Particle.news
Download on the App Store

AWS Restores Services After US‑EAST‑1 DNS Failure Triggers Global Outages

AWS cites a DynamoDB‑endpoint DNS fault in North Virginia as the cause, with residual backlogs clearing as it readies a fuller incident report.

Overview

  • The cloud provider reported increased error rates and latency across multiple services in US‑EAST‑1, leading to disruptions for consumer apps, games, fintech platforms and Amazon’s own services.
  • AWS identified DNS resolution problems affecting DynamoDB endpoints at about 09:26 local reporting, implemented fixes by roughly 11:24, and said all services returned to normal by 00:01 on Oct 21.
  • Recovery was slowed by follow‑on issues in EC2 instance starts and network load balancer health checks, prompting temporary throttles until systems stabilized.
  • Outage trackers logged millions of user reports across more than 2,000 sites, with affected services including Signal, Snapchat, Roblox, Fortnite, Zoom, Slack, Coinbase, Venmo and Perplexity.
  • AWS says most systems are operating normally now, though some customers may still see delays or elevated error rates as queues drain and a detailed postmortem is forthcoming.