Microsoft’s Azure and Microsoft 365 platforms suffered major outage on Wednesday, only hours before the company’s quarterly earnings call.
The issue hit Azure Front Door, Databricks, Maps, and Virtual Desktop, and Microsoft 365. Xbox and even the company’s investor site were unreachable for much of the day. By evening, Microsoft said latency and error rates were “back to normal,” though the outage lasted more than eight hours.
According to Downdetector, reports started around 11:40 a.m. ET and peaked at over 18,000 before gradually dropping.
The Week Before, It Was Amazon AWS Outage
If this sounds familiar, that’s because it is. Just last week, Amazon Web Services went down, taking thousands of sites and apps—including Snapchat and Reddit—with it. Earlier this year, Microsoft’s own Outlook and Teams platforms were hit by separate disruptions.
The world’s biggest tech companies keep promising resilience, but the same story repeats itself: a single configuration error can ripple across half the internet.
Together, Microsoft, Amazon, and Google now run about two-thirds of the world’s cloud infrastructure, according to Canalys—Amazon at 32%, Microsoft 23%, Google 10%. When any of them stumbles, so does everything that depends on them.
One Error, Global Consequences
The ripple effects reached far beyond Microsoft’s own systems. Alaska Airlines said the outage knocked out parts of its online operations, affecting both Alaska and Hawaiian Airlines. Heathrow Airport and Vodafone reported disruptions before recovering later in the day.
These are not small hiccups. Airlines, hospitals, banks, and governments now all rely on the same few commercial clouds. The promise of redundancy has given way to a different kind of fragility—a dependence on a handful of systems that can all fail in the same way.
The Dangers of Vendor Lock-In
In 2023, U.S. regulators started looking more closely at that dependence. The Federal Trade Commission asked for feedback on the competitive practices of large cloud providers. Microsoft and Amazon said everything was fine. Google accused Microsoft of making it impossible for customers to move to competing platforms.
They’re all right, and all wrong. Cloud providers have built ecosystems that are too expensive and too complex to leave. The result is predictable revenue for them and systemic risk for everyone else.
A single update can take down everything from airline check-in systems to emergency-service networks—and there’s no easy way to switch.
The Bigger Picture
The Azure outage was brief by global standards, but it fits a pattern that’s becoming hard to ignore. The same few companies run the servers, the data centers, and increasingly, the AI infrastructure that other firms depend on.
With each crash, the question comes up again: why do we keep centralizing the backbone of the internet into a few major players?
Until there’s real diversification—or regulation that forces it—these outages are going to feel a lot more routine.
More on IT and Networking News
- RPKI Explained – Resource Public Key Infrastructure
- Telecom at Risk: Addressing Evolving Cyber Threats
- Will Tariffs Impact BEAD Funds & Broadband?
- TLS: The Importance of Transport Layer Security
Discover more from Buy IP Addresses & Sell IPv4 Globally
Subscribe to get the latest posts sent to your email.
