On Tuesday, a significant disruption impacted a number of high-profile websites, including X and ChatGPT, as a result of problems with major internet infrastructure provider, Cloudflare. Many users reported difficulties accessing these sites and other online services shortly after 11:30 GMT. The outage monitoring site Downdetector became inundated with reports from concerned users experiencing issues.
According to a statement from Cloudflare, the problems began at approximately 11:20 UTC when the company observed a spike in unusual traffic directed at one of its services. This surge in traffic led to errors affecting the flow of data through its network. While the company is currently investigating the cause of this unusual traffic, it reassured users that its team is working diligently to ensure that all traffic is processed without errors.
As internet services gradually returned to normal, Cloudflare announced that it had implemented a change that successfully restored its dashboard services. However, users visiting X encountered a message indicating that there was a problem with its internal server, attributed to the errors originating from Cloudflare. Similarly, users of ChatGPT saw an error message prompting them to unblock challenges.cloudflare.com to continue using the service.
Cloudflare is a significant player in the realm of internet security, providing services that help verify that visitors to websites are human rather than automated bots. The company claims that approximately 20% of all websites globally utilize its services in some capacity. However, the extent of the impact from this recent outage on those websites remains unclear.
Interestingly, Downdetector itself experienced issues during this incident, displaying an error message as many users attempted to visit the site for updates on the ongoing outages. Alp Toker, director of NetBlocks, which monitors web service connectivity, noted that the outage highlighted a catastrophic disruption to Cloudflare's infrastructure. He remarked on the irony of the situation, stating that while many websites rely on Cloudflare to protect against denial of service attacks, this dependency has inadvertently turned the company into one of the internet's largest single points of failure.
This incident follows a previous outage affecting Amazon Web Services last month, which knocked over 1,000 sites and apps offline. Shortly after, another major web services provider, Microsoft Azure, also experienced disruptions. Jake Moore, a global cybersecurity advisor at ESET, emphasized that these recent outages have once again brought to light the fragile nature of the networks on which many businesses rely. With limited alternatives, companies often find themselves heavily dependent on Cloudflare, Microsoft, and Amazon for hosting their websites and services.