A widespread service interruption at Amazon Web Services (AWS), the dominant cloud provider powering much of the internet, took numerous websites and applications offline for over two hours on Monday. This incident marks the latest in a series of events exposing the inherent fragility of our global technology infrastructure.
The disruption, which impacted major banks, gaming platforms, and entertainment services, began just after 3 a.m. Eastern Time. By 5:27 a.m., Amazon confirmed that most affected services were back online, though engineers were still actively managing a backlog of queued requests.
Among the high-profile services affected were WhatsApp, the official British government website, government tax services, the payment app Venmo, the cryptocurrency exchange Coinbase, and even games hosted by The New York Times. Dozens of other prominent companies and retailers, including Amazon itself, Hulu, Snapchat, Ring doorbells, and McDonald’s, also reported significant service interruptions.
The precise cause of the outage remains unknown, but initial assessments found no evidence to suggest a cyberattack was responsible.
This incident, experts note, once again underscores the internet’s heavy dependence on a handful of major technology providers like Amazon, Microsoft, and Google. When any one of these giants experiences an issue, it can trigger widespread disruptions impacting millions of users globally. For instance, a similar, day-long internet outage last year was traced back to a faulty update from a lesser-known cybersecurity company, CrowdStrike.
Amazon Web Services boasts thousands of clients who rely on its infrastructure for complex, high-demand, data-intensive operations. These include everything from streaming video and running sophisticated web applications to storing colossal amounts of digital information. Amazon’s cloud-computing division has strategically placed its infrastructure worldwide, enabling companies to offer their products seamlessly to customers across the globe. By leveraging AWS instead of building their own costly data centers, clients gain the flexibility to rapidly scale their operations up or down without massive capital investments in hardware.
Harry Halpin, CEO of NymVPN, a virtual private network service, suggested that the problem might have originated from a technical fault within one of Amazon’s primary data centers. However, he emphasized the inherent lack of transparency in cloud platform operations, making it impossible to ascertain the true cause—including the possibility of a cyberattack—unless Amazon chooses to disclose it.
Halpin, whose company provides VPN services to soldiers in Ukraine, shared that he awoke to a flurry of emails from troops on the front lines, all inquiring about the cause of the disruption. He stressed that this vulnerability isn’t unique to Ukraine but extends to many other Western governments that similarly depend on these crucial cloud services.
“If your entire nation’s infrastructure relies on a few providers, all in the United States, and anything can go down at any moment, either for malicious reasons or just technical errors, that’s an exceedingly dangerous situation,” he stated.
“Everyone takes it for normal,” remarked Dr. Halpin, a former research scientist at the Massachusetts Institute of Technology, referring to the internet’s reliance on a few dominant providers. “But it’s not normal.”
In its initial public statement, Amazon confirmed an “operational” issue impacting multiple services in its Northern Virginia region. The company assured the public that its engineers were diligently working to mitigate the effects and pinpoint the root cause.
Amro Al-Said Ahmad, a computer science lecturer at Keele University in England, pointed out that Amazon’s “us-east-1” region in Northern Virginia houses one of its largest data centers. While cloud computing generally functions smoothly for daily operations, he explained, even a minor error, such as a flawed software update, has the potential to bring down the entire system.
Advocates for media freedom voiced concerns that the outage, which also disrupted secure communication apps like Signal and other digital tools, highlights the threat posed to free speech by the internet’s concentration of power among a few large technology companies.
“When a single provider goes dark, critical services go offline with it,” stated Corinne Cath-Speth, head of digital for Article 19, a free speech advocacy group. She emphasized the urgent necessity for greater diversification in cloud computing services.
She concluded, “The infrastructure underpinning democratic discourse, independent journalism and secure communications cannot be dependent on a handful of companies.”
Despite the widespread disruption, Amazon’s share price showed little movement in premarket trading, indicating that investors were not overly concerned by the outage. In the first half of the year, Amazon Web Services was a significant profit driver for the company, contributing nearly 20 percent of Amazon’s total sales and approximately 60 percent of its operating profit.
Melissa Eddy in Berlin and Andrés R. Martínez in Seoul contributed to this report.