Amazon has confirmed that a wave of drone strikes in the Middle East directly targeted critical components of its cloud and web infrastructure, forcing several data centers offline and triggering a cascading series of service disruptions across multiple countries in the region. The incident represents one of the most serious physical infrastructure events to affect the company’s regional cloud operations in recent years.
Early reports emerged describing visible “sparks and a fire” at one Amazon cloud facility in the United Arab Emirates (UAE), followed by what the company characterized as a precautionary, temporary, and deliberate shutdown of systems. Emergency protocols were reportedly activated as a safeguard to protect both hardware and personnel. When Reuters sought clarification on whether Iranian drone strikes were responsible for the incident, Amazon declined to confirm or deny the attribution, maintaining a cautious and noncommittal public stance regarding the origin of the attacks.
However, an updated notice posted to Amazon’s Health Dashboard suggests the scope and severity of the disruption is considerably broader than initially understood. According to the revised disclosure, the incident now spans three separate facilities across two countries. Two sites in the UAE sustained direct physical impacts from drone activity, while a third data center in Bahrain experienced indirect structural damage caused by a nearby explosion. The blast wave reportedly affected surrounding infrastructure, leading to secondary technical failures inside the facility.
Although the company has refrained from naming Iran or any other actor as the source of the drone operations, the strikes—regardless of attribution—appear to have caused significant structural compromise, localized fires, electrical failures, and extended power outages. In at least one affected facility, automated water-based fire suppression systems were activated in response to smoke and heat detection. While the systems functioned as designed to contain the fire risk, they also introduced additional water damage to sensitive server racks, networking hardware, and storage equipment, compounding the operational challenges.
Amazon further acknowledged that core cloud management tools have been disrupted. “We can confirm that the AWS Management Console and the command line interface (CLI) are disrupted due to the failure of two Availability Zones,” the company stated. The loss of these Availability Zones has impaired customers’ ability to deploy, scale, monitor, and manage workloads in the impacted region. Businesses dependent on real-time data processing, cloud-hosted applications, and distributed services may be experiencing degraded performance, latency issues, or temporary service unavailability.
The company has characterized the operational environment as volatile and unpredictable, noting that restoration timelines remain uncertain as engineering teams assess physical damage, restore electrical stability, and validate system integrity. Industry observers note that physical damage to cloud infrastructure—particularly when involving multiple facilities—can have ripple effects across redundancy architectures designed to ensure uptime.
The broader geopolitical backdrop adds further complexity to the situation. Inside Iran, internet connectivity has reportedly been almost entirely severed since the launch of a U.S.-Israeli military campaign on February 28. Digital rights organizations and telecommunications analysts attribute the near-total blackout to deliberate actions by Iranian authorities, who are believed to have implemented sweeping network restrictions and traffic filtering measures.
Simultaneously, there are mounting reports of an active cyber-offensive targeting Iranian digital assets. Hackers appear to have compromised multiple Iranian news websites and online platforms, temporarily overriding standard content with politically charged messages urging citizens to mobilize against the ruling regime. This combination of physical infrastructure attacks, regional instability, internet blackouts, and cyber operations underscores the increasingly blurred lines between kinetic warfare and digital conflict.
Amazon’s Health Dashboard currently lists the severity level of the UAE facilities as “Disrupted,” indicating significant operational impairment, while the Bahrain site is categorized as “Impacted,” suggesting partial service degradation. The company has advised customers operating in or relying on Middle Eastern infrastructure to take immediate precautionary steps. These include backing up mission-critical data to geographically distinct regions, migrating workloads outside the affected zone where feasible, validating failover configurations, and activating preexisting disaster recovery and business continuity plans.
Customers are also being encouraged to reassess redundancy strategies, evaluate cross-region replication setups, and monitor service health notifications closely. Given the fluid security environment and the potential for additional disruptions, Amazon has emphasized vigilance and preparedness as restoration efforts continue.
The incident highlights the growing vulnerability of global cloud infrastructure to geopolitical conflict, demonstrating that even highly distributed and resilient systems can face operational strain when confronted with coordinated physical and digital threats. As investigations and recovery operations progress, further updates are expected regarding attribution, damage assessment, and long-term mitigation measures.