Amazon Web Services Restored After Widespread Outage In US

Amazon Web Services (AWS) has recovered from an outage on Tuesday that was widespread in the US, and impacted big name websites.

The outage seemed to have impacted the eastern side of the United States on Tuesday, and according to Downdetector.co.uk, reports of problems first began appearing around 8pm (BST) and lasted beyond 11pm (BST).

The AWS outage reportedly impacted multiple services and websites. Apps impacted included MacDonald’s and Taco Bell, while websites belonging to the Verge, the New York Metropolitan Transportation Authority, Southwest Airlines, the Boston Globe, and the US securities regulator’s EDGAR system were also affected.

AWS outage

However it is next to impossible to count the number of services and websites the outage actually impacted.

Other Amazon services such as Amazon Music and Alexa were also impacted, according to Downdetector.

Several hours after Downdetector first began showing reports of outages, Amazon said, “the issue has been resolved and all AWS Services are operating normally.”

The AWS Health Dashboard meanwhile had noted problems of degradation on multiple services in the US-East-1 region that started around 3PM ET (8pm BST).

Amazon blamed the outage on “a subsystem responsible for capacity management for AWS Lambda.”

AWS Lambda is a service that lets customers run computer programs without having to manage any underlying servers.

“Between 11:49 AM PDT and 3:37 PM PDT, we experienced increased error rates and latencies for multiple AWS Services in the US-EAST-1 Region. Our engineering teams were immediately engaged and began investigating,” said Amazon. “We quickly narrowed down the root cause to be an issue with a subsystem responsible for capacity management for AWS Lambda… As of 3:37 PM, the backlog was fully processed. The issue has been resolved and all AWS Services are operating normally.”

According to Reuters, the disruption appeared smaller in time and breadth than one the company suffered in 2017 of its data-hosting service known as Amazon S3, representing the bread and butter of its cloud business.

The outage appeared to extend to AWS’s own web page describing disruptions in its operations, which at one point failed to load on Tuesday, Reuters witnesses saw.

Previous outages

Amazon had its last major outage in December 2021, when disruptions to its cloud services temporarily knocked out streaming platforms Netflix and Disney+, Robinhood, and Amazon’s e-commerce website ahead of Christmas that year.

AWS outages have happened before on multiple occasions prior to that, and experts point out that these service interruptions demonstrate the need for organisations to build in redundancies into their operations.

AWS outages have taken place over the past decade – sometimes as a result of outside incidents.

For example back in May 2010, Amazon’s EC2 service suffered a power outage after one of its data centres failed to cope with a power switch-over following a car crash, which triggered a local blackout.