AWS UAE Data Center Temporarily Shut Down After Objects Strike Facility
Amazon Web Services (AWS) has confirmed a significant incident at its data center in the United Arab Emirates, where objects struck the facility, leading to sparks and a fire. The event occurred at approximately 4:30 AM PST on March 1 in the availability zone mec1-az2, part of the ME-CENTRAL-1 Region. In response, the UAE fire department cut power to the building, plunging the zone into darkness. AWS stated that other zones remain operational, but restoration of the affected area is expected to take several hours.
Impact on Cloud Services and Customer Recommendations
The AWS Health Dashboard currently shows services at the data center as 'Disrupted,' with specific impacts on Amazon Elastic Compute Cloud (EC2) and Amazon Relational Database Service (RDS). AWS has provided multiple updates, advising customers to use alternate Availability Zones or other AWS Regions where possible. The company is actively working on mitigations, such as enabling customers to disassociate Elastic IP addresses from affected resources. As of the latest updates, power restoration is still pending, with no estimated time of arrival provided.
Regional Context and Security Concerns
This incident unfolds against a backdrop of heightened regional tensions, including recent missile and drone strikes by Iran following US and Israeli actions. When questioned by Reuters about a potential connection between the data center incident and these strikes, AWS did not confirm or deny any link. The broader Gulf region has been affected by these conflicts, with reports of strikes targeting airports, ports, and residential areas. The data center outage adds to the challenges faced by the UAE during this volatile period.
Technical Recovery Efforts and Updates
AWS has been providing detailed status updates on its Health Dashboard, highlighting progress in recovering APIs like AssociateAddress and AllocateAddress. The company emphasizes that customers running applications redundantly across zones are not impacted, but those with resources in the affected zone may experience longer recovery times. Recommendations include retrying failed API requests and restoring from backups in unaffected zones. AWS continues to monitor the situation closely, with plans for further updates as restoration efforts proceed.
