Tech Beetle briefing GB

Cloudflare Apologises After Latest Outage Takes Down LinkedIn and Zoom

Essential brief

Key facts

Cloudflare experienced a 30-minute outage affecting major sites like LinkedIn and Zoom, caused by firewall adjustments rather than an attack.

This is Cloudflare's second significant outage in under a month, following a larger disruption in November due to a configuration file issue.

The outages highlight concerns about the centralization and vulnerability of key internet infrastructure providers.

Cloudflare serves about 20% of all websites and has a vast global customer base, making its reliability critical to internet stability.

Experts suggest outages, while disruptive, also reveal the scale of Cloudflare's influence and may prompt improvements in resilience.

Highlights

Cloudflare experienced a 30-minute outage affecting major sites like LinkedIn and Zoom, caused by firewall adjustments rather than an attack.

This is Cloudflare's second significant outage in under a month, following a larger disruption in November due to a configuration file issue.

The outages highlight concerns about the centralization and vulnerability of key internet infrastructure providers.

Cloudflare serves about 20% of all websites and has a vast global customer base, making its reliability critical to internet stability.

Cloudflare, a leading web infrastructure and cybersecurity provider, experienced a significant outage on a recent Friday morning that disrupted popular websites such as LinkedIn, Zoom, Canva, Shopify, and Downdetector.

The outage lasted approximately 30 minutes and affected 28% of Cloudflare's traffic before being resolved shortly after 9am GMT.

Cloudflare clarified that the incident was not the result of a cyberattack but stemmed from adjustments made to its firewall to protect customers against a newly disclosed software vulnerability.

This event marks Cloudflare's second major outage within a month, following a larger disruption in mid-November that impacted high-profile services including X, OpenAI, Spotify, and multiplayer games like League of Legends.

That prior outage was caused by a configuration file exceeding its expected size, which triggered a crash in the system managing threat traffic.

Cloudflare has since engaged directly with hundreds of customers and outlined plans to prevent similar widespread impacts from single updates.

The recent outage has reignited discussions about the risks of centralization in internet infrastructure.

Experts like Steven Murdoch, a computer science professor at University College London, note that while Cloudflare markets itself on reliability and resilience, repeated outages may prompt customers to reassess their dependence on the company.

Michał Woźniak, an internet infrastructure expert, highlighted that this is the fourth major global internet outage since late October, underscoring the fragility of large tech platforms.

Cloudflare serves nearly 300,000 customers across 125 countries and claims to protect about 20% of all websites, blocking billions of cyberattacks daily and generating over $500 million in quarterly revenue.

Despite the negative impact of outages, some experts suggest there may be a silver lining; outages reveal the vast scale of Cloudflare's user base and the critical role it plays in internet operations.

Cloudflare has pledged to release more detailed information next week on measures to prevent future failures, emphasizing its commitment to improving system stability and customer trust.