On Monday, October 20, 2025, Amazon Web Services suffered a large-scale outage that caused millions of users to briefly go offline. Starting around 9 a.m. CET, the outage brought down a number of high-profile applications and websites, such as Snapchat, Roblox, Fortnite, Duolingo, and a few others, leaving many individuals unable to use their accounts until it came back online.
According to Downdetector, users faced issues with Snapchat login, Snapchat web, and reported “is Snapchat down” queries surged as the AWS outage spread. The problem originated from “significant error rates” in AWS’s DynamoDB service in the US-EAST-1 Region (northern Virginia), considered the global capital for data centers.
AWS status updates confirmed recovery was underway, with Amazon stating: “We continue to observe recovery across most of the affected AWS Services.” The downturn disrupted a number of high-profile services, such as the cryptocurrency exchange Coinbase, Epic Games, major UK banks, and the check-in systems of many airports.
Amazon identified a specific root cause of the problem and had its engineers working on several parallel lines to accelerate the recovery. This incident highlights the danger of having an individual, monopoly cloud provider. Experts in the industry recommend that an organization should distribute its workloads among various cloud platforms as a means of insurance in the event of future, systemic disruptions.
