An overheating incident at an Amazon Web Services data center in northern Virginia disrupted core trading services on Coinbase and FanDuel Thursday night, with AWS warning Friday morning that full recovery could take several more hours.
Amazon Web Services began reporting operational issues Thursday night after a thermal problem at a northern Virginia data center knocked out core services for millions of users, disrupting trading and access on platforms including Coinbase and FanDuel.
"Full recovery is still expected to take several hours," AWS wrote in an update posted at 9:51 a.m. ET on Friday.
The outage originated in AWS's US-East-1 region — its primary U.S. facility — and was isolated to a "single Availability Zone" within that region, the company said. AWS attributed the disruption to overheating at the affected data center and said engineers were working to restore cooling capacity.
"We are actively working to bring additional cooling system capacity online, which will enable us to recover the remaining affected hardware in the impacted zone," the company said in its 9:51 a.m. ET update.
AWS's health dashboard first flagged trouble at 8:25 p.m. ET Thursday, noting that it was "investigating instance impairments." The affected infrastructure includes EC2 instances, which provide virtual server capacity to a broad range of AWS customers.
FanDuel, the sports-betting platform, posted on X at 9:00 p.m. 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 problems were tied to the broader AWS outage. Users of the platform took to social media to complain about lost bets stemming from their inability to cash out.
Cryptocurrency trading platform Coinbase disclosed on X Friday that failures across multiple AWS zones "caused an extended outage of core trading services," adding that the primary issue had since been fully resolved.
AWS had no additional comment beyond its dashboard updates.
The scope of the disruption reflects the concentration of cloud infrastructure dependency among consumer-facing financial and gaming platforms. AWS accounts for roughly one-third of the global cloud infrastructure market, according to industry estimates, and provides backend services to millions of companies worldwide.
The Virginia outage comes as scrutiny of data center reliability and geographic concentration has intensified across the technology industry. A single thermal failure in one availability zone was sufficient to cascade into service interruptions affecting both cryptocurrency markets and sports wagering — underscoring the operational risk that centralized cloud infrastructure can pose to downstream platforms and their end users.
AWS has not provided a timeline for when all affected systems will be fully restored.
Jay Goldberg is a staff writer at TechEchelon covering technology, markets, and policy. He files the breaking news and deal coverage that move the publication's core desks.
More from Jay →Microsoft, Europol, and international partners disrupted the Amadey and StealC malware operations under Operation Endgame, seizing 326 servers, 142 domains, and identifying over $47 million in linked cryptocurrency.
CISA has confirmed active exploitation of CVE-2026-20253, a critical Splunk Enterprise flaw that allows unauthenticated remote attackers to manipulate files, and ordered federal civilian agencies to patch by Sunday under Binding Operational Directive 26-04.
Market intelligence platform Klue has confirmed attackers stole OAuth tokens connected to customer Salesforce environments on June 12, as the newly emerged Icarus extortion group publicly claims responsibility and the list of affected organizations continues to grow.
Written for readers who already know the basics — markets, AI, and the policy decisions that shape both.