This is not the first time issues with Amazon’s key services have caused widespread disruptions.
WASHINGTON — Amazon Web Services suffered a major outage early Monday morning, cutting off access to many of the internet’s most popular websites for hours.
The problem appeared to be largely resolved Monday, until reports of problems accessing websites spiked again around 10 a.m. By noon, at least some services appeared to be restored, but the outages show how the technology underpins the vast majority of the internet, and how AWS failures can lead to major problems across the globe.
What is AWS?
Amazon Web Services, commonly referred to as AWS, is a cloud computing platform owned and operated by Amazon.
In layman’s terms, AWS is a cloud host, like a much more powerful Google Drive or Squarespace, where companies can deploy their websites and programs without having to pay for expensive servers and physical infrastructure. Instead, they pay a monthly subscription fee to Amazon to use their services instead.
It offers cloud services, including computing power, storage and databases, to companies across the globe. From Netflix to Venmo to the British government, it’s likely that at least some of the websites any person goes to on a daily basis are powered by AWS.
The cloud-service operation is a huge profit center for Amazon. It holds roughly a third of the $152 billion market for cloud services, according to a report by Synergy Research — a larger share than its closest rivals, Microsoft and Google, combined.
What happened Monday morning?
The outage appears to have begun around 3 a.m. Eastern Monday morning, according to Downdetector.com, which tracks online outages.
Major sites and services such as Slack and Microsoft365 began reporting issues, all apparently caused by the outage at Amazon’s data center.
On its website, Downdetector acknowledged the widespread problem with a banner on top of the site.
“User reports indicate issues at Amazon Web Services (AWS) in the US-East-1 region,” the site said. “These problems are impacting multiple services that depend on AWS infrastructure. We’re monitoring the situation: check your local Downdetector site for the latest updates.”
But by 6 a.m., some of the issues appeared to be resolving. Amazon said in a statement that sometime after 5 a.m., it had applied “initial mitigation” and confirmed that some websites were beginning to come back online.,
“We are seeing significant signs of recovery,” the company said in a statement. “Most requests should now be succeeding. We continue to work through a backlog of queued requests.”
Amazon pinned the outage on issues related to their domain name system, an apparatus that converts web addresses into IP addresses so websites and apps can load on internet-connected devices.
For most users logging on in the morning, the situation appeared to be resolved. But a few hours later, around 10 a.m., users began noticing another uptick in outages.
Sites such as Adobe and Amazon went down again for several hours while AWS underwent another round of emergency maintenance.
Hasn’t there been AWS outages before?
Yes, a few times in the past five years. Many popular internet services were down after a brief outage in 2023. The longest major outage was in December 2021, when companies — everything from airline reservations and auto dealerships to payment apps and video streaming services — were affected for more than five hours. Before that, there was one in November 2020.
Most of these outages appear to have affected US-EAST-1, the same data center that this new outage affected. Based on the east coast of the United States, US-EAST-1 is one of Amazon’s largest data centers and one of the hearts of the AWS network.
Some cybersecurity experts have warned for years about the potentially ugly consequences of allowing a handful of big tech companies to dominate key internet operations.
“So much of the world now relies on these three or four big (cloud) compute companies who provide the underlying infrastructure that when there’s an issue like this, it can be really impactful across a broad range, a broad spectrum” of online services, said Patrick Burgess, a cybersecurity expert at U.K.-based BCS, The Chartered Institute for IT.
AWS powers some 76 million websites, according to analytics tracker Builtwith. While Amazon doesn’t give hard numbers for how much of the internet is powered by its cloud service, every outage in the past five years has brought down a number of high profile sites.
In Monday’s outage, Amazon itself, as well as Venmo, Hulu, Snapchat, Ring, Roblox, Slack, Fortnite and Microsoft365 were some of the major websites and web services affected, as some outlets reported that “half the internet” was offline.
The Associated Press contributed to this report.
