Amazon Web Services (AWS), the dominant cloud computing platform powering much of the internet, experienced a major outage on Monday, October 20, 2025, causing widespread disruptions to numerous popular websites, applications, and services globally.
What Happened
The outage began around 12:11 AM ET (midnight Pacific time) and primarily affected AWS’s US-EAST-1 Region in northern Virginia, one of Amazon’s largest and most critical data centers. AWS reported “increased error rates and latencies” across multiple services, with 28 services initially impacted.
The root cause was identified as a Domain Name System (DNS) failure. DNS functions as the internet’s directory, translating user-friendly web addresses (like amazon.com) into numerical IP addresses that computers can understand. When this system failed, it prevented websites and applications from accessing their data stored in AWS’s DynamoDB database, even though the data itself remained secure.
Services and Platforms Affected
The outage disrupted a wide range of major services across multiple sectors:
Gaming Platforms:
-
Fortnite (Epic Games)
-
Roblox
-
Clash Royale and Clash of Clans
-
Pokémon GO
-
EA App and Battlefield 6
Financial Services:
-
Coinbase (cryptocurrency exchange)
-
Robinhood (trading app)
-
Venmo (PayPal)
-
Chime
Social Media and Messaging:
-
Snapchat
-
Signal
-
Reddit
Other Popular Services:
-
Duolingo
-
Canva (design platform)
-
Perplexity (AI startup)
-
Lyft
-
Ring (digital doorbell)
-
Amazon’s own services: Amazon.com, Prime Video, Alexa
-
Disney+, Hulu
-
McDonald’s app
-
United Airlines and Delta Airlines (customers unable to check in or find reservations)
Global Impact
The disruption affected millions of users, particularly those on the US East Coast during their morning commute, who found themselves unable to order coffee via mobile apps or access essential services. The outage impacted everything from banking to social media, airline reservations, and online shopping.
Financial services, gaming platforms, and businesses relying on AWS infrastructure experienced significant operational challenges. Cryptocurrency exchange Coinbase reassured users that “all funds are safe” despite access issues. Canva reported “significantly elevated error rates” affecting its functionality.
AWS Response and Recovery
AWS engineers were immediately engaged to mitigate the issue and understand its root cause. The company worked on “multiple parallel paths to accelerate recovery“.
Timeline of Recovery:
-
12:11 AM ET: Initial problem reported
-
1:26 AM ET: AWS confirmed notable error rates for DynamoDB endpoint in US-EAST-1
-
Early morning: AWS identified potential cause and began implementing fixes
-
3:03 AM PT (approximately 6:03 AM ET): Some services returned to normal
-
6:35 AM ET: Amazon addressed the DNS issue and advised companies to clear their cache (temporary storage files) to expedite service restoration
By mid-morning, AWS announced: “The underlying DNS issue has been fully mitigated, and most AWS Service operations are succeeding normally now“. However, the company noted that some requests might be throttled during full resolution, and certain services continued working through backlogs of events.
Why It Matters
This incident highlights the vulnerability of internet infrastructure and the global dependency on cloud service providers. AWS is the world’s most popular cloud services provider, generating $107-108 billion in revenue in 2024 and accounting for the majority of Amazon’s profits.
Rob Jardin, digital officer at cybersecurity firm SymVPN, explained: “The internet was initially designed to be decentralized and resilient, yet a significant part of our online ecosystem is now concentrated in a limited number of cloud regions. When one of these areas encounters an issue, the repercussions are immediate and extensive“.
Experts noted there was “no indication” the outage resulted from a cyberattack, instead appearing to be a technical malfunction. Professor Alan Woodward of the University of Surrey suggested the cause would likely be “something relatively simple like a misconfiguration in DNS”.
