Amazon's cloud unit, AWS, said on that power to its data centre in the United Arab Emirates was shut down temporarily after objects struck the facility, triggering sparks and a fire. AWS said the fire department shut off power to the facility and its generators as crews worked to extinguish the blaze.
Amazon's status page indicated a 'localized power issue has affected a single Availability Zone in the ME-CENTRAL-1 Region (mec1-az2).' AWS later clarified that the availability zone "was impacted by objects that struck the data centre, creating sparks and fire."
Services Disrupted Across the Region
Bloomberg reports that the affected zone, mec1-az2, experienced connectivity issues impairing Elastic Compute Cloud (EC2) instances, Elastic Block Store volumes, database services, and networking-related APIs. AWS reported increased error rates for several EC2 networking functions, including AllocateAddress and AssociateAddress.
Customers also reported errors when calling EC2 APIs, specifically networking-related APIs including AllocateAddress, AssociateAddress, DescribeRouteTable, and DescribeNetworkInterfaces.
The Geopolitical Backdrop
While it remains unclear whether the incident is directly linked to the broader regional conflict, the fire broke out on the same day Iranian projectiles struck the UAE in retaliation for US and Israeli strikes that killed Supreme Leader Ayatollah Ali Khamenei and other senior officials. The Iranian strikes hit airports, ports, and residential areas across the country and the wider Gulf.
On Saturday alone, Iran fired 137 missiles and 209 drones across the UAE, according to its Ministry of Defence; the number of attacks on Sunday had not yet been confirmed. At least one person was killed at Abu Dhabi's airport, and other critical infrastructure sites were hit.
When Reuters asked AWS whether the incident at the data centre was connected to the strikes, the company did not confirm or deny any connection.
Latest update: Signs of recovery, but no restoration timeline
AWS said it was seeing 'significant signs of recovery' for some systems, but power remained down at the centre, with the company saying it did not have an estimated time for power restoration.
AWS said customers could create and associate new network addresses in unaffected zones, and highlighted that other Availability Zones in the region remained operational. The company said it was routing traffic away from the affected data centre. AWS operates 123 such groups across 39 regions globally.
In a statement posted to its Health Dashboard, AWS said, "We are still awaiting permission to turn the power back on, and once we have, we will ensure we restore power and connectivity safely. It will take several hours to restore connectivity to the impacted AZ. The other AZs in the region are functioning normally."
As of the time of writing, full power had not been restored and no definitive timeline had been given.
