Amazon Identifies the Issue That Broke Much of the Internet, Says AWS Is Back to Normal
On October 20, 2025, a significant outage at Amazon Web Services (AWS) sent shockwaves across the global internet, disrupting everything from social media and gaming to financial services and airline ticketing. By the afternoon of October 21, Amazon confirmed that services were largely back to normal, but not before the incident had exposed—once again—the fragility of the modern digital ecosystem and the outsized role played by a handful of cloud providers[4][5][11].
What Happened?
The outage began at approximately 07:55 UTC, primarily affecting AWS’s Northern Virginia (US-EAST-1) region, one of the company’s most critical data center clusters[1]. This region hosts a vast array of services and acts as a backbone for countless applications and websites. The root cause was traced to a Domain Name System (DNS) issue, specifically impacting the DynamoDB API—a core database service used by many AWS customers to store and retrieve user data[1][5][11]. When DNS resolution failed, applications could no longer find the correct backend servers, effectively severing the connection between front-end services and their data stores[1][5][12].
Third-party monitoring by ThousandEyes confirmed that the problem was internal to AWS’s infrastructure, with no coinciding network events that might have pointed to external causes[1]. This made troubleshooting and recovery more complex, as the issue was not a simple connectivity problem but a breakdown in the fundamental mechanisms that allow cloud applications to locate and communicate with their backend services.
Global Impact
The effects were immediate and widespread. Major platforms such as Slack, Atlassian, Snapchat, Reddit, Facebook, Fortnite, Disney+, Hulu, Delta Air Lines, United Airlines, McDonald’s, Coinbase, and even AI firm Perplexity reported service disruptions[1][5][12]. In cities from London to Tokyo, workers were locked out of their digital tools, businesses struggled to process payments, and consumers found themselves unable to complete everyday tasks—from changing airline tickets to paying hairdressers[5][11]. Even services like Venmo and Zoom experienced lingering difficulties as the recovery process continued[5][11].
The outage was described as the largest internet disruption since the 2024 CrowdStrike incident, which had previously hobbled hospitals, banks, and airports[5][11]. The scale of the disruption highlighted how much of the world’s digital infrastructure is concentrated in the hands of a few cloud providers, and how quickly a single point of failure can cascade into a global crisis.
The Recovery Process
AWS began rolling out fixes around 09:22 UTC, with the most severe issues clearing by 09:35 UTC[1]. However, full normalization took much longer. Amazon stated that “services returned to normal operations” by 6 p.m. Eastern Time, but acknowledged that some AWS services still had a backlog of messages to process, which could take several additional hours[5][7]. Users continued to report residual delays and intermittent glitches as engineers worked through the recovery process[7][12].
Cybersecurity expert Mike Chapple compared the recovery to restoring power after a city-wide blackout: “While a city’s power is coming back online, neighborhoods may see intermittent glitches as crews finish the repairs”[7]. This analogy underscores the complexity of modern cloud infrastructures, where restoring service is not a single switch-flip but a phased process that can expose users to ongoing instability.
Why Did This Happen?
Amazon attributed the outage to a DNS resolution failure affecting the DynamoDB API[1][5][11]. DNS acts as the internet’s address book, translating human-readable domain names into machine-readable IP addresses. When this system fails, even robust backend systems become inaccessible because applications cannot find their data stores. In this case, the failure was internal to AWS’s architecture, not the result of a cyberattack or external network event[1][12].
This is not the first time the US-EAST-1 region has been at the center of a major internet outage. It is at least the third such incident in five years, raising questions about the resilience of this critical infrastructure hub[5][11]. Amazon has not provided additional clarity on why this region seems particularly vulnerable, leaving customers and industry observers to speculate about underlying architectural or operational weaknesses[5][11].
Broader Implications
The October 2025 AWS outage is a stark reminder of the concentration risk inherent in today’s cloud computing landscape. As more businesses, governments, and critical services migrate to the cloud, they become increasingly dependent on a small number of providers. AWS, Microsoft Azure, and Google Cloud dominate the market, and their operational health is now synonymous with the health of the global internet.
This incident also highlights the importance of redundancy and disaster recovery planning. Organizations that rely on a single cloud provider—or a single region within that provider—are especially vulnerable to cascading failures. Best practices increasingly recommend multi-cloud strategies, geographic distribution of services, and robust failover mechanisms to mitigate the impact of such outages.
Looking Ahead
In the immediate aftermath, Amazon has assured customers that data remained secure throughout the incident, and no evidence of a security breach has been reported[12]. The company is likely to conduct a thorough post-mortem to identify the exact sequence of events and implement safeguards to prevent recurrence. However, given the history of similar outages, the broader tech community will be watching closely to see whether meaningful changes are made to improve the resilience of critical cloud infrastructure.
For now, the digital world is breathing a sigh of relief as services return to normal. But the lesson is clear: in an age where the internet is the backbone of commerce, communication, and critical infrastructure, the stakes for reliability and redundancy have never been higher.
Conclusion
The AWS outage of October 2025 was a dramatic demonstration of both the power and the peril of centralized cloud computing. While Amazon has identified and resolved the technical issue, the disruption serves as a wake-up call for businesses and governments worldwide. As society’s dependence on cloud services grows, so too does the need for robust, redundant, and resilient digital infrastructure. The internet may be back to normal—for now—but the conversation about how to prevent the next great outage is just beginning.
Original source: TechCrunch – Amazon identifies the issue that broke much of the internet, says AWS is back to normal
The post Amazon’s AWS Outage Sparks Global Internet Chaos; Services Restored, Highlights Cloud Dependency Risks first appeared on Limited Liability Solutions.
Source: Read More