TL;DR: AWS suffered a massive DNS outage starting October 19, 2025 that lasted over a day. Ten days later they published a root-cause analysis blaming a misstep in Dynamo’s DNS configuration, a cascade of EC2 errors, and faults in the Network Load Balancer.
The report walks through how DynamoDB lost its DNS entries, how EC2 instances went haywire under retry storms, and how the NLB failures compounded the whole mess—offering a step-by-step look at the sequence that brought large parts of AWS to a standstill.
Watch on YouTube
Top comments (0)