Amazon Web Services is showing “significant signs of recovery” after a major outage early Monday that impacted sites and services including Facebook, Snapchat, Coinbase and Amazon itself — reviving concerns about the internet’s heavy reliance on the cloud giant.
The company attributed the failure, which began around 12:11 a.m. PDT, to a problem in its Northern Virginia (US-EAST-1) region. In an update shortly after 2 a.m. Pacific, AWS blamed a DNS resolution issue, meaning the internet’s phone book temporarily failed to find the correct address for DynamoDB, a core database service that thousands of apps use to store and find data.
OpenAI’s ChatGPT was among the sites impacted, The Verge reported. Check-in kiosks went down at LaGuardia Airport, with long lines starting to form earlier this morning, the New York Times reported.
DownDetector showed problems for financial apps like Venmo and Robinhood, gaming services such as Roblox and Fortnite, the Signal messaging app, and productivity tools including Slack and Canva.
In an update at 3:35 a.m., Amazon confirmed that the core DNS issue was “fully mitigated,” and that most services were operating normally.
However, the company said it was still working through a backlog of requests for services like Lambda, its serverless computing platform. It warned that customers in the US-EAST-1 region would still see increased error rates when trying to launch new instances in its core cloud computing service, EC2.
US-EAST-1 is AWS’s oldest and largest cloud region, a popular nerve center for online services, which has made it an Achilles heel for the internet over the years. Major outages originating from this same region also caused widespread disruptions in 2017, 2021, and 2023.
The latest outage suggests that many sites have not adequately implemented the redundancy needed to quickly fall back to other regions or cloud providers in the event of AWS outages.
