Amazon Web Services says it’s dealing with disruptions in its Middle East (UAE) Region after “objects” struck a data center and triggered a fire, forcing an emergency power shutdown.
In an update on its AWS Health Dashboard, the company said the incident impacted one of its Availability Zones (mec1-az2) at around 4:30 a.m. PST, creating “sparks and fire.” AWS said the fire department shut off power to the facility and generators while crews worked the scene, and that AWS was still awaiting permission to restore power safely.
APC UPS 1500VA Battery Backup Surge Protector (BR1500G)
A consumer UPS that can keep networking gear and a small workstation running through short outages, and provides basic surge protection for home office setups.
Check Price on AmazonAWS has been steering customer traffic away from the impacted zone where possible, but warned that recovery could take hours as power and connectivity are brought back in a controlled way. In its status updates, AWS also urged customers to consider shifting workloads or ingest paths to alternate regions depending on their architecture and dependency on affected services.
The company has not publicly attributed what the “objects” were. Reuters reported that when asked whether the incident was connected to weekend strikes across the region, AWS neither confirmed nor denied a link.
AWS’ public health updates and reporting also pointed to separate issues in Bahrain, where power problems have affected parts of AWS’ footprint there as well—adding another complication for customers relying on Middle East regional redundancy.
Belkin 12-Outlet Surge Protector Power Strip
A multi-outlet surge protector meant for desks and small setups, helpful for reducing risk from power spikes when you have multiple devices in one place.
Check Price on AmazonThis is the kind of incident that illustrates why cloud “regions” matter in practice. A single Availability Zone going dark can be survivable—if customers are engineered for it. But when the disruption is tied to physical infrastructure and power restoration, the clock is often set by safety constraints rather than software rollbacks.
It’s also a reminder that AWS outages can ripple far beyond enterprise IT. A major AWS incident in October 2025 disrupted a wide range of consumer apps and services, underlining how many everyday products ultimately sit on the same cloud plumbing.