Amazon.com, the main retail page of the Seattle, Wash.-based company, suffered an outage that left millions of users unable to access the site.
The retail giant's main U.S. site went offline at approximately 3 p.m. ET on Monday. The cause of the outage remains unclear, but Amazon Web Services (AWS), including EC2, remained up and working on the most part unaffected, according to its status page.
However, after we first published this post, at 3:14 p.m. ET, the AWS Management Console was experiencing elevated error rates across all regions of the world, including North America and Europe. There was also a flagged issue with a North Virginia data center that was quickly resolved.
Affected sites were restored about 25 minutes later, as was the AWS Management Console, though it was still experiencing "elevated error rates" during its recovery. Amazon said it experienced "an increased error rate for CreateTags and DeleteTags APIs in the US-EAST-1 region."
After the first few minutes of the outage, browsers attempting to access the sites would continually load. Placement text was soon added to say "some improvements" were being made. Shortly after, the text changed to say Amazon was "having trouble."
Regional sites, such as Amazon.co.uk, appeared unaffected by the U.S. website's outage. ZDNet confirmed Amazon-owned Audible.com is down, but Goodreads.com is not. Services such as Netflix, which run on AWS, were working throughout the retail site's outage, while others, such as the Amazon Cloud Player, were not.
Just a quick update: http://t.co/hpfjSVhrDI is currently down and we are fixing it as quickly as we can. Thanks for your patience!
Amazon receives approximately 77 million people visiting from the U.S. per day, and is one of the top ten websites in the world in terms of traffic. It's not clear how much revenue Amazon may lose as a result of the outage. One report estimates it to be in the region of more than $1,100 in net sales per second.