Access to popular websites and services worldwide, such as social media platform X, OpenAI’s ChatGPT, and online gaming services, was affected Tuesday morning by a global outage at internet infrastructure provider Cloudflare. At 11:48 UTC, the San Francisco-based business recognized the problem and said it was looking into issues that “potentially impacts multiple customers.”
Thousands of websites that relied on Cloudflare’s services went offline at the same time, highlighting the internet’s significant reliance on centralized infrastructure providers. Cloudflare reported “widespread 500 errors” affecting its dashboard and API, while users trying to visit the impacted sites saw “internal server error on Cloudflare’s network” notifications.
Impact Worldwide Across Key Platforms
By midday, DownDetector had over 10,000 user complaints across several areas, with 61% of them pertaining to problems using mobile applications, 28% to accessing websites, and 11% to problems with server connections. Users in the US, UK, and India were impacted by the outage; the first reports of issues were made around 6:08 a.m. ET.
The impacted services went beyond social media and included Donald Trump’s Truth Social platform, League of Legends, Spotify, and Letterboxd. Ironically, because DownDetector depends on Cloudflare’s security infrastructure, it had issues. DownDetector is the main service consumers utilize to check outage reports.
Issues With Infrastructure Dependency
With an average of 78 million HTTP requests every second, Cloudflare serves millions of websites globally with essential internet infrastructure services. The company offers technologies to keep websites up and running during moments of high traffic as well as protection against cyberattacks.
According to one source, Cloudflare said it was “continuing to investigate” the situation by 12:03 UTC and attempting to “understand the full impact and mitigate this problem.” Later in the day, services started to recover, but users were still seeing “higher-than-normal error rates.”
The event happened soon after last month’s significant Amazon Web Services outage that impacted several sites. Because single points of failure can spread to thousands of seemingly unrelated websites and services, these interruptions have rekindled worries about the internet’s reliance on a limited number of infrastructure providers.