Amazon.com Inc.’s cloud-computing arm is working to resolve an outage that disrupted a swath of websites and services, including the New York transit agency’s status updates.
Amazon Web Services said it began investigating “increased error rates and latencies” in one of its data center clusters shortly after 3 p.m. New York time. The failures affected the company’s US-EAST-1 region, which is centered in northern Virginia and is Amazon’s most important data center hub.
AWS is the world’s largest seller of on-demand computing power and software services, which it delivers from a network of vast server farms. That means its outages can ripple across the internet, creating headaches for a range of companies and industries. The Verge news site, for instance, said Tuesday that the disruption left it unable to update its home page.
On Twitter, technologists reported not being able to log into the AWS console, the portal they use to manage AWS services. And the MTA took to the social media site to report that train service information on its website and MYmta app were unavailable because of the AWS outage.
AWS’s status health page reported that the failures stemmed from an issue with the company’s Lambda service, a product that spins up computing power in response to events. Dozens of other AWS services were affected, the company said.
In an update to its status page posted at 4:48 p.m. in New York, AWS said it was seeing a “recovery” of Lambda and other services.
“We are continuing to monitor closely as we work towards full recovery across all services,” the notice said. A company spokesperson said the status page had the latest information.
