Coinbase suffered a near six-hour trading outage after an overheating incident at an Amazon Web Services data center in Northern Virginia disrupted core cloud services. The thermal event affected AWS’s use1-az4 availability zone in the us-east-1 region, impairing EC2 instances and EBS volumes and stopping order flow across Coinbase’s platform.
AWS said the incident began on May 7, 2026, at 4:20 PM PDT, when elevated temperatures overwhelmed cooling systems in a single availability zone and triggered power interruptions. Coinbase notified users of degraded performance at 6:53 PM PDT and moved into “Cancel Only” mode, allowing order cancellations while blocking new trades.
Trading Halt Hits Users Across Core Functions
Coinbase began restoring trading services on May 8 at 00:49 AM PDT, leaving the publicly acknowledged disruption at about 5 hours and 56 minutes. During that window, users were unable to place trades, verify balances or withdraw assets, while delays were also reported for certain blockchain integrations, including Solana and ALEO.
All markets have been re-enabled for trading on Coinbase Exchange. https://t.co/WQtg90bNqC
— Coinbase Support (@CoinbaseSupport) May 8, 2026
The exchange did not report any customer fund losses or data breaches tied to the outage. Still, the inability to trade or access balances created direct client friction, especially for users relying on real-time execution during volatile market conditions.
The disruption came during a difficult period for Coinbase. Days earlier, the company announced roughly 700 job cuts, equal to about 14% of its workforce, and reported first-quarter 2026 results showing a $394 million net loss and $1.41 billion in revenue, down 31% year over year.
That timing made the outage more commercially sensitive. A trading halt removes transaction activity precisely when volume pressure and weaker revenue were already weighing on the business, turning a cloud incident into a platform availability and revenue-risk event.
Centralized Cloud Risk Moves Back Into Focus
Operationally, the outage highlighted the crypto industry’s reliance on centralized cloud infrastructure. Even though cloud regions are designed to limit cross-zone impact, a localized physical failure involving cooling and power systems still cascaded into material service disruption for a global exchange.
The incident also raises broader infrastructure questions as data-center demand increases. Industry reporting has tied some of that pressure to AI workloads, making thermal and power resilience a growing operational concern for platforms that depend on high-availability cloud services.
Backup systems, failover design and incident response plans must account for availability-zone failures that can impair core trading, custody and withdrawal workflows.
Coinbase’s outage shows how platform availability can quickly become a market-structure issue. The practical takeaway is that continuity disclosures and post-incident root-cause reporting will remain central to assessing operational resilience across regulated crypto venues.