On a Thursday afternoon in November, a large share of the world's workplace email and collaboration tools fell silent at once, as Microsoft's Outlook, Teams and Copilot stumbled together under a shared infrastructure failure. Servers had dropped out of service and failed their own health checks, leaving the machines still running to carry a load they were never meant to bear. As the afternoon wore on, the wave of reported trouble began to ebb, though the company had yet to say when normal service would return or what first set the failure in motion.
Microsoft Outage Disrupts Outlook, Teams and Copilot as Tens of Thousands Report Problems
The machines that stayed up were being asked to carry more than they should.
Mark
Start with the basics. What actually broke on Thursday?
Mimi
Microsoft said some of its servers went offline and were failing automatic health checks. That hit Outlook first, but Teams, Copilot and other Microsoft 365 services were affected too.
Mark
And how big was it?
Mimi
Downdetector reports for Outlook reached about 130,000 shortly after 1pm Eastern, then fell to about 77,000 by just before 3pm.
Luke
I'd be careful with those numbers. Downdetector counts user reports, not confirmed affected accounts. They show the shape of the outage, not its exact size.
Mark
Fair. What was Microsoft's explanation?
Mimi
That servers were unavailable, which put pressure on the ones still working. It was testing a configuration change to redistribute traffic and restarting affected services.
Mark
So the surviving servers were overloaded?
Luke
That's the company's description of pressure on remaining infrastructure. It hasn't said what took the first servers offline, so the root cause is still unknown.
Mark
Did Microsoft say when things would be back to normal?
Mimi
No. It gave no timeline for full restoration.
Luke
Which means the drop to 77,000 is encouraging but not proof of recovery. We don't yet know how many people were still affected after that.
Der Puls
- Reports of Outlook trouble surged past 130,000 on Downdetector shortly after 1pm ET, with some users locked out of their inboxes entirely.
- The disruption spilled across Teams, Copilot and other Microsoft 365 services, with no single region spared.
- Failing servers pushed extra traffic onto the machines still standing, deepening the strain on an already fragile system.
- Microsoft is testing a traffic redistribution change and restarting affected services, and by 3pm ET reports had eased to roughly 77,000, though full recovery remains unconfirmed.
On a Thursday afternoon in November, a large share of the world's workplace email and collaboration tools fell silent at once, as Microsoft's Outlook, Teams and Copilot stumbled together under a shared infrastructure failure. Servers had dropped out of service and failed their own health checks, leaving the machines still running to carry a load they were never meant to bear. As the afternoon wore on, the wave of reported trouble began to ebb, though the company had yet to say when normal service would return or what first set the failure in motion.
Shortly after 1pm Eastern on Thursday, the outage tracker Downdetector began filling with complaints. About 130,000 people said Outlook was not working. Some could not send or receive email; others could not open their inboxes at all.
The trouble did not stop at email. Users also struggled to reach Teams, the workplace messaging app, and Copilot, Microsoft's AI assistant, along with other Microsoft 365 services. The company acknowledged the problems could touch users in any region, so there was no single place where the disruption could be contained.
Microsoft explained on X that some of its servers had become unavailable and were failing automatic health checks. That left the surviving servers absorbing more demand than they could comfortably handle. The company said it was testing a configuration change to redistribute traffic and restarting the affected services, saying, "We're working to safely return them to service."
By just before 3pm Eastern, reports of Outlook problems had dropped to about 77,000, a meaningful easing though still a large number of people. Microsoft has not said when full service will be restored, nor why the servers went offline. Downdetector figures are user-submitted, so they signal scale and direction rather than an exact count. What remains to be seen is whether Microsoft confirms full recovery and explains the root cause.
Shortly after 1pm Eastern Time on Thursday, the number of people telling the outage tracker Downdetector that Outlook wasn't working climbed to about 130,000. Some said their emails weren't going out or coming in. Others couldn't get their inboxes to open at all.
The trouble reached well beyond email. Users also reported problems reaching Teams, Microsoft's workplace messaging app, and Copilot, its AI assistant, along with other Microsoft 365 services. The company acknowledged that the issues could affect users in any region, so there was no single place where the disruption was contained.
Microsoft addressed the problem publicly on X. It said some of its servers had become unavailable and were failing automatic health checks, which in turn put added pressure on the servers still running. In other words, the machines that stayed up were being asked to carry more than they should.
The company described two steps it was taking. It was testing a configuration change meant to redistribute traffic, which had become elevated on parts of its infrastructure. It was also restarting the affected services to bring them back online. "We're working to safely return them to service," Microsoft said.
The figures on Downdetector suggest the situation eased somewhat as the afternoon went on. By just before 3pm Eastern (20:00 GMT), reports of Outlook problems had fallen from the earlier peak to about 77,000. That is a substantial drop, though it still represents a large number of people reporting trouble.
What Microsoft did not say is when services would be fully restored. The company also did not offer an explanation of why the servers went offline in the first place, beyond saying they had unexpectedly become unavailable.
Downdetector counts are user-submitted reports rather than a precise tally of affected accounts, so the 130,000 and 77,000 figures indicate scale and direction more than an exact headcount. What to watch next is whether Microsoft announces full recovery and whether it later explains what caused the servers to fail.
Bemerkenswerte Zitate
We're working to safely return them to service.— Microsoft, on X
Möchten Sie die ganze Geschichte?
Das Original lesen bei Al Jazeera ↗