beyond Microsoft's own apps. OpenAI said it was investigating broken Microsoft connectors, and Anthropic said an issue with an upstream provider was affecting its Microsoft 365 connector. That is the shape of a modern cloud outage: one company's email failure becomes another company's broken integration within the hour. For individual users the experience was mostly confusion — an error message, a spinning inbox, a meeting link that went nowhere — with no way to tell whether the problem was their own connection or something much bigger. For IT teams it was worse: a service they cannot fix themselves, with updates arriving on a public social-media account and in the admin centre, and no workaround to offer staff except to wait.
Was it a cyberattack?
Nothing Microsoft has published points to one, and that matters more than the rumours. Microsoft has not attributed the incident to an attack, and none of its updates describe malicious activity. The cause it has described — its own servers becoming unexpectedly unavailable — is an infrastructure failure, not a breach. That has not stopped speculation, and the speculation had a ready-made hook: the same week, Microsoft patched an actively exploited Outlook security flaw. The flaw, tracked as CVE-2026-124194, is a network-spoofing vulnerability in Outlook with a severity score of 6.5 out of 10, and the US Cybersecurity and Infrastructure Security Agency confirmed it was being actively exploited. Microsoft fixed it in its November Patch Tuesday updates on November 10, two days before the outage. But there is no confirmed connection between the flaw and the outage, and the two look like different kinds of problem: the flaw is a client-side Outlook issue that could expose authentication data, while the outage is a cloud connectivity failure in Exchange Online infrastructure. Microsoft has not linked them, and the reporting that mentions both is careful to say the connection is unconfirmed. The same Patch Tuesday also caused a separate headache — the November update KB5132061 failed to install on some Windows 11 machines that had the October preview update installed, with Microsoft confirming the failures and pushing an automatic server-side repair — but again, that is a Windows update problem, not an email outage, and the two share a week, not a confirmed cause. The disciplined way to read all of this is that Microsoft had a bad week across several unrelated systems, and the outage should be judged on its own evidence, which so far shows a failure, not an attack.
What to do if your Outlook is still not working
- Check Microsoft's service health first: the Microsoft 365 admin centre lists the incident as EX1550543 for Exchange Online and MO1550821 for the wider Microsoft 365 effect, and the Microsoft 365 Status account posts updates as the situation changes. If the incident is still open there, the problem is not your computer.
- Try Outlook on the web as well as the desktop app, and try a different network if you can. If the web version works and the desktop app does not, the fault is local to your machine or your connection, not the outage.
- Do not repeatedly re-enter your password or reset your credentials because of an outage. A service failure is not a sign your account was compromised, and a password reset in the middle of one only adds confusion.
- If you are an IT admin, check the service health dashboard, tell your users plainly what is down and what is not, and hold off on your own deployments and configuration changes until Microsoft says the incident is resolved.
- Keep a backup channel for anything urgent — a phone call, a personal email address, or a messaging app — and do not assume a message sent during the outage actually went out. Check your Sent Items and your Outbox once service returns.
- After service comes back, verify your calendar: re-check meetings scheduled during the outage, confirm invites went out, and watch for duplicate messages sent by retries.
- Make sure the November security updates are installed, including the fix for the actively exploited Outlook flaw CVE-2026-124194. The outage and the flaw are separate problems, but both deserve attention in the same week.
What this outage really shows
Step back from the timeline and the incident numbers, and the November 12 outage is a reminder of a risk the industry keeps rediscovering: when email, calendars, meetings, files and AI assistants all live in one vendor's cloud, a failure in one part of that cloud is a failure of the whole working day. Microsoft's own numbers underline the concentration — its service-health documentation reports worldwide Microsoft 365 uptime of 99.526 percent for the first quarter of 2026 and 99.994 percent for the second — which sounds reassuring until you remember that the remaining fractions are measured in hours of downtime for millions of people, and that a single bad afternoon can erase a quarter of good numbers. There is also an irony worth noting: the failing component appears to have been involved in processing server health checks — the system that is supposed to notice when servers are in trouble was itself part of the problem. That is exactly the kind of detail a post-incident review needs to explain, because it determines whether the fix is a patch or an architecture change. For users and businesses, the practical lesson is not to abandon Microsoft 365 — there is no rival that is immune to this kind of failure — but to have a plan for the afternoon it stops. A second communication channel, a local copy of the day's critical documents, and an IT team that can tell staff plainly what is down are worth more than any uptime percentage. Outages like this one are also a test of how a vendor communicates: Microsoft's updates were frequent but vague, and the most useful information — the revised diagnosis that traffic had not surged and servers had simply dropped out — arrived hours after the failure began. The review Microsoft owes its customers is not just about what broke, but about how long it took to say so clearly.
Bottom line
On Thursday, November 12, 2026, a failure inside Microsoft's cloud email service knocked Outlook offline for tens of thousands of users across the Americas and beyond, with more than 130,000 problem reports on Downdetector at the peak and Teams, SharePoint, Copilot and Defender listed as potentially affected. Microsoft says a large number of its own servers became unexpectedly unavailable — not a surge in traffic, and with no evidence of a cyberattack — and it spent the afternoon pausing deployments, moving traffic and restarting servers while service gradually returned. The full root cause has not been published yet, and the separate Outlook security flaw patched two days earlier has no confirmed connection to the outage. If your inbox is still misbehaving, check the Microsoft 365 service health page under incident EX1550543 before touching your own settings, try the web version of Outlook, and confirm anything urgent went out once service is back. The outage is a reminder that cloud email is a single point of failure for most working days — and that a backup way to reach people is worth having before the next one.
Reference sources
- Reuters — Microsoft 365, Outlook down for thousands of US users, Downdetector shows
- HCAMag — Microsoft outage disrupts Outlook and Teams for businesses worldwide
- Tech Prolonged — Outlook outage: Microsoft says servers are unexpectedly unavailable
- Kevin T. Swaney — Microsoft 365 Is Down for Tens of Thousands, and This Week's Security Patch Is Breaking Windows Too
- Ransom News — Microsoft 365 outage: was it a cyberattack? What we know
- WOM Technology — Outlook Outage Nov. 12, 2026: What Microsoft Got Wrong
- Boston MIT — Microsoft 365 Outage Nov 12: What Happened, What to Do
- Ynetnews — Microsoft services back online after global outage
Leave a public comment