Coinbase says a heat problem in an Amazon Web Services (AWS) data center caused its latest round of service disruptions. The exchange says the root cause was a problem inside Amazon Web Services’ US‑EAST‑1 cloud region, not an internal security breach.
How AWS Cooling Problems Hit Coinbase
AWS reported higher temperatures in a single data center in Northern Virginia, which hosts part of the US‑EAST‑1 region. The heat damaged some EC2 instances and EBS volumes, and AWS warned that services in that zone could face “impairments.”
Coinbase said the overheating in use1‑az4 caused degraded performance across parts of its exchange operations. Users struggled to place trades, check balances, and complete withdrawals during the disruption. However, Coinbase stressed that customer funds stayed safe and that the outage affected access and transaction processing, not custody.
AWS tried to stabilize the situation by restoring cooling systems and shifting traffic away from the affected zone. The company said cooling progress was “incremental” and slower than expected, so some services stayed impaired for several hours.
Why This Outage Lasted Longer Than Usual
According to Coinbase, the May 7 incident involved failures affecting more than one AWS zone in the region. Because the trouble spread beyond use1‑az4, some of the usual failover paths did not work as expected. As a result, core trading services experienced an extended outage rather than a brief slowdown.
While the incident was unfolding, users reported errors when placing trades, checking balances, and submitting withdrawals. Coinbase states that customer funds remained safe in custody systems and that the disruption mainly affected access and transaction processing. At the same time, AWS teams worked to restore temperature controls and other managed services tied to the affected infrastructure.
Once Coinbase understood the nature of the problem, it began bringing the platform back online in stages. The exchange used a phased recovery plan to avoid overloading systems while parts of the underlying cloud environment were still unstable. In similar incidents, Coinbase Exchange has sometimes used “cancel only” modes to let customers cancel open orders before resuming full trading.
Coinbase stated that the main outage had been fully resolved by the end of the incident. Thanking customers for their patience, the business advised that anyone with account-level inquiries contact its support staff. Support employees are ready to review certain account histories and transactions that may have been affected during the disruption window.
READ MORE: MSTR Stock Retreats as Bitcoin Price Stalls at $80k: Here’s Why it May Rebound