Amazon Web Services Outage Disrupts Global Websites, Services

A major outage at Amazon Web Services (AWS), stemming from a domain name system (DNS) issue in its US-EAST-1 region, disrupted a vast array of global websites and services through Monday, with recovery efforts concluding by late evening.

The disruption, which began around 2311 UTC Sunday (3:11 a.m. Eastern Time, Monday), affected critical platforms including Snapchat, Roblox, Fortnite, Signal, Coinbase, Robinhood, Venmo, McDonald’s app, United Airlines, and the British government’s HM Revenue and Customs website, as well as Amazon’s own services like Ring and Alexa.

The root cause was identified as an underlying internal subsystem responsible for monitoring the health of network load balancers, which impacted core services like DynamoDB and led to widespread API errors and connectivity issues. The outage originated in AWS’s key data center in northern Virginia, a hub that supports much of the world’s cloud infrastructur.

Over 11 million user reports were logged on DownDetector, with more than 2,500 services affected globally. The disruption had significant ripple effects, including students being unable to access learning platforms like Canvas, which is used by 50% of college and university students in North America.

Financial and government services were impacted, with Medicare enrollment websites inaccessible and United Healthcare’s provider search tool malfunctioning. The outage also affected media organizations, including The New York Times, The Wall Street Journal, and Disney, which reported issues with their online content and services.

AWS confirmed the DNS issue was fully mitigated by midday, and most services were operating normally by 1900 UTC (6 p.m. Eastern Time), though some throttling and minor delays persisted into the evening.

This is not the first major AWS outage; similar widespread disruptions occurred in 2021 (lasting over five hours), 2023, 2020, and 2017, highlighting recurring vulnerabilities in the centralized cloud infrastructure.

Experts like Patrick Burgess of BCS, The Chartered Institute for IT, emphasized that while such outages are typically resolved within hours, they underscore the fragility of a global digital ecosystem dependent on a few dominant cloud providers.