Amazon.com’s cloud computing unit said Monday it had largely contained a widespread internet outage originating from a data center in northern Virginia, but lingering issues continued to frustrate users of popular apps like Snapchat, Reddit and Venmo.
The disruption, which AWS attributed to problems in its Domain Name System that blocked applications from accessing the DynamoDB API — a key cloud database for user data — affected thousands of websites and services worldwide. It marked the third major meltdown linked to AWS’s US-EAST-1 region in five years, underscoring the fragility of the global digital infrastructure.
AWS, the world’s largest cloud provider ahead of Microsoft’s Azure and Alphabet’s Google Cloud, powers everything from food delivery apps to airline booking systems. The outage knocked workers offline from London to Tokyo, halting routine tasks like paying hairdressers or rebooking flights.
“Many applications were gradually coming back online” in the U.S. afternoon, AWS said in updates posted to its status page. However, the company acknowledged elevated error rates persisting in several services.
Outage tracking site Downdetector, owned by Ookla, reported over 4 million user complaints and impacts on at least 1,000 companies. While Reddit and Roblox had mostly stabilized, Snapchat and language app Duolingo saw resurgent problems. Video calling platform Zoom and digital wallet Venmo also drew complaints of slow performance or failures.
In Britain, banks Lloyds and Bank of Scotland, telecoms Vodafone and BT, and the HMRC tax authority website were among those hit, per Downdetector’s U.K. tracker.
Experts warned the episode exposed overreliance on a handful of cloud giants. “This outage once again highlights the dependency we have on relatively fragile infrastructures,” said Jake Moore, global cybersecurity adviser at ESET.
Nishanth Sastry, research director at the University of Surrey’s computer science department, pointed to a core issue: “The main reason for this issue is that all these big companies have relied on just one service.”
The meltdown evoked last year’s CrowdStrike software glitch, which crippled hospitals, banks and airports globally — the largest tech disruption in recent memory.
Wall Street shrugged off the news, lifting Amazon shares 1.6% to $216.48.
AWS said it had identified and addressed the root DNS issue, with full resolution nearing. The incident renewed calls for diversified cloud strategies amid growing warnings about single points of failure in an increasingly interconnected world.
