Microsoft Outage Ends After 36+ Hours: Outlook, Teams Restor

Microsoft has confirmed that every service hit by its sprawling global outage is working again, closing out one of the longest disruptions to its cloud productivity tools in recent memory.

Microsoft said on Saturday that all services knocked out by a global outage that began on Thursday, including Outlook and Teams, have been restored. By the company’s own timeline, the disruption ran for well over 36 hours. For many businesses, schools and government offices, it was a stretch in which email went dark, meetings failed to start and everyday work slowed to a crawl.

How the outage unfolded

The first signs of trouble appeared on Thursday, November 12, around midday to early afternoon US Eastern time. In India, that was late Thursday night, around 11:30 pm. Microsoft’s service-status account acknowledged that it was investigating reports of users having trouble reaching online mailboxes. Within hours, complaints were pouring in from around the world.

Outage trackers showed reports climbing steadily, with tens of thousands of users flagging problems at the peak. Outlook was the most visible casualty, but it was far from the only one. Microsoft listed a broad set of services as potentially affected:

Exchange Online, which powers business email
Microsoft Teams
SharePoint Online
Microsoft 365 Copilot, the company’s AI assistant
Microsoft Defender XDR and Defender for Office 365
Microsoft Planner and Microsoft To Do
Microsoft Intune, the device-management service

As the first day dragged on, the story became one of partial recoveries and setbacks. Some users regained access while others stayed locked out. At times, one service appeared to improve just as another deteriorated. By late Thursday night, Microsoft said it was seeing “partial recovery across regions” and that its traffic optimisation work was producing positive results for stability and capacity.

A long second day

By Friday morning, the disruption had turned into a second day of work. Microsoft reported “encouraging recovery progress,” with availability and email delivery improving across a growing share of the affected infrastructure. It cautioned that some users might continue to see intermittent problems as recovery actions continued.

Recovery was deliberately slow. The company said that, given the scale of demand, it restored services gradually and validated stability at each stage. That approach is standard in large-scale recoveries, because bringing everything back at once can overwhelm systems that have just stabilised. As capacity returned, users were told they might see a gradual recovery as their clients reconnected.

A second problem emerged in the form of a backlog. Millions of emails had been stuck in queues, and Microsoft extended the timeout window for email delivery to 48 hours so that messages would not be bounced back to senders. By Friday, the company said these queues were significantly reduced and approaching normal levels, and many people reported receiving a wave of delayed messages all at once.

What Microsoft says caused it, and what it says did not

Microsoft has been careful about what it has said regarding the cause. It described the incident as a technical issue and stated explicitly that it was not related to cyber activity or to AI agents. The statement matters because large outages immediately raise suspicion of a hack, and because Microsoft has been pushing AI agents deeply into its products. The company also apologised to customers for the disruption.

What the company has not provided, at least so far, is a detailed technical explanation of the root cause. Microsoft typically publishes a formal post-incident review for major outages, usually within days, laying out the sequence of events, the underlying fault and the steps it will take to prevent a repeat. Until that report appears, the precise trigger remains unconfirmed, and any claims beyond the company’s own statements are speculation.

Who felt it the most

Few software tools are as embedded in daily life as Outlook and Teams. When they stop, the effects spread quickly:

Offices and enterprises. Companies that run their communication entirely on Microsoft 365 lost email and the ability to hold virtual meetings at the same time. Teams that depend on shared files and calendars found those unreachable.

IT administrators. Intune and Defender sit behind the scenes of device management and security. Disruption to these tools makes the job of IT teams harder, and a prolonged problem raises uncomfortable questions about visibility during an incident.

Customer-facing work. Support desks, sales teams and operations staff who rely on email to communicate with clients faced delays and missed messages.

The online public. Social media filled with jokes and memes as the outage passed the 12-hour mark, a reminder of how normalised these tools have become and how disruptive their absence is.

The bigger issue: concentration risk

Beyond the immediate frustration, the episode renews a debate that follows every major cloud failure. When so many organisations rely on a single provider for email, collaboration, identity, security and now AI, one fault can ripple across industries at once. The incident is not the first this year, with an earlier January disruption also leaving users unable to log in to Microsoft 365 and Teams, though that one was brief by comparison.

Resilience experts regularly advise organisations to keep a plan for when a core platform goes down. Practical steps include:

Maintaining an alternative communication channel for emergencies, such as a separate messaging or conferencing tool.
Keeping critical contact lists and incident runbooks outside the primary platform.
Testing what happens to mail flow and calendars during a prolonged outage.
Reviewing service-level agreements and understanding what compensation, if any, is available.

What users should do now?

For individuals and IT teams returning to work, a few checks are worthwhile:

Look for delayed emails and confirm that important messages sent during the outage were delivered.
Re-check meeting invites and calendar entries that may have failed to sync.
Sign out and back in if apps behave oddly, as clients reconnect at different speeds.
Watch the official service-health dashboard and status account for any follow-up notices.

What comes next

With services restored, attention turns to the full post-incident review. The key questions are what failed, why recovery took so long, why different services seemed to fail and recover unevenly, and what changes will stop a repeat. Customers, particularly large enterprises with contractual protections, will be looking for clear answers and, in some cases, service credits.

For the millions who spent a day and a half without their inbox, the immediate message is simple: the tools are back. The longer lesson is about how much of modern work sits on a handful of cloud platforms, and how quickly it can stall when one of them stumbles.

Frequently asked questions

When did the Microsoft outage start?
It began on Thursday, November 12, 2026, around midday to early afternoon US Eastern time, which was late that night in India.

Which Microsoft services were affected?
Outlook and Exchange Online, Teams, SharePoint Online, Microsoft 365 Copilot, Defender, Planner, To Do, and Intune, among others.

Was the Microsoft outage caused by a cyberattack?
Microsoft says no. It described the incident as a technical issue unrelated to cyber activity or AI agents.

Are all Microsoft services working again?
Microsoft says all affected services have been restored. Some users may still see lingering delays as clients reconnect.

Will I lose emails sent during the outage?
Microsoft extended its delivery timeout window to 48 hours to help prevent messages from bouncing, and queues were reported to be returning to normal.