An outage affecting cybersecurity firm Cloudflare took down huge swathes of the internet with it on Tuesday, once again highlighting how a handful internet services allow the entire web to stay online.
Among the websites affected by the outage are gigantic services including X-formerly-Twitter, OpenAI’s ChatGPT, and Spotify.
Even Downdetector, a website that tracks internet outages via user-submitted reports, was temporarily knocked down. (The website appears to have come back online, once again allowing users to report other websites that were taken down by the Cloudflare outage.)
Cloudflare said that it’s working hard to bring services back online and is “seeing services recover,” as Bloomberg reports.
While we await a postmortem of what actually happened, a Cloudflare spokesperson told the outlet that it saw a “spike in unusual traffic” at around 6:20 am Tuesday morning.
“We do not yet know the cause of the spike in unusual traffic,” the spokesperson told Bloomberg. “We are all hands on deck to make sure all traffic is served without errors.”
Cloudflare offers a service that acts like a buffer between websites and users. The goal is to protect web hosts from being overwhelmed by traffic, for instance, when it comes to more malicious attempts to knock out a service through the use of a distributed denial of service (DDoS) attack.
But as the latest outage demonstrates, Cloudflare has become immensely popular, and its services are being used by hundreds of thousands of companies around the world. That means that it’s now a load-bearing piece of online infrastructure, so when it runs into trouble, tons of stuff on the internet fails.
The news comes less than a month after Amazon’s Web Services suffered its own hours-long outage, similarly taking out huge chunks of the internet. In a blog post, AWS revealed that a problem with the service’s automated domain name system (DNS) management system had triggered a cascading series of events.
While we await news about the cause of the “spike in unusual traffic,” University of Surrey Centre for Cyber Security professor Alan Woodward told The Guardian that it’s unlikely to be a cyberattack, given the scale of the outage, which is unlikely to have a single point of failure.
... continue reading