A glitch at an Amazon Web Services (AWS) data centre in the US left millions around the world unable to access half the internet, but how exactly did it cause the widespread disruption?
AWS recorded"increased error rates" yesterday evening that continued for more than 15 hours, affecting some of the most popular social media platforms, websites and online services.
It has been described as the largest disruption since theCrowdStrike outage of 2024.
As services return to normal, here is everything we know about what caused yesterday's mammoth shutdown.
What is Amazon Web Services?
Let's start at basics.
AWS is a subsidiary of billionaire Jeff Bezos' Amazon and the world's most broadly adopted cloud service — a way of storing files and applications over the internet rather than on physical devices.
It provides power, storage and databases to companies, universities and governments.
"We offer the greatest choice of innovative cloud capabilities and expertise, on the most extensive global infrastructure with industry-leading security, reliability, and performance," AWS says on its website.
AWS controls more than 41 per cent of the global market, with Microsoft and Google as the next biggest competitiors.
With some of the world's largest companies relying on the service, it's no surprise that a glitch would cripple half the internet.
What is the current status of the AWS outage?
Amazon said it first began seeing errors to AWS services about 5.50pm AEDT yesterday.
Three hours later, the underlying issue was located and crews saw"significant signs of recovery".
Amazon later said it was still experiencing"significant API errors and connectivity issues across multiple services" and began investigating the "root cause".
About 9am today, all AWS services returned to normal and the operation issue has been marked as resolved.
What caused the AWS outage?
AWS' DynamoDB is a database that hosts information for companies, including customer data.
This was failing to work because of an error to the domain name system (DNS) at its data centre in northern Virginia — one of the most important cloud hubs in the world.
DNS is the service that translates internet addresses into machine-readable IP addresses that connects browsers and apps with websites and online services.
Essentially, it acts as a phonebook, connecting devices to servers.
What websites and apps were down because of the outage?
DownDetector, a website that tracks online outages, recorded more than 11 million reports to hundreds of services.
Social media and chat platforms Snapchat, Reddit, Signal and Whatsapp were affected.
Users reported outages to Disney+, 9Now, Hinge, Tinder, Coinbase, Robinhood, Life360, Tidal, EA Games, Roblox, Fortnite, Venmo, Xero Canva, Slack, Zoom, and Duolingo.
Starbucks, DoorDash and Lyft confirmed briefly seeing issues.
Amazon's own website, PrimeVideo and Alexa also faced problems.
In the United Kingdom, Lloyds Bank, the Bank of Scotland, Vodafone and BT were impacted.
Yesterday's outage is a prime example of the warnings experts have been raising for years about the potentially dire consequences of allowing a handful of companies to dominate major internet operations.
Has this happened in the past?
The last time there was a disruption of this magnitude was in 2024.
A faulty software update bycybersecurity company CrowdStrike impacted devices running Microsoft's Windows, which caused issues to airports, hospitals, banks and other services across the world.
This is also not the first time there has been an outage related to Amazon’s key services.
Many popular internet services and publishers were down after a brief outage in 2023.
AWS’s longest outage in recent history affected companies, from airline reservations and auto dealerships to payment apps and video streaming services, for more than five hours in late 2021.
Other major outages happened in 2017 and 2020.
