Source: Sharecast
According to Downdetector - a website which allows people to report outages - some of the world’s biggest brands and organisations were affected.
As well as Halifax, Snapchat and Ring, they included Amazon, Lloyds Banking Group, HM Revenue & Customs, Zoom, Vodafone, Life360, Fortnight and Coinbase, among others.
Check-in facilities as some airports, including New York’s La Guardia, also appeared to have been affected.
In a statement, Amazon's cloud division said: "We can confirm increased error rates and latencies for multiples AWS services in the US-East-1 region. This issue may also be affecting case creation through the AWS support centre or the support API.
"We are actively engaged and working to both mitigate the issue and understood the root cause."
US-East-1 region refers to its data centre in North Virginia.
AWS later confirmed its engineers had identified the cause of the problem. In a statement shortly after, it added that a fix had since been applied and "early signs of recovery" were being seen.
However, the system now needs to clear a substantial backlog of requests that had built up during the morning before all sites and apps will be up and running again.
As at 1400 BST, some sites, including HMRC, Lloyds and Duolingo, were coming back online. But others continued to report problems.
Companies use AWS’s cloud infrastructure to run their websites, apps and other platforms. It is one of Amazon’s most important, and profitable, divisions.
Other companies that do not use AWS were unaffected by the outage, including Google, which runs its own cloud service, Facebook-owner Meta and Elon Musk’s X.
Michael Hewson at MCH Market Insights said: "This outage raises serious questions as to the wisdom of companies outsourcing some or all of their core business infrastructure to a small subset of third-party vendors, in order to save money on hosting it themselves.
"Amazon, and all other cloud services, need to do better, and businesses that use their services need to hold them account for today’s outage as well as any future issues."
Rob Jardin, chief digital officer at NymVPN, said: "There’s no sign this AWS outage was caused by a cyberattack – it looks like a technical fault.
"Outages of this scale expose our over-reliance on centralised infrastructures. The internet was originally designed to be decentralised and resilient, yet today so much of our online ecosystem is concentrated in a small number of cloud regions.
"When one of those regions experiences a fault, the impact is immediate and widespread."
