Dozens of major websites and apps went down on Friday, potentially preventing millions of people from using everyday online services.
Reports of trouble began at about 6.40am ET, with outage tracker Downdetector showing problems across services such as Apple Pay, DoorDash, Reddit and Hulu.
The broad disruption was linked to connectivity issues at Amazon Web Services, better known as AWS, which said shortly after 9am ET that the problem had been fixed.
The incident hit several AWS services tied to the company’s US-West-2 region, located in Oregon.
AWS publicly acknowledged the issue at 4.40am PDT, or 7.40am ET, though service interruptions may have started before the company issued its alert.
In total, seven AWS services were affected: AWS Direct Connect, Global Accelerator, Internet Connectivity, IoT Core, Amazon API Gateway, Elastic Compute Cloud and Elastic Container Service.
Because AWS supplies cloud computing capacity and internet infrastructure for a vast number of websites, apps and companies, even a regional problem can ripple across many platforms that appear unrelated to users.
Amazon EC2 provides the virtual servers that keep websites and applications running, while API Gateway helps separate online systems exchange data and communicate.
Problems affecting AWS Internet Connectivity and Direct Connect can prevent businesses and users from reaching services hosted in the affected region, even when the websites’ own systems remain operational.
Issues began around 6.40am ET, affecting a range of services including Apple Pay, DoorDash, Reddit and Hulu streaming service, according to Downdetector, which monitors online outages
The widespread disruption appeared to be linked to Amazon Web Services (AWS), which confirmed it was investigating connectivity problems affecting multiple services in its US-West-2 region in Oregon
An AWS spokesperson confirmed to the Daily Mail that the company had experienced connectivity problems affecting multiple services in its Oregon region, but said the issue was resolved shortly after 9am ET.
AWS expanded the number of affected services from five to seven during the outage, adding Global Accelerator, which directs users along the fastest available routes to websites, and Elastic Container Service, which runs applications across Amazon’s servers.
The additions suggested that the failure was affecting multiple layers of AWS’s Oregon network rather than a single product.
AWS labeled the incident ‘impacted,’ confirming that customers were experiencing problems, although there was no indication that data had been lost or that Amazon’s entire global network had failed.
Companies using AWS posted apologies on X while their websites were unavailable, with one stating: ‘There is a massive AWS outage in America which is causing interruptions.’
Users also shared their frustrations during the disruption, with one writing: ‘AWS is down and is literally breaking the internet. Please hang on.’
Another posted: ‘This is not a drill, AWS in us-west-2 is offline for everyone.’
PlayStation users were among those reporting problems, with one player claiming they had been abruptly kicked out of a game on Friday morning.
Dozens of popular websites crashed on Friday, possibly leaving millions without access
‘Anyone else’s PlayStation connection down just tried to start a stream and I can’t play the game,’ another X user shared.
According to Downdetector, 30 percent of users reported problems with AWS, citing issues involving Amazon EC2, the service that provides virtual servers for websites and applications.
Another 29 percent reported problems with AWS’s US-West-2 region in Oregon, where Amazon confirmed it was investigating connectivity failures. A further 23 percent cited issues involving US-East-1, the company’s Northern Virginia region.
