🎙 The inbox that would not load | The Receipts - Dataspheres AI

The inbox that would not load (6 min), with captions The story Microsoft 365 users across the Americas could not reach their mailboxes on Thursday, Novembe...

The inbox that would not load (6 min), with captions The story Microsoft 365 users across the Americas could not reach their mailboxes on Thursday, November 12, after a number of Microsoft servers became unexpectedly unavailable. Microsoft's first public post came at 12:54 p.m. Eastern, and by 2:31 p.m. it had named the cause. Nearly 130,000 problem reports for Outlook reached the tracker Downdetector just after 1 p.m. Eastern, along with more than 46,000 for Microsoft 365. Those figures count complaints, not customers. Microsoft reported more than 450 million paid commercial seats in January 2026. In its posts through the afternoon, Microsoft said it was investigating mailbox access in the Americas at 12:54 p.m., extended the scope to any region at 1:07 p.m., cited elevated load at 1:35 p.m., began moving traffic at 1:48 p.m., and paused all deployments and restarted servers at 2:12 p.m. At 2:31 p.m. it said the elevated load was not driven by an increase in traffic but by servers being unexpectedly unavailable, straining the remaining infrastructure. Microsoft has not said why the servers became unavailable. The company's service level agreement sets 99.9 percent uptime a month for Exchange Online, the service that holds mailboxes for organizations. November has 30 days, so that allows about 43 minutes of downtime. From Microsoft's first post to its 2:31 p.m. post is 97 minutes, with the incident still open. Missing the 99.9 percent threshold carries a credit of 25 percent of that month's fee, rising to 50 percent below 99 percent and 100 percent below 95 percent. Downtime is counted per user, in minutes, when mail cannot be sent or received with Outlook on the web, so each customer's number differs. Customers must submit a claim to Microsoft support with the time, duration and number and location of affected users; for a November incident the claim is due by December 31. Thursday was the second long mail outage of 2026 that Microsoft traced to its own infrastructure. On January 22, Outlook and other Microsoft 365 services failed for nearly 10 hours, which Microsoft blamed on a portion of service infrastructure in North America not processing traffic as expected. Microsoft has not linked the two outages. Microsoft had not declared the incident resolved in its 2:31 p.m. post. It promises a preliminary review within 48 hours of resolving a broad incident and a final review within five business days, leaving the reason its servers dropped out still unexplained. Highlights Highlight In pictures Listen The episode as audio