The article details a service disruption of Amazon DynamoDB in the Northern Virginia region on October 19-20, 2025, caused by a latent defect in the automated DNS management system. This resulted in increased API error rates, connection errors for Network Load Balancers, and failed EC2 instance launches, affecting customer applications. The incident highlighted the importance of DNS automation for maintaining service reliability and performance.