It has been experienced by a global failure that has left several websites, including the social network X, Canva and ChatGPT, out of work for four hours.
Cloudflare said at 12:48 p.m. (CET) that it was aware of the problems it was causing for its customers, which include several of the world’s largest digital companies, and that it was “investigating the incident.”
According to the specialized page, which records incidents online, the problems began to be noticed around 12:30, although the peak occurred around 1:00 p.m. The situation has been improving little by little and, around 1:15 p.m., many of the affected websites were already beginning to function normally, although later the problems returned.
“We are seeing services recover, but customers may continue to experience more errors than usual as we continue to work to resolve the situation,” the company said in a statement at 1:20 p.m. After hours of irregular operation, Cloudflare considered the incident resolved at 4:00 p.m., although more than an hour later, according to their own statements, they were still working on it.
The problems have affected several sectors equally. Generative artificial intelligence (AI) tools have failed en masse. OpenAI, the company responsible for ChatGPT, was one of the first, but Claude (from Anthropic), Perplexity and the Chinese DeepSeek have also fallen. Gemini, from Google, and Copilot, from Microsoft, are among the few that have continued to function normally.
Video games have also experienced setbacks. League of Legendsone of the most popular, was one of the first to suffer from them. It didn’t take long for him to fall too Valuing. And, later, incidents were recorded on Battlenet, the online gaming platform of the production company Blizzard.
The social network The problem has also spread to banking services, according to some users.
A pillar of the internet
Cloudflare is a specialized cloud computing company that acts as an intermediary between content providers and websites. Specifically, it has a large content distribution network (CDN), which it makes available to those who want to offer their services online. The American company provides service to millions of pages around the world, hence a failure in its systems is immediately transferred to a multitude of online sites.
The failure of Cloudflare systems, Amazon’s cloud computing subsidiary, occurs three weeks after the same thing happened with Azure, Microsoft’s cloud. The three incidents took with them a multitude of online services and show the vulnerability of the global internet infrastructure, which depends on the proper functioning of a handful of companies.
Although Cloudflare has not yet revealed the cause of the problem, there is speculation in the industry that it may be due to a failure in the updates. “Our informed speculation, given the history and complexity of these systems, is that the incident could be related to one of the numerous planned updates expected today, including deployments in key regions such as Tahiti, Los Angeles, Atlanta and Santiago de Chile,” ventures Diego Suárez, chief technology officer of the cybersecurity company Transparent Edge.
“The fall of Cloudflare follows the same pattern that we already saw with the recent service interruptions of AWS and Azure. These platforms are vast, efficient and widely used,” explains Eusebio Nieva, technical director of Check Point for Spain and Portugal. “The upside is clear: its scale keeps costs down and provides even the smallest organizations with a level of performance that would have been impossible before. The downside is just as clear: when a platform of this size fails, the impact spreads widely and quickly, and everyone feels it at once.”
