Amazon sent millions of Americans offline when its Amazon Web Services (AWS) arm suffered one of its longest outages in recent memory.
The chaos started shortly after midnight on the East Coast, when a minor update to the Domain Name System broke Amazon’s DynamoDB service. The database is one of the most widely used on the planet, quietly running everything from Alexa’s voice recognition to the logistics systems that move parcels around.
Almost immediately, things began to unravel. Amazon’s delivery systems failed at 2 am, followed by Slack, Venmo, Zoom, and airlines, which saw more than 4,000 flights delayed. By 3 am, the mess had spread across the US, hobbling financial platforms like Coinbase and Robinhood, and leaving ordinary users unable to trade, study, or even print shipping labels.
One bewildered small-business owner in Maryland told the Wall Street Journal: “We’re not getting our orders today and we’re not getting anything out. I had no idea that the loss of one web cloud service would chip away at my small business and give me a Monday morning from hell.”
By the end of the day, 142 AWS services had gone dark. Amazon admitted that customers were struggling to start new cloud servers and move workloads between regions. With DNS misdirecting traffic, data centres on the East Coast were essentially unreachable.
Ponce Bank boss Carlos Naudon said the outage stopped him charging his electric car and cost the bank between $50,000 and $100,000 in lost transactions.
“Honestly, it’s more unsettling than anything else. You can’t assume that something like that is going to be foolproof and nothing can happen.”
Analyst Jacob Bourne from eMarketer warned: “Even if just briefly, major providers like AWS going down represent vulnerabilities in what have become critical infrastructure for organisations and, in some cases, governments globally.”
By late afternoon, Amazon said it had mostly restored service, but the damage was done. The outage showed how dependent the world has become on a handful of tech giants running the so-called “public cloud.”
AWS controls roughly a third of the cloud market, making it the digital backbone for everything from trading platforms to takeaway apps. When it sneezes, the internet catches a cold.
Some firms that spread their workloads across multiple clouds managed to dodge the worst of it. McKenney’s CIO Shaun Hunt said his company avoided full downtime thanks to diversification. “We have a strategy to diversify our risk,” he said.
Still, the event echoed other high-profile internet collapses caused by DNS errors, such as Facebook’s 2021 blackout and the massive 2016 outage that broke half the web. It was yet another reminder that the internet, for all its complexity, can still be felled by a single bad line of code from the wrong corporate giant.
