The Daily Ledger

TechnologyNews

Nadella: Microsoft outage was a bug, and old systems can't be patched

After a week of failures at Microsoft, the FAA, Apple and the NSA, Microsoft's chief blamed a hidden bug and warned about unfixable connected hardware.

The Daily Ledger Staff · Updated 3 min read
Flat-vector illustration of a connected city and cloud data centre with operational and ageing server racks, a software bug workstation, and a subtle amber network fault warning.
Illustration generated with AI

A string of tech outages in about a week has hit Microsoft, the US air traffic system, JPMorgan Chase, United Airlines, Apple and the National Security Agency. Organisations that have explained their own failures point to ordinary faults and maintenance, not attacks. Microsoft's chief executive, Satya Nadella, addressed the Microsoft outage on Wednesday.

What each organisation has said

According to Scientific American, Microsoft's cloud software was unavailable for almost 24 hours. The company says the problems were not caused by malicious activity or AI agents. A preliminary investigation found that a software bug introduced in July went unnoticed until a routine change pushed more traffic onto the affected servers. Recovery took most of a day because restarting servers kept triggering the same fault.

The Federal Aviation Administration reported on November 13 that data flooded its telecommunications network during work that included installing a new router at a remote site in Arkansas. The disruption affected 17 of its 22 regional air traffic control centres and led to a nationwide ground stop. The agency found no sign of a cyberattack. Apple's outage followed an internal update, the report says. The NSA said its trouble on a classified communications network arose during scheduled maintenance.

Nadella on a "perfect storm"

Speaking at The New York Times' DealBook Summit, Nadella blamed the Outlook disruption on a "latent bug" that produced a "perfect storm," TechCrunch reports. He said Microsoft had run an incident response and would apply lessons to its upstream architecture and operations. He also called lots of critical infrastructure "unpatchable," naming firmware, chillers and HVAC equipment that are connected to the internet.

AI tools are now a very powerful way to find vulnerabilities, he said, and because flaws are being exploited, a patch has to move from availability to production almost instantly. That speed, he said, brings instability.

Are outages really rising?

Experts quoted by Scientific American urge caution. An Uptime Institute analysis found that data-centre outage rates per site fell in 2026 for a fifth straight year. Laura Nolan, a software reliability engineer, used a hypothetical figure of 52 major outages a year to explain that it would not mean one a week, because incidents cluster, and that clustering does not necessarily point to a shared cause. Kelly Shortridge, chief product officer at Fastly, offered a hypothesis, which she did not present as proof: the yearly rush to ship software before holiday update freezes may play a part. Nancy Leveson, an MIT professor who studies system safety, said changes to complex systems often have results no one intended.

Shortridge's advice is to isolate components so one failure does not bring down the rest, and to test backup systems under stress.

Sources

Written with automated tools from the sources above and published without editor review. Spotted an error? Tell us · Corrections Policy

Report a problem with this story · Copyright & takedown