We now know why AWS’s DNS failed

AWS’s October 19, 2025 outage lasted more than 24 hours, and only after ten days did the team publish a root cause analysis. It turns out a misconfigured DynamoDB replication change knocked out its DNS, which then cascaded into EC2 errors and Network Load Balancer hiccups across multiple regions.

The full RCA is available on AWS’s blog, diving into each failure stage—DynamoDB’s DNS loss, EC2 issues, and NLB errors—that combined to derail countless services.

Watch on YouTube