Parts of the internet stop working globally after major network goes down

metro.co.uk
Toronto, Canada - September 24, 2023: Popular social media apps on an Apple iPhone: Facebook, Instagram, YouTube, Pinterest, X (formerly Twitter), LinkedIn, Reddit, TikTok, and Threads. X is not loading properly for some users (Picture: Getty Images)

Large parts of the web are down for thousands of people – including X, formerly called Twitter – amid a problem with Cloudflare.

Issues with the platforms, which also include PayPal, ChatGPT, Letterboxd and bet365, began at around 11.20am.

The list of impacted sites includes the Scottish Parliament, Vinted, and one of the world’s most popular video games, League of Legends.

Problems with ID.me, the US government’s verification website, have also been flagged by users.

Have you been impacted? Get in touch with us by emailing josh.milton@metro.co.uk

Metro reader Marion Fenlon, who is in her 60s and lives in East Sussex, said she is struggling to pay her parking ticket and can’t buy anything online.

Why has some of the internet gone down?

The platforms’ servers are hosted on Cloudflare, which provides cybersecurity tools and helps nearly two in 10 sites to load content.

More than 11,000 people were reporting issues with Cloudflare at 11.40am, according to outage tracker Downdetector.

Cloudflare outages reported in the last 24 hours Outage reports have begun to ease (Picture: Downdetector)

Cloudflare said on its status webpage that the issue has been fixed, adding: ‘We believe the incident is now resolved.

‘We are continuing to monitor for errors to ensure all services are back to normal.’

The company’s CTO, Dane Knecht, posted a lengthy apology on X following the fix.

‘I won’t mince words: earlier today we failed our customers and the broader Internet when a problem in @Cloudflare network impacted large amounts of traffic that rely on us,’ he said.

‘The sites, businesses, and organisations that rely on Cloudflare depend on us being available and I apologise for the impact that we caused.’

He added that the team will share a breakdown of the cause of the outage ‘in a few hours’, stressing that it was not a cyber attack.

OpenAI, the start-up behind ChatGPT, earlier confirmed that the chatbot was down for some users.

When loading up the AI assistant, the screen read: ‘Please unblock challenges.cloudflare.com to proceed.’

While trying to access bet365, it showed visitors an error message: ‘Sorry, you have been blocked. You are unable to access bet365.com.’

While both the chatbot and the betting platform told users they had been ‘blocked’, this isn’t the case – it’s just Cloudflare not working.

Is the outage a privacy risk at all?

Rob Jardin, Chief Digital Officer at NymVPN, warned to Metro that the answer might be yes.

‘As Cloudflare is the intermediary between millions of users and the web, including through many VPN and crypto services, this can reveal the real IP addresses of people, creating opportunities for specific cyber attacks like a denial of service attack (DDoS),’ he said.

”Imagine you’re texting a friend, and suddenly the person managing the phone network can see who you are and what you’re doing. That’s essentially what happens.

‘Your real physical location (IP address) and the list of every website you try to visit (DNS queries) can be briefly exposed to anyone watching, including hackers or surveillance systems.’

Rotterdam, Netherlands - April, 2023 : Social media application on the smartphone screen Cloudflare is an invisible force keeping about two in 10 websites afloat (Picture: Getty Images)

Jardin added that today’s disruption is similar to the outage that shook Amazon Web Services, a giant cloud computing service.

AWS suffered a glitch during an update, causing an hours-long outage that dragged government websites, banks, games, streaming services, airlines and crypto platforms offline.

Ookla, which runs Downdetector, previously told Metro that the outage impacted at least 16 million people across 60 countries.

The AWS outage, as well as today’s with Cloudflare, show what happens when too many websites rely on one giant company’s infrastructure, software artefact management firm Cloudsmith’s CTO, Lee Skillen, told Metro.

‘Today’s Cloudflare, tomorrow might be Fastly,’ he said, referring to the US cloud platform used by 1,200 companies.

‘Although outages are not uncommon, a global “completely down and out” outage like this is absolutely highly unusual, and there is no doubt that this has a wide-reaching impact worldwide for businesses and their users.’

Georgina O’Toole, a chief analyst and partner at TechMarketView, added to Metro: ‘Hidden infrastructure, like Cloudflare, is easy to forget about when it is working, but as soon as it fails, the implications are widespread.

‘The topic of organisational resilience will once again rise to the top of boardroom agendas, with conversations turning to how best to spread risk, reduce impact and mitigate against issues that can’t be prevented.’

Get in touch with our news team by emailing us at webnews@metro.co.uk.

For more stories like this, check our news page.

News Updates

Stay on top of the headlines with daily email updates.