AWS Overheating Outage Disrupts Coinbase
May 8, 2026 – 11:18 am
A single data center’s cooling system fell behind in northern Virginia, leading to an overheating issue for Amazon Web Services (AWS). This disruption caused AWS to throttle and partly reroute traffic away from the affected Availability Zone, impacting various customer workloads.
According to Amazon Web Services, additional cooling capacity came online a couple of hours after the initial impact, with "early signs of recovery" appearing shortly afterward. However, bringing the site fully back online took longer than expected, leaving AWS unable to provide a specific restoration time.
Coinbase Confirms AWS-Related Outage
Coinbase confirmed that its trading platform issues were caused by the AWS event. After several hours of degraded markets, all markets were re-enabled and trading returned to normal.
The CME Group also reported issues with its CME Direct platform during the same period, attributing it to "essential maintenance." While they didn’t explicitly mention the AWS event as a factor, the timing suggests a potential connection.
Northern Virginia Cluster Affected
The affected data center is part of AWS’s oldest and busiest region, US-East-1 in Northern Virginia, which groups multiple physical data centers designed to operate independently. Customers running in the impacted zone were advised to failover to other zones.
This incident echoes a larger outage last October where a DNS resolution failure in DynamoDB caused disruptions across more than 100 AWS services, affecting platforms like Snapchat, Reddit, United Airlines, and Coinbase. Similarly, CME suffered an extended trading outage in early November due to a cooling failure at a CyrusOne data center in Chicago.
The recurring nature of these events highlights the vulnerability of single physical or logical sites becoming bottlenecks for significant public-facing traffic.