Mass internet meltdown as world's most popular apps and websites go offline
Dozens of websites and apps including Lloyds Bank and Snapchat were knocked offline today. Amazon Web Services later said it was showing ‘significant signs of recovery’
A major outage caused widespread internet disruption across the world and left millions of people unable to access sites including HMRC, Snapchat, Fortnite, and Duolingo.
Listen to this article
The outage also affected work apps such as Slack and Canva, National Rail, phone and broadband companies including BT, and EE and Amazon services including Amazon.com, Amazon Alexa, Ring and Amazon Prime Video.
The outage is caused by issues with Amazon Web Services (AWS) in the US which are impacting apps which depend on its infrastructure, according to service disruption platform Downdetector, which showed a spike in failures reported from 8am (midnight PST).
After roughly three hours of disruptions, systems were gradually coming back online, with AWS saying it was seeing “significant signs of recovery” for some impacted services.
Major British banks including Lloyds and Halifax were among the services also affected, as well as GOV.UK, which is needed for applying for visas, renewing passports, and managing taxes.
HMRC confirmed on Monday that some of its customers have been impacted by the outage, with a spokesperson saying: "We're aware that customers are having problems accessing our online services, as part of global issues affecting Amazon Web Services. We're working urgently with them on this matter.
"Our phonelines are currently busy as a result, so for anything that isn't urgent we recommend calling at a later time."
Billions of users are now thought to be affected, with a spike in reports of users being unable to access a host of popular sites began on Monday morning.
Shortly after 10.30am GMT, AWS confirmed they were seeing "significant signs of recovery" as they attempted to overcome issues.
The problem was linked to an issue at Amazon Web Services (AWS), with their infrastructure supporting many of today's best known internet sites and apps.
It took the company 78 minutes to identify the issue and begin rolling out a fix.
Read more: Reform UK suspends four councillors following leaked meeting recording
Amazon said there have been “increased error rates” and delays with “multiple AWS services”, in a message posted to its service status page.
"We have identified a potential root cause for error rates," Amazon said in an update at 10am GMT on Monday.
"Based on our investigation, the issue appears to be related to DNS resolution of the DynamoDB API endpoint in US-EAST-1," it said.
"We are working on multiple parallel paths to accelerate recovery," adding that the "issue also affects other AWS Services".
It added: "We will continue to provide updates as we have more information to share, or by 2:45 AM."
By midday (UK time), AWS said it had observed "recovery across most of the affected AWS Services".
Amazon Web Services is known as "cloud service" and props up a host of internet infrastructure.
Their services are relied on by much of the internet - including sites not traditionally linked to the Amazon brand.
In simple terms, AWS relies on companies hiring its computers and servers to run their apps and websites off.
The AWS Health Dashboard says there is an "operational issue" in Virginia, that means servers are running slower or seeing hiccups.
"We are actively engaged and working to both mitigate the issue and understand root cause," it said at about 8.50am.
"We will provide an update in 45 minutes, or sooner if we have additional information to share."
It added the issue lies with DynamoDB, a database that stores apps and websites, according to Amazon.
As a result, much of the internet will be impacted heading into Monday lunchtime.
It is now the most popular provider of "cloud services" in the world. It made $108 billion last year, and it now accounts for the majority of Amazon’s profits.
It comes as Amazon to take on 15,000 seasonal workers to deal with festive rush.
The outage has affected a growing list of sites, with the likes of HMRC, Roblox, Canva and My Fitness Pal hit by the outage.
According to Down Detector, online games including Clash Royale and Roblox have been impacted.
Other affected sites include: Snapchat, Ring, Roblox, Clash Royale, Life360, My Fitness Pal, Xero, Canva, Amazon, Amazon Web Services, Amazon Music, Prime Video, Clash of Clans, Fortnite, Wordle, Duolingo, Coinbase, HMRC, Playstation, Pokémon Go.
Current signage on the Duolingo indicates that the language learning app is undergoing a ‘maintenance break’.
More than 6,500 people are struggling to access Lloyds, according to Downdetector.