Amazon Web Services has declared it cannot restore access to customer resources and data in parts of its Middle East cloud infrastructure after datacenters were damaged during the US war with Iran, according to an update posted to the company's AWS Health Dashboard. The cloud giant confirmed that anything hosted exclusively in its Bahrain Region remains permanently inaccessible, while one Availability Zone in the United Arab Emirates has been declared unrecoverable. The announcement marks a rare instance of a major cloud provider acknowledging the total loss of customer data due to physical infrastructure destruction.
The damage began in March when the first Bahrain Availability Zone was struck, prompting AWS to advise customers to migrate their workloads to other Regions. Most customers followed that guidance before additional attacks in April disabled a second Availability Zone, rendering the entire Bahrain Region—designated me-south-1 in AWS's system—unavailable. In the UAE, two AWS facilities were hit by drones in March as Iran retaliated against US-Israeli strikes on its territory. One of the three Availability Zones in the UAE Region, identified as mec1-az2, has now been deemed unrecoverable. AWS says it continues working to restore resources in the two other affected UAE Availability Zones, mec1-az1 and mec1-az3, while noting that most customers have restarted operations in other AWS Regions by restoring from backups or copying accessible data.
According to the Health Dashboard update, "the damage to our infrastructure spanned multiple Availability Zones and exceeded what our regional and multi-AZ services are designed to withstand." After completing a thorough assessment, the company determined it's unable to restore access to resources and data hosted exclusively in the Bahrain Region. AWS and several other tech companies advised customers after the first strikes to recover whatever resources they could and replicate them in alternate Regions, ideally in Europe, rather than wait for services in the Middle East to return. That guidance has proven well-founded.
The incident exposes the limits of AWS's standard high-availability architecture, which distributes applications across multiple Availability Zones within a single geographic Region to withstand the loss of a single location. Each Zone consists of one or more discrete datacenters with independent power, cooling, and network connectivity. That approach can't protect against a Region-wide failure, however. Workloads that must survive such an event require a disaster recovery plan involving another Region entirely. AWS told customers in March to enact their disaster recovery plans, recover from remote backups stored in other Regions, and update their applications to direct traffic away from the affected Regions.
The company says its support teams remain available to help customers move their applications to alternate Regions, though it declined to provide additional comment beyond the Health Dashboard information. AWS's acknowledgment that some customer data is permanently lost underscores the distinction between cloud redundancy within a Region and true geographic disaster recovery. Organizations operating in geopolitically volatile areas now face a hard calculation about where physical infrastructure can survive and where only distributed, multi-Region architectures offer genuine protection. The cloud's promise of resilience ultimately depends on hardware that remains vulnerable to missiles.