Amazon’s AWS outage hits UAE and Bahrain after data center fire triggered by ‘objects’; link to Iranian strikes unclear

AWS customers across the Middle East woke up to disruption this week after a fire at one of the company’s regional data centers knocked parts of its cloud network offline.

Amazon Web Services said its United Arab Emirates facilities were hit by power and connectivity problems on Monday after the site was struck by what the company described as “objects,” which sparked a fire and forced systems offline. The incident rippled across the company’s UAE and Bahrain regions, leaving some customers scrambling for workarounds.

According to AWS’ public status page, two availability zones in the UAE region lost power. Availability zones are clusters of data centers designed to operate independently, so multi-zone outages tend to draw close attention from enterprise customers who rely on the architecture for redundancy.

AWS first disclosed Sunday that one UAE zone had been affected after “objects” struck the data center, sparking fires and creating sparks, triggering a shutdown. On Monday, the company reported further trouble.

“We can confirm that a localized power issue has affected another availability zone” in the UAE region, AWS said, according to a report from Reuters.

In a later update, the company reported increased error rates tied to Amazon S3 in the ME-CENTRAL-1 region after two availability zones, identified as mec1-az2 and mec1-az3, became impaired. AWS said S3 is built to withstand the loss of a single zone, though the second failure pushed error rates higher for data ingest and egress.

The company urged customers to route S3 traffic to another AWS region where possible. It added that the AWS Management Console and command-line interface (CLI) were experiencing disruptions due to the dual-zone failure.

AWS did not confirm any connection between the incident and Iranian strikes reported across parts of the Gulf, including the UAE and Bahrain, following recent U.S. and Israeli attacks on Iran. The cause remains officially unlinked.

The company did not confirm any connection between the incident and Iranian strikes reported across parts of the Gulf, including the UAE and Bahrain, following recent U.S. and Israeli attacks on Iran. For now, the cause remains officially unlinked.

AWS said partial recovery had begun in the UAE earlier Monday, though the company warned customers that full restoration could take time. It advised affected users to shift workloads to other regions where possible.

The cloud division said full recovery for both the UAE and Bahrain regions was expected to be “many hours away.”

The disruption appeared to ripple beyond AWS customers. Abu Dhabi Commercial Bank said Monday that technical problems were affecting some of its platforms and its mobile app. The bank did not say whether the issues were tied to the AWS outage.

The incident is a reminder of how localized infrastructure events can cascade across cloud-dependent services, particularly in regions where fewer availability zones serve large volumes of traffic. For companies running mission-critical workloads, the episode is likely to renew focus on multi-region failover strategies and geographic redundancy.

AWS continues to work on restoring full service.