AWS’s Day-Long Outage Explained

On October 19, 2025, AWS went down for over 24 hours when Dynamo’s DNS inexplicably lost its name-server records—knocking out name resolution across the board and taking critical services offline.

Ten days later the official RCA lays it all out, showing how that initial DNS failure cascaded into EC2 hiccups and Network Load Balancer errors. The deep-dive video timestamps: 0:00 summary, 5:30 DNS failure mechanics, 13:41 EC2 errors, and 16:16 NLB errors.

Watch on YouTube