Amazon Web Services (AWS), the world’s largest cloud computing platform, confirmed that it has restored normal operations following a massive outage that began early Monday, October 20, 2025, and lasted approximately 15 hours. The disruption affected thousands of popular websites, apps, and digital services worldwide, underscoring how dependent the internet infrastructure has become on a single cloud provider.

The outage originated around midnight PT (3 a.m. ET) in AWS’s primary data center in Northern Virginia, known as the US-EAST-1 region. Amazon reported an “operational issue” involving its Elastic Compute Cloud (EC2) internal network and a critical Domain Name System (DNS) subsystem responsible for routing requests to DynamoDB, AWS’s key cloud database service. The interruption caused cascading failures impacting 14 AWS services that many major companies rely on for their web presence and applications.

During the outage, users experienced widespread problems accessing services from major tech firms and platforms including Snapchat, Reddit, Roblox, Coinbase, Venmo, Robinhood, Amazon.com, Prime Video, Ring doorbells, and financial institutions like Lloyds Bank in the UK. Media companies such as Disney, The New York Times, and NPR also reported difficulties. Airlines including United and Delta noted interruptions in their online services.

According to Amazon’s status updates, the root cause was traced to a network health monitor subsystem that disrupted traffic load balancing within the EC2 infrastructure. Amazon’s public AWS Health Dashboard documented efforts to mitigate increased error rates and latency. By late Monday afternoon (around 6 p.m. ET), Amazon announced all AWS services had “returned to normal operations,” while acknowledging some services would take additional hours to clear backlogs created during the outage.

Tech experts like Mike Chapple, IT professor at University of Notre Dame, described the recovery phase as “slow and bumpy,” likening it to restoring power after a large blackout where smaller disruptions are common during the fix. Computer science professor Ken Birman of Cornell University emphasized the need for cloud customers to implement stronger fault tolerance and backup strategies across multiple providers to avoid total dependence on one service failing, warning that “companies that cut corners and skip these steps get scrutinized after outages.”

The scale of the outage demonstrated just how deeply integrated AWS is in the web ecosystem—its infrastructure supports government agencies, businesses, universities, and individual app developers worldwide. In 2024 alone, AWS generated $107 billion in revenue, representing 17% of Amazon’s total earnings, highlighting its critical commercial role.

This incident marks at least the third major disruption in the past five years linked to the US-EAST-1 data center, raising questions about resiliency and the concentration of cloud infrastructure in single locations. While Amazon has declined detailed public comment beyond technical disclosures, the company reaffirmed its commitment to restoring service reliability and transparency through its AWS Dashboard updates.

For businesses and consumers, the outage was a stark reminder of the vulnerabilities in the internet’s backbone and the risks posed by centralization of cloud services. Experts recommend multi-cloud strategies and comprehensive disaster recovery plans to mitigate future risks.

As AWS continues to clear service backlogs in the coming hours, clients are advised to monitor updates on the AWS Health Dashboard and review their systems for any residual disruptions. Amazon’s engineering teams are expected to conduct a thorough post-mortem to prevent recurrence.

In summary, while Amazon Web Services has returned to normal operations following its longest outage in recent years, the incident has spotlighted the critical importance of cloud infrastructure stability and the growing urgency for businesses to build resilience into their digital operations. Continued vigilance and strategic planning will be key next steps for cloud-dependent enterprises moving forward to safeguard against similar global disruptions.