Outlook down on November 12: failed health checks overloaded Microsoft servers

Microsoft's Outlook email service went down for many users on Thursday, 12 November 2026, as a Microsoft 365 outage cut connections to Exchange Online mailboxes, the cloud email system that sits behind Outlook for businesses. Microsoft first reported trouble reaching mailboxes in the Americas and later said users in any region could be affected. It said servers had been taken out of service after failing automated health checks, which overloaded the remaining capacity. By the afternoon in the United States, Microsoft had restarted the affected servers and its service-health page showed products as working again; reports said more than 130,000 users had been hit.

Most companies no longer run their own email servers. They rent email, calendars and chat from cloud providers such as Microsoft, which runs them in large shared data centres. That move saves money and effort, but it also concentrates risk: when one part of a provider's system fails, thousands of organisations lose email at the same moment, and none of them can fix it themselves.

This outage shows how that can happen without any attack or traffic surge. Cloud systems constantly check whether each server is healthy and automatically remove the ones that fail, so that users are moved to working machines. According to Microsoft, that safety system was part of the problem this time: servers failed the health checks and were taken out of service, and the work they had been doing landed on fewer machines, which then struggled. Microsoft said the failed checks might come from a specific component that processes server health, rather than from the servers themselves.

For business technology teams, the episode is a reminder of three practical points that come up after every such outage: keeping a second way to reach staff and customers when email is down, knowing where the provider posts official status updates (Microsoft uses its service-health page and the Microsoft 365 admin centre), and checking what the service contract says about compensation for downtime.

Microsoft's account changed as the day went on. It first linked the trouble to elevated load in part of its infrastructure, then clarified that the load came from servers being unexpectedly unavailable, not from more users. Other explanations circulated online, including a claim about faulty spam filtering, but Microsoft had not confirmed any of them, and readers should treat such claims as unverified.

Peepals timeline: On Thursday, 12 November 2026, Microsoft opened incident EX1550543 after users in the Americas could not reach Exchange Online mailboxes through several connection methods. It then widened its warning to say Microsoft 365 connections in any region could be affected, under incident MO1550821. As the day went on it explained that servers had failed automated health checks and been removed from service, and it began rolling back a network configuration change that had coincided with the start of the trouble. By about 4:45 pm US Eastern time the affected servers had been restarted and the outage appeared to be winding down, and the service-health page later showed products as operational.

Short-term: Businesses, schools and government offices that rely on Outlook lost access to email for part of the day, and some had to wait for mail clients to reconnect gradually after servers came back.

Long-term: Repeated cloud outages strengthen the case for companies to plan for provider failures, through backup communication channels and clearer service agreements, rather than treat cloud services as always available.

Who is affected: Organisations that use Microsoft 365 and Exchange Online for email; IT and support teams who handle user complaints during outages; and customers of businesses whose communication depends on that email.

Questions and Answers