Catenaa, Monday, October 27, 2025- Amazon Web Services said a software defect in its automated DNS management system caused the global outage that crippled parts of the internet last week.
The disruption, centered in the US-East-1 region, halted access to AWS’s DynamoDB database service for several hours on Monday, freezing websites, cloud platforms and news outlets, including Telecoms.com.
The company said the problem originated from a “latent race condition” in the automation process that generated incorrect and empty DNS records for DynamoDB endpoints.
AWS said its automation tools, which normally update and distribute DNS data to maintain capacity and performance, failed to correct the faulty records.
That error cascaded across its infrastructure, blocking access to critical data and applications.
In a detailed postmortem, AWS apologized for the failure and said steps are being taken to strengthen monitoring and recovery systems.
The company said it “recognizes how critical” its services are and pledged to improve reliability.
Analysts said the incident raised concerns about heavy reliance on automated systems in cloud infrastructure. The outage also reignited debate over redundancy and the resilience of public cloud networks.
AWS said an automation flaw in its DNS management system triggered a global outage, disrupting services and raising doubts over cloud reliability.
