Overview
- AWS reports a bug in DynamoDB’s DNS system in US‑EAST‑1 created an invalid empty record that blocked endpoint resolution and required manual intervention.
- The disruption lasted roughly 15–16 hours and cascaded across thousands of services worldwide, with Brazilian platforms such as PicPay, MercadoPago, iFood and Mercado Livre among those affected.
- EC2 in the same region struggled to start new instances during the incident, compounding recovery for workloads that needed additional capacity.
- Analyst tallies place AWS near one‑third of the global cloud market, including 37% per Gartner and 30% per Synergy Research, heightening systemic impact when core services fail.
- Recent guidance from coverage urges redundancy across multiple cloud providers, stronger continuity plans and using financial services with alternative systems to reduce future exposure.