Amazon Web Services (AWS), the cloud infrastructure used by thousands of companies worldwide, suffered a major outage today, Monday, October 20, affecting millions of websites and applications, including giants like Amazon.com, Prime Video, Mercado Libre, Perplexity AI, Canva, and others.
The outage is impacting users in different regions across the globe, particularly in the United States and Europe.
According to AWS’s official status page, Amazon is already aware of the disruption and has acknowledged that several of its services are experiencing issues. While there is still no official cause, all signs point to the origin being in the US-East-1 region, one of the most heavily used zones by major tech platforms.
Among the hardest hit are Mercado Libre, Snapchat, Fortnite, Roblox, Duolingo, and Zoom, according to user reports and data collected by Downdetector, the site that monitors the real-time status of thousands of services.
The main cause of this global outage is believed to be a failure in AWS data centers located in the United States, particularly in the US-East-1 region. According to Downdetector, 65% of the outage reports originated from that region, and issues persisted as of 8:34 a.m. (Central Mexico time), despite Amazon stating that it was actively working to resolve the incident.
The impact was widespread and took multiple apps, websites, and digital platforms offline for several hours. These services rely on AWS to operate. Among the most affected were:
Mercado Libre
Mercado Pago
Snapchat
Fortnite
Roblox
Duolingo
Zoom
League of Legends
Brawl Stars
Airbnb
Amazon.com
Prime Video
Users reported issues such as blank screens, login failures, inability to load content or complete transactions, and even services going completely offline.
In the case of Mexico and Latin America, one of the most noticeable outages was that of Mercado Libre, the region’s largest e-commerce platform, along with its payment service, Mercado Pago. For several hours, users were unable to access their accounts or complete purchases, leading to a wave of reports on social media.
Screenshots and posts on platforms like X (formerly Twitter), Facebook, and TikTok showed connection errors, “service unavailable” messages, and frustration from users who were unsuccessfully trying to use these services.
The AI startup Perplexity AI, known for its conversational search engine, also confirmed that its system went offline due to the AWS incident. Its CEO, Aravind Srinivas, shared on X:
"The issue is due to an AWS outage. We’re working to resolve it."
Despite recovery efforts, many services are still not operating normally. Canvas, the educational platform widely used by schools and universities in the United States, continues to display error messages when attempting to log in.
This is just one of the many services affected by the outage, which has so far taken platforms like Mercado Libre, Fortnite, Zoom, Reddit, Roblox, and several others offline or left them with performance issues.
Downdetector played a key role in illustrating the extent of the outage. From early on, the site gathered thousands of reports showing real-time issues with some of the world’s most widely used apps and websites.
Among the platforms with the highest number of reported issues were:
Amazon and Prime Video
Mercado Libre and Mercado Pago
Snapchat
Roblox
League of Legends
Duolingo
Fortnite
The concentration of these services on a single infrastructure highlights just how dependent today’s digital ecosystem is on cloud providers like Amazon.
Although Amazon Web Services (AWS) claims to have mitigated the initial issue related to its DNS system, it is now facing new complications with its network load balancers, which continue to cause widespread service disruptions for many companies relying on its cloud infrastructure.
In an update posted on its official status page, AWS stated:
"We have taken additional steps to help recover the internal subsystem that monitors the health of our network load balancers. We are now seeing signs of improvement in connectivity and in APIs for several AWS services."
They also noted they are working to resolve another major limitation:
"We identified an issue affecting the launch of new EC2 instances and are already taking action to mitigate this behavior. We will provide another update before 10:00 a.m. PDT."
This is a developing story...