Amazon cloud goes dark in UAE after data center struck during Iran missile offensive

Amazon Web Services reported a major cloud infrastructure outage across its Middle East operations beginning on Sunday, March 1, 2026, after unidentified objects struck one of the company’s data center facilities in the United Arab Emirates, triggering a fire and forcing local emergency services to cut power to the affected site. The incident, which spread to a second Availability Zone the following day and produced connectivity disruptions in Bahrain, has drawn sustained attention from the global technology, financial services, and defense policy communities due to its proximity to simultaneous Iranian missile and drone strikes across the Gulf region.

Amazon Web Services published an initial status update on its health dashboard at 5:19 a.m. Pacific Standard Time on March 1, 2026, noting that connectivity and power issues had affected application programming interfaces and instances in a single Availability Zone, designated mec1-az2, within the Middle East Central 1 region. A subsequent update posted at 9:41 a.m. Pacific Standard Time confirmed that Availability Zone mec1-az2 had been impacted by objects that struck the data center, creating sparks and fire. The fire department shut off power to the facility and its backup generators while working to extinguish the fire. Amazon Web Services said the company was awaiting permission from authorities to restore power and that, once granted, it would restore power and connectivity safely. The company said it would take several hours to restore connectivity to the impacted Availability Zone, adding that other Availability Zones in the region were functioning normally at that time.

How the outage spread from mec1-az2 to multiple availability zones in the UAE region

Subsequent Amazon Web Services status updates confirmed that a second Availability Zone, designated mec1-az3, had also been affected by a localized power issue. With two of the three Availability Zones within the Middle East Central 1 region significantly impacted, Amazon Web Services reported that customers were experiencing high failure rates for data ingestion and egress. Customers in the remaining zone, mec1-az1, reported increased Elastic Compute Cloud application programming interface errors and instance launch failures.

Amazon Web Services stated that the disruption initially rendered Amazon Elastic Compute Cloud instances, Elastic Block Store volumes, and Relational Database Service databases unavailable within the affected zone, leading to elevated error rates across other services. The company confirmed that customers running workloads redundantly across multiple Availability Zones were not impacted by the outage.

A detailed service impact update published on March 2, 2026, at 2:53 a.m. Pacific Standard Time stated that when mec1-az2 was powered off at approximately 4:00 a.m. Pacific Standard Time on March 1, the Simple Storage Service continued to operate normally. As the second Availability Zone became impaired, Simple Storage Service error rates increased. Amazon Web Services advised customers to update their applications to ingest Simple Storage Service data to an alternate Amazon Web Services region and stated that restoration of the two Availability Zones would include a careful assessment of data health and any necessary repair of storage systems.

According to industry publication Data Center Knowledge, 38 services across the Middle East Central 1 region saw major disruptions during the outage, including Lambda, Elastic Kubernetes Service, and Virtual Private Cloud. Amazon Relational Database Service and DynamoDB experienced heavy degradation. Amazon Web Services restored 32 services including Simple Storage Service and CloudWatch, but core Elastic Compute Cloud networking application programming interfaces failed for several hours. Amazon Web Services said full recovery from issues in both the United Arab Emirates and Bahrain was expected to be many hours away and advised customers to back up critical data and shift operations to servers in unaffected Amazon Web Services regions. An Amazon Web Services representative declined to comment when contacted by Bloomberg.

In a separate update, Amazon Web Services said it was also investigating connectivity and power issues in its Middle East South 1 region in Bahrain. The company did not provide details on the cause of the Bahrain disruption.

What UAE financial institutions reported after the AWS Dubai disruption

Abu Dhabi Commercial Bank issued a statement on Monday saying that technical issues were affecting some of its platforms and mobile application users. The bank described the disruption as a region-wide information technology issue and did not link it directly to the Amazon Web Services outage. Multiple United Arab Emirates-based investment and banking platforms also reported service disruptions on Monday, though those institutions did not formally attribute the outages to the Amazon Web Services incident. Amazon Web Services has confirmed that its United Arab Emirates customer base includes Al Ghurair Investment and Dubai Islamic Bank.

