AWS Resolves Major Outage After Nearly 24 Hours of Service Disruption

Amazon Web Services resolved a major outage in its US-EAST-1 region that lasted nearly 24 hours and affected over 140 services, with ongoing backlogs for some components after the primary restoration. The disruption began late on 19 October 2025 and was fully resolved by the afternoon of 20 October, after a sequence of cascading failures centred on DNS resolution for DynamoDB endpoints.

The incident started at about 11:49 PM PDT on 19 October, with AWS engineers observing increased error rates and latencies across multiple services, and by 12:26 AM on 20 October the trigger had been identified as DNS resolution issues affecting regional DynamoDB service endpoints.

After resolving the DynamoDB DNS issue at 2:24 AM, AWS confronted further impairments in EC2’s internal subsystem and Network Load Balancer health checks, which led to throttling of operations such as EC2 instance launches and Lambda-related processes. By 3:01 PM PDT on 20 October all AWS services had returned to normal operations, though some services including AWS Config, Redshift and Connect continued processing backlogs for several hours.

AWS has committed to sharing a detailed post-event summary and recommends customers configure Auto Scaling Groups across multiple Availability Zones to improve resilience.