In the early hours of October 20, 2025, Amazon Web Services (AWS) suffered a major outage that disrupted websites and digital services worldwide, affecting entertainment, banking, travel, and messaging platforms.

The disruption began in AWS’s US-EAST-1 region, home to many of its core systems.

An issue in the company’s Domain Name System (DNS) and DynamoDB database service interrupted key connections between applications and their data.

Although Amazon said it “fully mitigated” the problem within a few hours, the ongoing outage reveals how much of the internet relies on a single provider and how fragile that setup can be.

Why It Matters: What began as a disruption in one AWS region has cascaded into a global event still unfolding in real time. Critical systems remain unstable, and many platforms are struggling to restore functionality. As AWS races to stabilize its core services, millions of users and organizations are being reminded that the convenience of the cloud comes with a dangerous concentration of risk.

• DNS and DynamoDB Malfunctions: AWS originally traced the disruption to an issue with its DNS service and DynamoDB, a database system widely used to store and manage information. DNS, which converts human-readable URLs into machine-recognizable IP addresses, stopped working. Mike Chapple of the University of Notre Dame compared the effect to “large portions of the internet suffered temporary amnesia.”

• Services Large and Small Affected: The initial fallout was widespread. Snapchat, Reddit, Canva, Facebook, and Fortnite went dark, along with Coinbase, Perplexity, Amazon’s own site, Prime Video, and Alexa. Messaging apps such as Signal were disrupted, and major UK banks, including Lloyds and Halifax, reported outages. Even AWS’s own support ticketing system went offline, adding to customer frustration.

• Continued Disruption and Unstable Recovery: Despite AWS’s early statement claiming the issue was “fully mitigated,” widespread instability persists. Core services like EC2, CloudTrail, and Lambda remain degraded, with users reporting slow responses, throttled requests, and connection errors. Many organizations continue to experience partial outages and performance failures hours after AWS’s initial announcement. Far from over, the recovery process has proven inconsistent and incomplete.

• Centralization and Fragility: The outage has reignited debate about the centralized nature of cloud infrastructure. AWS’s US-EAST-1 region proved again how disruptions can have global consequences. Lance Ulanoff of TechRadar highlighted how AWS has become the invisible backbone of the web, saying it “sits in the middle of everything.”

• Not Just About Uptime: Beyond commerce, the outage disrupted how people communicated and accessed essential information. Messaging app Signal went offline, prompting digital rights group Article 19 to call the event a “democratic failure.” Media outlets such as The New York Times, Disney, and the Associated Press were also affected. Some activated backup systems to keep breaking news flowing.

Go Deeper -> Amazon Web Services says issue ‘mitigated’ as sites hit by global internet outage – CNN

Web services begin to recover after major outage takes leading websites offline – NBC News