A Single AWS Failure Knocked a Huge Chunk of the Internet Offline

A technical failure in Amazon's critical Northern Virginia data center region triggered a massive AWS outage on Monday, knocking major services like Snapchat and Disney+ offline for hours and exposing the fragility of the web's core infrastructure.

• A cascade of errors: The disruption began in the pre-dawn hours with a DNS resolution issue for Amazon's DynamoDB database service. The initial problem created a domino effect, impairing network load balancers and the EC2 service responsible for launching cloud servers. While Amazon announced an initial fix within hours, the subsequent failures meant recovery was a day-long effort, with some backlogs persisting even after services returned to normal.

• Domino effect: The outage demonstrated the internet's deep reliance on AWS, taking down a wide swath of services. Gaming platforms like Fortnite and Roblox went dark, while streaming services went down, as reported by Cord Cutters News. The failure even hit Amazon's own internal warehouse and seller systems, according to CNBC.

• Déjà vu in Virginia: Monday's failure is the latest in a series of major outages originating from the US-EAST-1 region. The Northern Virginia data center, one of Amazon's oldest and largest, was also the source of similar widespread disruptions in 2021 and 2023, raising questions about the systemic risk of having a single, critical point of failure.

The outage serves as another stark reminder of how much of the digital world is built on infrastructure controlled by just a handful of companies.

The incident has reignited the debate over internet centralization and the risks of relying on a few tech giants. Some are speculating that an ongoing "brain drain" at Amazon could be a factor in these repeated failures, while others are now focused on the financial fallout as companies begin to calculate their losses.