AWS Outage Takes Down Big Chunk Of Internet; Fortnite, Roblox, Amazon, Banks, And More Affected

By Elijah Gonzalez | October 20, 2025 | 10:30am

Video Games News Fortnite

AWS Outage Takes Down Big Chunk Of Internet; Fortnite, Roblox, Amazon, Banks, And More Affected

This morning, an Amazon Web Services (AWS) outage brought down several popular sites and apps, affecting everything from video games to airlines. According to Amazon’s AWS Health Dashboard, they started experiencing an increase in errors starting at around 3:11 a.m. ET.

Looking at the site Downdetector, which tracks downtime across various sites and services, we can see downage spikes in games like Fortnite, Roblox, Wordle, Death By Daylight, and Battlefield. The PlayStation Network appears to have had outages as well, as did Steam around the same time. Websites like Reddit and Amazon were affected, as were banks like Lloyds Bank.

At around 5:27 a.m. ET, Amazon wrote that it was seeing “significant signs of recovery” after applying a change at 5:22 a.m. As of 8:48 a.m. ET, they wrote that “new events published to these services are being delivered normally and are not experiencing elevated delivery latencies.”

For those unaware, AWS is Amazon’s cloud computing technology which is used worldwide to ostensibly lease out servers and more. AWS is the biggest cloud computing service in the world, with 29% of the global cloud market share.

As for what went wrong, if you forgive the tech jargon, Amazon wrote, “based on our investigation, the issue appears to be related to DNS resolution of the DynamoDB API endpoint in US-EAST-1.”

Most Popular

While this may make it sound like the issues were constrained to the eastern United States, this was very much a global outage. US-EAST-1 is the code for servers based in a northern Virginia datacenter, but this datacenter is used by services around the world. As for DynamoDB, this is a database that hosts information for various companies, like customer data, for instance. DNS stands for Domain Name System, and the common analogy used to explain how it works is that it’s like a phone book—essentially, it converts website domain names, which are easily human-readable, into something that sites and applications can understand, called an IP address, allowing computers to properly connect to the right place.

So basically, anything that was relying on this specific data center in northern Virginia to retrieve stored information couldn’t find the data due to the DNS issue, resulting in downages across a variety of sites and apps.

As for why this affected so many services, this speaks to AWS having the biggest share of the cloud computing picture; basically, one company powers an astounding percentage of the Internet.