AWS Outage Knocks Out Coinbase, FanDuel After Virginia Data Center Overheats

Alexa Alix Last Updated: May 10, 2026 3 minutes read A data center cooling failure at one of Amazon Web Services' northern Virginia facilities knocked out trading and sports betting services for millions of users across Thursday night and into Friday, May 8, 2026.

AWS confirmed the outage originated from a thermal event inside a single availability zone and took over 12 hours to bring services back to stable levels.

What Caused the Outage AWS engineers first detected thermal problems at approximately 5:25 PM PDT on Thursday, May 7, in availability zone use1-az4 of the US-EAST-1 region, the company's oldest, largest, and most heavily used data center hub in northern Virginia.

The overheating triggered a power loss that damaged the EC2 virtual server instances and EBS storage volumes housed in the affected hardware racks. Amazon described the incident as a “thermal event resulting in a loss of power” at the facility, and said its main recovery effort centered on restoring cooling system capacity.

By May 8 at 1:50 PM, the company said it stabilized cooling to pre-event levels, which allowed the majority of impaired EC2 instances and EBS volumes to come back online.

According to AWS's official health dashboard, AWS noted in a 3:29 PM ET update on Friday that “full recovery is still expected to take several hours,” adding that “efforts are slower than we had previously anticipated.” A small number of instances and EBS volumes remained impaired even after cooling systems were restored.

Who Got Hit and How Hard The disruption hit cryptocurrency exchange Coinbase hardest, knocking out core exchange functions for more than five hours. Other reported victims included the CME Group trading platform and major gambling company FanDuel.

As CNBC reported, FanDuel posted on X at 9:00 PM ET Thursday that its team was “aware and investigating the current technical difficulties prohibiting users from accessing our platform.” Two hours later, the company confirmed the issue was tied to the broader AWS outage, with gamblers complaining about lost bets from being unable to cash out.

Coinbase users reported failed transactions, delayed withdrawals, and difficulty accessing trading services throughout the disruption. Some blockchain integrations, including Solana and ALEO transfers, also experienced delays. Throughout the outage, Coinbase maintained that customer funds remained safe and secure.

The affected services covered a wide range of AWS products.

The outage disrupted the following categories of infrastructure within the use1-az4 availability zone: EC2 compute instances and EBS storage volumes Amazon S3, DynamoDB, and dependent services such as Lambda and Kinesis API Gateway, ElastiCache, and Elastic Load Balancing A Rough Week for Coinbase The outage compounded what was already a difficult stretch for Coinbase.

The exchange cut 700 employees on Monday, May 5, and reported a $394 million quarterly loss on Thursday, May 7, before its platform went dark for seven hours on Friday. As The Next Web noted, the timing compounded operational challenges for the exchange, which had experienced previous outages tied to AWS infrastructure issues, including one in October 2025.

This was the second major AWS outage in seven months. Coinbase confirmed all customer funds remained secure and that all markets were re-enabled once services recovered.

The Bigger Picture for Cloud Reliance AWS accounts for roughly one-third of the global cloud infrastructure market, which means a single availability zone failure in northern Virginia has the reach to disrupt financial services, crypto exchanges, and consumer applications simultaneously.

Denmark has paused grid connections for new data centers as AI infrastructure strains the country's power grid, and the thermal failure in Virginia points to the same underlying tension from the opposite direction. Data centers generate enormous amounts of heat, and when the systems built to remove that heat fail, the servers shut down.

As more industries shift their core operations onto cloud infrastructure, the physical requirements of keeping that infrastructure cool become as critical as the software layer running on top of it. Recovery Status AWS said the majority of affected services returned to normal operations by the afternoon of May 8.

A small subset of EC2 instances and EBS volumes continued to show impairment as engineers worked through remaining hardware recovery in a controlled manner. As IT Pro detailed in its breakdown of the event, AWS brought affected racks back online progressively to avoid additional instability.

For e-commerce operators, platform builders, and digital businesses running workloads in the US-EAST-1 region, this outage serves as a concrete reminder to test multi-region failover configurations and ensure your disaster recovery plans account for extended