Microsoft's Outage Updates Point to Missing Servers and an Overwhelmed Secure-Connection Process
Microsoft's running updates describe failed health checks, a network change being rolled back and restored servers going unresponsive again, while Australian firms and agencies lost email for hours.
Outspoken Digest Technology Desk
Friday, November 13, 2026/2 min read
Microsoft's own updates describe an outage that changed shape through the night: from elevated load, to missing servers, to failed health checks, to a process that sets up secure connections being overwhelmed, with restored servers failing again once traffic returned. No root cause had been published by Friday morning, so the technical story so far is Microsoft's running commentary rather than a finding.
This desk has covered the outage's scope and status separately. This piece sets out the engineering detail Microsoft has given and how organisations outside the UK coped.
What did Microsoft say went wrong technically?
Microsoft's Exchange Online incident, EX1550543, began with users in the Americas unable to reach mailboxes, SecurityWeek reported. Microsoft said that a large number of servers had become unexpectedly unavailable, pushing work onto the remaining infrastructure, and that overall traffic had not risen.
Just before 3 p.m. Eastern on Thursday, or 20:00 UTC, Microsoft said the affected servers were failing automated health checks, that it was testing a workaround that bypasses the component processing those checks, and that it was looking at high processor use, SecurityWeek said. It also began reverting a recent network configuration change that coincided with the start of the disruption. Microsoft had not said why the health checks failed.
Later updates, quoted by Computerworld, said a process responsible for establishing secure connections became overwhelmed, causing servers to stop processing requests. A companion incident, MO1550821, noted a surge in outbound calls from routing services to backend servers, which Microsoft tied to traffic rebalancing it had done during the event.
Why did recovery stall?
Restarting servers showed signs of improvement at about 9 p.m. Eastern (02:00 UTC), according to SecurityWeek. But Microsoft later said restored servers were becoming unresponsive again, and it responded by restricting some traffic and reintroducing connections gradually, testing a configuration change for the secure-connection behaviour on a subset of servers first, Computerworld reported. Microsoft described its latest fixes as not taking as expected, Computerworld said, and ABC reported that it later spoke of positive results for server stability.
The services hit were wider than email. Computerworld's account of Microsoft's notices lists Teams calendar and scheduling, Copilot, Defender for Office 365 Safe Links, Intune enrolment and parts of Purview, while direct access to SharePoint files and core Teams messaging largely kept working.
How did organisations outside the UK cope?
The picture from ABC News in Australia is of businesses brought to a standstill while waiting for email. Stores lost email-based sales, patients lost telehealth access and some remote schools closed for the day. The broadcaster said Australia's spy agency and federal police were affected, Japan's government said its IT infrastructure was hit and the US House of Representatives said its users were affected. Home Affairs Minister Tony Burke said the government activated a cyber agreement with Microsoft to prioritise Australian interests in restoration, ABC reported.
Greyhound Research chief analyst Sanchit Vir Gogia said, as Computerworld reported, that enterprises have treated Microsoft 365 as a licence rather than a single point of failure, and argued the answer is planned fallbacks, not a second suite: an out-of-band channel for incident command, mail continuity for critical roles and a decision on whether security controls fail open or closed.
The next marker is Microsoft's promised post-incident report, which should explain why the servers dropped out in the first place.
Published in The Outspoken Digest
Editorial desk
Outspoken Digest Technology DeskSoftware, hardware, artificial intelligence and what they change for everyone else.
Newsletter
The Digest, in your inbox
One edition, sent when it is ready. No noise, and your address is never passed on.
Read Next
More Technology →
Starship Flight 15 Still Awaits an FAA Licence Before Tonight's 4:45 p.m. CT Window and First Double Catch Attempt
Nov 13, 2026/2 min read
Nvidia Heads Into Kress's First Week With a Six-Week Slide, a Staff Meeting on 18 November and Earnings Tuesday
Nov 13, 2026/3 min read
EU Digital Fairness Act: Commission Proposes Fines of Up to 6% and a Switch-Off Right for Addictive Features
Nov 12, 2026/2 min read
Agile Robots Raises $500 Million at an $8.5 Billion Valuation as SoftBank and Nvidia Back Its Physical AI Push
Nov 12, 2026/2 min read