AWS outage exposes global internet fragility
Amazon Web Services experienced a major outage that disrupted a large share of the internet for hours, affecting consumer apps, enterprise tools, media sites, financial services, airlines, and smart home devices before services were restored. The disruption centered on AWS’s oldest and largest region in Northern Virginia. A failure in resolving addresses to a core database service prevented applications from reaching their data, which cascaded across interdependent services and customers. The scale of disruption highlighted the concentration of critical internet workloads on a small number of cloud platforms and regions.
The incident was tied to a Domain Name System failure affecting access to the DynamoDB API in the US-East-1 region. DNS is the directory that translates service names into IP addresses. When resolution to the database endpoint failed, applications could not locate back-end resources even if those resources were healthy. That mismatch produced surges in timeouts and error rates across systems that rely on DynamoDB and related control planes. Because many customers centralize production workloads in this region and depend on shared primitives, the fault propagated until mitigations and cache clearing restored normal address resolution.
A wide range of services reported problems as the outage rippled outward. Consumer and social platforms affected included Snapchat, Reddit, Pinterest, Facebook, Instagram, and Apple TV. Messaging and collaboration tools saw interruptions, including WhatsApp, Signal, Zoom, Slack, Microsoft Teams, and Chime. Gaming platforms and services were hit, including Fortnite, Roblox, Xbox services, Playstation, and titles such as Rainbow Six Siege and PUBG Battlegrounds. Streaming and media platforms encountered issues, including Netflix, Hulu, Roku, Spotify, Disney outlets, ESPN, and major publishers and newsrooms such as the Associated Press, The New York Times, and The Wall Street Journal. Commerce and delivery platforms experienced disruption, including Amazon’s retail site, DoorDash, Grubhub, and Etsy. Smart home and device ecosystems such as Ring doorbells and Alexa assistants were impacted, with some users reporting alarms and automations failing to trigger. Financial services and fintech were affected, including Venmo, Coinbase, Robinhood, banks and payment portals, and U.K. banks such as Lloyds and Halifax. Airlines including Delta and United reported operational issues, and government portals such as HMRC experienced access problems. Additional apps and platforms cited by outage trackers included Perplexity, Duolingo, Canva, Goodreads, Life360, CollegeBoard, Wordle, T-Mobile, Verizon, and Starbucks.
AWS indicated services returned to normal later the same day, with the caveat that some services would process backlogs for several hours as queues drained and networking stabilized. This pattern aligned with a DNS-related incident, where restoration of correct resolution enables gradual normalization as caches refresh and dependent services recover. While core functionality resumed, the event exposed the operational risk of heavy dependence on a single region and shared control planes for critical services.
