Ookla, using Downdetector data from 2025, analyzed millions of user-submitted reports to identify the largest website and service outages of the year across global and regional markets. The analysis shows that while digital platforms remained essential in 2025, they also proved fragile—particularly when failures occurred at the infrastructure level.
Outages across cloud services, gaming networks, streaming platforms, and communication apps demonstrated how a single point of failure can cascade across multiple dependent services simultaneously.
Downdetector by Ookla
Downdetector, operated by Ookla, tracks real-time service disruptions across thousands of digital platforms worldwide using consumer-submitted reports. Its enterprise offering, Downdetector Explorer, enables continuous monitoring, near real-time alerting, and correlation of outages across upstream and downstream service dependencies.
The World’s Largest Global Outages of 2025
The most severe outages of 2025 were driven by cloud infrastructure failures and large platform-level disruptions, impacting millions of users worldwide.
AWS — October 20, 2025
The largest outage of the year resulted from a failure at Amazon Web Services, generating over 17 million Downdetector reports across Amazon services and dependent platforms.
The outage lasted more than 15 hours and was traced to an automated DNS management issue tied to DynamoDB in the US-EAST-1 region. The failure disrupted multiple services, including Snapchat, Netflix, and several e-commerce platforms.
PlayStation Network — February 7, 2025
The second-largest global outage affected PlayStation Network, generating over 3.9 million reports. The disruption lasted more than 24 hours, preventing access to online gaming services, including titles such as Call of Duty and Fortnite. Downdetector attribution analysis indicated the issue originated internally within PSN, with no major cloud or ISP involvement.
Cloudflare — November 18, 2025
A global outage at Cloudflare generated over 3.3 million reports. The disruption lasted nearly five hours and impacted websites, applications, and APIs that rely on Cloudflare’s infrastructure, highlighting the global dependence on centralized cloud services.
Largest Outages by Region
United States and Canada
- PlayStation Network (Feb 7, 2025): 1.6 million reports
- YouTube (Oct 15, 2025): 1.5 million reports
- AWS (Oct 20, 2025): 1.2 million reports
- Snapchat (Oct 20, 2025): 944,675 reports
- Starlink (Jul 24, 2025): 583,989 reports
- Verizon (Aug 30, 2025): 515,923 reports
Europe (EU)
- PlayStation Network (Feb 7, 2025): 1.7 million reports
- Snapchat (Oct 20, 2025): 989,559 reports
- Vodafone (Oct 13, 2025): 833,211 reports during a UK-wide broadband, 4G, and 5G outage linked to a non-malicious vendor software issue
- WhatsApp (Feb 28, 2025): 621,763 reports
- Spotify (Apr 16, 2025): 468,334 reports
- Odido (June 15, 2025): 357,685 reports
- Odido (June 25, 2025): 382,003 reports
Asia Pacific (APAC)
- X (Mar 10, 2025): 645,395 reports
- Snapchat (Oct 20, 2025): 399,108 reports
- YouTube (Oct 15, 2025): 245,087 reports
- AWS (Oct 20, 2025): 175,380 reports
- AWS (Apr 15, 2025): 106,667 reports
Latin America (LATAM)
- YouTube (Oct 15, 2025): 183,672 reports
- AWS (Oct 20, 2025): 164,011 reports
- WhatsApp (Feb 28, 2025): 87,265 reports
- WhatsApp (Apr 2025): 57,095 reports
- Banco Itaú (Oct 6, 2025): 73,745 reports
Middle East and Africa (MEA)
- du (Feb 8, 2025): 28,444 reports
- Cloudflare (Nov 18, 2025): 28,016 reports
- Snapchat (Oct 20, 2025): 26,392 reports
Outlook
The 2025 outage landscape highlights the growing systemic risk associated with centralized digital infrastructure. As more platforms rely on shared cloud and network services, large-scale outages are likely to remain a recurring challenge, reinforcing the need for redundancy, monitoring, and infrastructure resilience.