The Amazon Web Services outage knocked out web services around the world. The epicenter of the outage is Northern Virginia.
FAIRFAX, Va. — An Amazon Web Services outage is causing major disruptions around the world. The service provides remote computing services to many governments, universities and companies, including The Associated Press.
On DownDetector, a website that tracks online outages, users reported issues with Snapchat, Roblox, Fortnite, online broker Robinhood, the McDonald’s app and many other services. As we work to learn more about the cause of the outage, and a timeline for restoration, here’s what we know right now.
The nature of the outage
AWS first reported the issue at 12:11 a.m. ET, describing it as an “operational issue” affecting 14 key services, including DynamoDB, Lambda, and EC2. Engineers confirmed increased error rates and latency spikes, disrupting both infrastructure systems and customer workloads.
Brian Vaughn of Bethesda’s Technology Transition Paradigm said the problem affected services for its DNS resolution.
“If you think of DNS, (the) domain main name systems as the landlord for the internet, when your landlord stops taking rent and allowing people to function is really what happened here,” he said. “So the websites simply stopped working.”
By 2:27 a.m., AWS’s health dashboard reported “significant signs of recovery.”
Northern Virginia impact
The source of the outage is known as the US-EAST-1 region. That region is based in Northern Virginia, which has become a hub for data centers. Data centers have turned Northern Virginia into the capital of data.
“Even though it is in Northern Virginia, that US-EAST-1 region is massive,” Rich Mogull of the non-profit Cloud Security Alliance explained. “So even if you’re using a data center in Oregon, or Frankfurt, Germany, or out in Singapore, it has some things that require it to still be able to talk to US-EAST-1 (in Northern Virginia.)”
Back in June, the D.C. Convention Center was home to Amazon’s AWS Summit offering a rare look into the industry’s plans for data, energy, and AI.
“Northern Virginia is the largest cluster of data centers that we have anywhere in the world,” said Amazon VP Kevin Miller at the time.
Data center debate
For years, data centers have fueled both economic booms and community concerns in Northern Virginia. With artificial intelligence accelerating and data consumption rising, the D.C. region is at the center of a growing debate: What’s the future of data centers here?
Curry Roberts, president of the Fredericksburg Regional Alliance, says more than 20 million square feet of new data centers have already been approved across the Fredericksburg region.
Roberts said it’s more than just big buildings and blinking servers — it’s big tax money.
“There is no industry that’s gonna generate the kind of tax revenue the data center industry generates with as little demand on services,” Roberts said.
“For every $13 in taxes that data centers generate, they only demand $1 in services.”
That kind of return has made data centers a hot ticket in some Virginia communities. But in others, especially in more developed Northern Virginia suburbs, residents have raised alarms about energy demand and the rapid pace of development.
Amazon says it’s listening.
“First and foremost, we always want to be really good citizens and neighbors with our local communities,” Miller said. “We are at this point the world’s largest corporate purchaser of renewable energy… including projects in Virginia and the surrounding area.”
Ripple effect
Because many major internet services depend on the Northern Virginia region for cloud hosting, the outage rippled across sectors:
-
Consumer apps: Fortnite, Snapchat, Roblox, and Duolingo went offline.
-
Enterprise services: Canva, Airtable, and Robinhood experienced disruptions.
-
Amazon’s core operations: Prime Video, Alexa, Ring, and even Amazon.com suffered downtime, with smart devices becoming unresponsive.
-
Crypto and finance platforms: Coinbase and Venmo temporarily halted transactions.
How long will the outage last?
“We can confirm global services and features that rely on US-EAST-1 have also recovered. We continue to work towards full resolution and will provide updates as we have more information to share,” AWS said.