Why the AWS UAE outage is significant in the context of the US-Iran conflict and Gulf security in 2026

Amazon Web Services did not identify the nature of the objects referenced in its status updates, nor did the company attribute the incident to any external actor. The outage occurred on the same day that Iran launched large-scale missile and drone attacks across the Gulf region. Iran’s Ministry of Defence stated that 137 missiles and 209 drones had been fired across the United Arab Emirates on Saturday alone. Iranian officials described the attacks as retaliation for United States and Israeli military strikes that killed Supreme Leader Ayatollah Ali Khamenei and other senior Iranian officials. The Gulf states, including Saudi Arabia, Qatar, Oman, Bahrain, Kuwait, and the United Arab Emirates, repelled the attacks. At least one person was killed at Abu Dhabi International Airport, and other critical infrastructure sites were reported to have been hit. Smoke was observed over a port in Dubai on Sunday after the government media office said the facility caught fire due to debris from an aerial interception.

Chris McGuire, Senior Fellow for China and Emerging Technologies at the Council on Foreign Relations, stated on social media platform X that if the disruption was caused by an Iranian drone strike, it would represent the first time a commercial data center had been physically targeted in an active armed conflict. Amazon Web Services has not confirmed or denied any connection between the data center incident and the broader regional military situation, and declined to comment to media on the matter.

What Amazon Web Services’ Middle East Central 1 region is and how it serves Gulf businesses

Amazon Web Services launched its Middle East Central 1 region in the United Arab Emirates in 2022, making it one of the company’s more recently established global infrastructure regions. An Availability Zone within Amazon Web Services infrastructure is defined by the company as one or more connected physical data centers featuring redundant power supply and networking capabilities. Availability Zones are designed as separate, isolated locations within each Amazon Web Services region with the intent that a failure in one zone will not affect the others. The Middle East Central 1 region contains three Availability Zones, of which two were simultaneously disrupted during the March 1, 2026, incident.

Amazon Web Services operates 123 Availability Zones across 39 regions globally. The United Arab Emirates region serves as a critical node for commercial cloud services across the Gulf Cooperation Council. Gulf Cooperation Council governments and enterprises have significantly expanded their reliance on hyperscale cloud providers over the past five years, with Amazon Web Services, Microsoft Azure, and Google Cloud all operating or developing data center capacity within the region.

The incident is described by technology industry analysts as the most significant Amazon Web Services outage since October 2025, when an operational issue in the United States East 1 region in Northern Virginia caused global service disruptions for several hours. Unlike the October 2025 outage, which stemmed from an internal operational failure, the March 2026 United Arab Emirates incident involved physical damage to the facility itself. The physical targeting of cloud infrastructure, whether by military ordnance, intercepted debris, or any other external object, has drawn renewed attention to the adequacy of existing civil-military coordination frameworks for protecting commercial critical infrastructure during armed conflict.

What the AWS UAE data center outage means for cloud infrastructure security and Gulf region businesses

• Amazon Web Services confirmed that unidentified objects struck its mec1-az2 Availability Zone in the Middle East Central 1 region in the United Arab Emirates on March 1, 2026, triggering a fire that forced the fire department to cut power to the facility and its backup generators.
• The outage expanded to a second Availability Zone, mec1-az3, causing high failure rates for data ingestion and egress across 38 Amazon Web Services services including Elastic Compute Cloud, DynamoDB, Lambda, and Elastic Kubernetes Service in the Middle East Central 1 region.
• Amazon Web Services also reported connectivity and power disruptions in its Middle East South 1 region in Bahrain, with full restoration in both regions expected to take many hours.
• The incident coincided with large-scale Iranian missile and drone strikes against Gulf Cooperation Council states, including the United Arab Emirates, in retaliation for United States and Israeli military strikes that killed Supreme Leader Ayatollah Ali Khamenei; Amazon Web Services has not attributed the data center damage to any external actor.
• The disruption affected major United Arab Emirates financial institutions including Abu Dhabi Commercial Bank, and has prompted analysis from defense and technology policy experts on the vulnerability of commercial cloud infrastructure to physical damage in active conflict zones.