Overview
- AWS said the incident was resolved late Oct. 20 after roughly fifteen hours, with some residual slowdowns as systems clear queued requests.
- Initial status updates pointed to DNS issues and elevated DynamoDB errors before AWS identified the internal monitoring subsystem as the root cause.
- Disruptions hit widely used services including Snapchat, Airbnb, Fortnite, Roblox, Reddit, Prime Video, Signal, Perplexity, and Coinbase, which told users that all funds were safe.
- Downdetector recorded thousands of outage reports centered in the United States, especially Virginia, consistent with problems in the US‑East‑1 region.
- Experts said the episode underscored concentration risk in cloud infrastructure and urged resilience strategies such as multi‑region or multi‑provider designs.