Overview
- Amazon Web Services reported increased error rates in its US‑East‑1 region at 3:11 a.m. ET, initially tying the disruption to a DNS resolution issue affecting the DynamoDB API.
- AWS said the DNS issue was fully mitigated by about 6:35 a.m. ET, then later identified EC2 internal-network failures tied to a subsystem that monitors network load balancers.
- Engineers applied mitigations, processed backlogs and temporarily rate‑limited new EC2 instance launches in US‑East‑1, with most service operations returning to normal.
- The outage cascaded across sectors, interrupting services such as Snapchat, Fortnite, Roblox, Coinbase, Signal, Canva and UK government sites, with banks and some airline check‑in systems also affected.
- Outage trackers logged millions of reports for more than 1,000 services, and experts said the incident underscores dependence on a few cloud providers rather than indicating a cyberattack.