Full Breakdown
AWS Overheating Outage in US-East-1 Disrupts Major Services
5/9/2026, 1:36:00 AM
Outage Overview and Context
On May 7 2026 a data centre in Northern Virginia (availability zone use1-az4 of the US-East-1 region) experienced a temperature rise that caused cooling-system failure and power loss. Amazon Web Services (AWS) said the thermal event impaired EC2 instances, EBS volumes and other services. US-East-1 is a heavily used AWS region that previously suffered large-scale outages in 2021, October 2025 (attributed to a DNS issue), and a 2024 CrowdStrike-related disruption.
Affected Companies, Services, and Impact
The outage hit customers. Coinbase saw delayed transactions and a temporary “cancel-only” mode; CME Group experienced latency and required maintenance before log-ins; FanDuel reported access problems. AWS restored EC2, EBS, IoT Core, NAT Gateway, Elastic Kubernetes Service, Elastic Load Balancing and Redshift, while ElastiCache, Managed Streaming for Apache Kafka, OpenSearch Service and SageMaker stayed impaired. The disruption delayed crypto trades, affected futures and options processing, and blocked gaming access, highlighting systemic risk from reliance on a single cloud provider.
Timeline of the Incident
- 5:25 PM PDT May 7 – AWS detects overheating in use1-az4.
- 6:47 PM PDT – Mitigation begins; AWS warns of elevated error rates.
- 8:06 PM PDT – “Incremental progress” reported; traffic shifted away.
- 5:11 UTC May 8 – Additional cooling capacity online, some racks recovered.
- Later May 8 – Coinbase confirms all markets re-enabled.
Official Statements & Responses
AWS described the event as an “increase in temperatures within a single data centre” causing “impairments for instances in the Availability Zone” and rerouted traffic to other zones. Coinbase said all markets were back online and funds remained safe. CME noted essential maintenance was completed and users could log in to CME Direct. FanDuel said its team was investigating the technical difficulties.
Criticism & Opposition
The Register called US-East-1 a “problem child,” suggesting the region’s scale adds stress. AWS executives replied the region is not inherently more fragile than other AWS locations.
On-the-Ground Reports
An AWS user reported, “All my servers in that region have just gone inaccessible, and AWSDashboard is misbehaving, with even the status page timing out.”
Conflicting Reports & Gaps
AWS has not disclosed the precise cause of the temperature rise nor a timeline for full recovery. Sources differ on whether the overheating stemmed solely from cooling-system failure or from a combination of power loss and thermal-management issues.
Verbatim Quotes
- “All markets have been re-enabled for trading on coinbase.com and in the Coinbase iOS and Android apps. Coinbase customers can log in to trade.” — Coinbase, Official Statement
- “We have experienced an increase in temperatures within a single data center, which in some cases has caused impairments for instances in the Availability Zone,” — Amazon Web Services, Status Report
- “Customers may experience longer than usual provisioning times.” — Amazon Web Services, Update (The Register)
- “Your funds are safe.” — Coinbase Support, X post
What’s Next
AWS will continue adding cooling capacity and advises customers to shift workloads to unaffected zones within US-East-1. No definitive ETA for full recovery has been provided.
