Amazon cloud recovering after outage disrupts websites

Amazon’s cloud services unit AWS is struggling to recover from a widespread outage that disabled thousands of websites as well as Snapchat and Reddit, some of the world’s most popular apps, and disrupted businesses around the world.

The turmoil marks the largest internet outage since last year’s CrowdStrike glitch disrupted technology systems in hospitals, banks and airports, and underscores the fragility of the world’s interconnected technologies.

After around three hours of outage, systems gradually began coming back online from 6am (9pm AEDT on Monday) and AWS said it was seeing “significant signs of recovery” for some affected services.

“Most requests should now succeed. We continue to work on queued requests,” said the latest update on the outage posted on the status page.

AWS provides on-demand computing power, data storage and other digital services to companies, governments and individuals.

Outages in its servers can cause outages on websites and platforms that rely on cloud infrastructure.

AWS competes with cloud services from Google and Microsoft.

Asked for comment on the outage, AWS referred Reuters to its status page.

Amazon did not respond to a request for comment.

Software engineer, cyber expert and Institute of Engineering and Technology Fellow Junade Ali said the problem occurred in one of the network systems AWS uses to control a database product.

“Since this issue can usually be resolved centrally… as long as there are no other issues identified, it should be possible to alleviate the issue in the coming hours,” he said.

Ookla, which owns outage tracking site Downdetector, said more than 4 million users reported problems due to the incident.

Problems on some apps and websites, including Snapchat, Roblox, streaming site Max and PayPal’s Venmo, were showing signs of easing, according to Downdetector.

There were over 4,000 reports about Downdetector on Snapchat last; this number was down from the previous peak of 22,000; Reports about Roblox dropped from a peak of over 12,600 to below 500.

But other services remained affected, with thousands of reports for social media app Reddit and financial platform Chime from Downdetector.

AI startup Perplexity, cryptocurrency exchange Coinbase, and trading app Robinhood all experienced platform outages and connected them to AWS.

Amazon’s own services, including its shopping site Prime Video and Alexa, were also affected, but most recently saw a decrease in the severity of Downdetector.

Among the game platforms affected were Clash Royale and Clash of Clans, along with Fortnite, owned by Epic Games.

Uber’s rival Lyft was also toppled in the US.

Signal’s president, Meredith Whittaker, confirmed in a post about X that the messaging app was also affected by the outage, but X’s owner, billionaire Elon Musk, said his platform remains operational.

In the UK, Lloyd Bank, Bank of Scotland and telecoms service providers Vodafone and BT were also facing problems, according to Downdetector’s UK website; UK tax, payments and customs authority as on HMRC’s website.

Experts and academics said the issue underlined how interconnected everyday digital services have become and how reliant they are now on a small number of global cloud providers, a disruption that negatively impacts business and daily life.

“The root cause of this problem is that all these big companies are relying on just one service,” said Nishanth Sastry, Research Director at the Department of Computer Science at the University of Surrey.

AWS competes with cloud services from Google and Microsoft.