A massive failure in Amazon Web Services (AWS) this Monday caused a significant portion of the Internet to stop working, affecting everything from banking portals to public service platforms. Although the company has already identified the origin of the failure, it confirmed that work to fully normalize its systems continues.
According to an official statement from Amazon, the interruption was due to issues with the Domain Name System (DNS), the mechanism responsible for translating the web addresses users type into numerical identifiers that computers can process.
The company detailed that high error rates were recorded in several services, specifically linked to the DNS resolution of the DynamoDB APIs in its main data center in Northern Virginia. Although the technical DNS defect was resolved around 2:24 a.m. (Pacific Time), the process of fully restoring all services is not yet complete.
Those affected by the outage included the main Amazon.com site, various subsidiaries, and AWS customer support channels. The company recommended its users consult the AWS Health Dashboard for real-time updates.
In the early hours of Monday, Amazon reported that the incident had been “fully contained” and that most services were resuming their normal operation, after several hours of widespread instability online.
The outage, which began around 3:00 a.m. on the US East Coast, was attributed by the tech giant to DNS failures, a type of problem that, depending on its complexity, can take hours to resolve.
During the collapse, high-traffic applications and services such as Coinbase, Fortnite, Signal, Zoom, and Amazon’s own Ring devices experienced prolonged interruptions.
It is worth remembering that AWS is the largest cloud infrastructure provider globally, with a market share exceeding 30%. Millions of companies and institutions depend on its servers to operate their websites, mobile applications, and critical systems.
So far, Amazon has not provided explanations for the specific trigger of the failure.
This event is reminiscent of two other major recent Internet collapses: the one in 2024, when a faulty update to CrowdStrike’s software rendered millions of devices worldwide inoperative and caused massive airport delays; and the one in 2021, in which an error at DNS provider Akamai took giants like FedEx, Steam, and the PlayStation Network offline for hours.
With information taken from TechCrunch