Sector: Technology
Microsoft 365 Outage Enters a Second Day, With Email Still Not Back to Normal
When the world's office email fails for a second day, a stock exchange tells its listed companies to pick up the telephone.
A global Microsoft 365 outage entered a second day on Friday, with Outlook email still disrupted and Microsoft saying its capacity had not yet returned to normal.
Microsoft's latest update, at 6:11 a.m. ET on Friday, reported additional progress toward recovery.
Traffic was being gradually restored across portions of the affected infrastructure, the company said. It pointed to improvements in mailbox access and delayed email processing, while warning that some intermittent issues may continue.
An hour earlier, at 5:28 a.m., Microsoft had said recovery actions were progressing in phases across the affected infrastructure. Its capacity had not yet fully returned to baseline, it said.
To protect delayed mail, the company extended the email timeout window to 48 hours. Delayed messages will keep being retried while the queue and backlog are processed.
Servers dropped out, and the rest strained
The outage began on Thursday afternoon, when users started losing access to Exchange Online, the cloud email system behind Outlook.
Microsoft acknowledged the problem at 12:54 p.m. ET, saying it was investigating issues accessing Exchange Online mailboxes in the Americas. The disruption then spread to users in every region.
Microsoft later said the elevated load was not driven by an increase in traffic. A number of servers had become unexpectedly unavailable, it said, straining the remaining infrastructure.
The servers had dropped out after failing automated health checks, the routine tests that pull unresponsive machines out of service. Microsoft said the failures may originate from a specific component responsible for processing server health.
A core process on the affected servers was also becoming overwhelmed, causing a further loss of capacity. Microsoft said it had not identified any recent change correlated with the start of the problem.
At the Thursday peak, outage tracker Downdetector logged more than 130,000 reports of Outlook problems, and more than 46,000 reports for Microsoft 365. Reports came from 74 countries and territories, and around 15,000 Outlook reports were still being logged on Friday at about 6 a.m. ET.
The fixes did not hold on the first attempt
By Thursday night, Microsoft conceded that the changes it had applied to improve serviceability were not taking as expected. It said it was identifying alternate workstreams that could be applied to increase capacity.
Early on Friday, the company reported early stabilization and capacity improvements as recovery actions continued. The updates that followed stayed cautious, describing progress in phases rather than a finish line.
Microsoft is tracking the incident as EX1550543 for Exchange Online and MO1550821 for the wider Microsoft 365 disruption. It said it would treat the incident with urgency and priority until full resolution.
Microsoft said it had no indication that the outage was the result of malicious cyber activity.
An exchange told its listed companies to use the telephone
Euronext, which runs exchanges in Paris, Brussels, Lisbon, Amsterdam, Oslo, Athens and Milan, advised listed companies on Friday morning to use the telephone to communicate corporate information.
Its corporate actions and listing teams are currently experiencing an email communication disruption and are unable to receive emails, the company said in a corporate filing. For urgent matters such as suspensions, issuers were told to contact the relevant exchange operator by telephone.
In Taiwan, fund management operations at securities firms, life insurers and investment trust companies were disrupted.
Governments were hit across three continents. Australia's intelligence agency and federal police were among those affected, Japan's government information systems were impacted, and Singapore's government emails were down.
In the United States, the House of Representatives said the outage had affected its users and its engineers were working with Microsoft on the problem. Britain's National Health Service said some of its services were affected.
OpenAI and Anthropic each reported problems connecting with Microsoft 365 tools, a reminder that other companies' AI assistants ride on the same email infrastructure. Workhorse Group postponed its third-quarter results announcement from November 12 to November 16, citing business disruptions caused by the outage.
The airlines mostly kept flying. Air India said its flight operations were unaffected and that it had temporarily rerouted its email services, while IndiGo reported disruptions to its email support services and said its flights were unaffected.
Shares fell before the open; Friday's close is the verdict
Microsoft shares closed at $523.00 on Thursday, up 0.36 percent on the day, before the outage entered its second day.
Dow Jones Newswires reported at 5:31 a.m. ET that the stock had declined 2.7 percent in premarket trading, to trade around $509 a share. That put the company on track to lose over $103 billion in market value.
Premarket trading is an early indication, not a settled verdict. The first close that counts comes when Friday's regular session ends at 4 p.m. ET.
"It's a reminder that concentration of critical cloud services in a few places creates systemic vulnerabilities," said Mikhail Zverev, a fund manager at Amati Global Investors. He expected the impact to be short-lived, with a fix likely to come soon.
The timing sharpens the pressure. Anthropic launched a new suite of office tools on Tuesday, and the market saw it as a direct challenge to Microsoft, Barron's noted.
More than 430 million people use Microsoft 365 services, alongside governments and organizations around the world.
What remains unknown
The root cause is unknown. Microsoft has described how servers dropped out of service, but not why the health checks failed or why a core process on the affected servers became overwhelmed.
The full scale is unknown, too. Microsoft has not announced how many users were affected, and the Downdetector figures are reports filed by users, not a count of everyone who lost email or meetings.
Whether an update caused the failure is unknown. Australia's home affairs minister, Tony Burke, told reporters Microsoft had compared the problem to the 2024 CrowdStrike outage, where an update went awry; Microsoft itself has not confirmed any update as the cause.
What customers are owed is also unknown. Microsoft has not announced service credits, compensation or a date for a full public account of the failure.
Three checks will show when this is really over
The first check is an all-clear from Microsoft itself. An update saying capacity has returned to baseline, with the delayed-mail backlog cleared, is what ends the incident — not another progress report.
The second check is the root-cause account. A post-incident review that explains the failed health checks would separate a one-off failure from a weakness in how Microsoft runs its email infrastructure.
The third check is Friday's settled close. If the shares recover in regular trading, investors will have treated the outage as a bad day; if the premarket fall holds into the close, the market is pricing a longer problem.
Sources & evidence
Source checked7 sources · 11 figures
Figure receipts
Sources
Sources used for this piece.
Microsoft share price, settled close, Thursday November 12, 2026
Dow Jones Newswires
Microsoft On Track to Lose $100 Billion in Value as Widespread Outage Continues
Associated Press
Microsoft outage hits companies and governments worldwide as it struggles to restore services
USA Today
Outlook still down for many as Microsoft outage lingers into Friday
people who use Microsoft 365 services (more than this figure)
email timeout window Microsoft extended during the outage to protect delayed mail
Reuters
Euronext asks listed companies to phone after Outlook outage
All dates 2026.
Microsoft Corporation
Stock · Nasdaq · MSFT
Discuss this story. Join TickerGrove on Discord to talk companies, earnings, and markets, or request future coverage.

