Cloudflare incident on February 6, 2025

Cloudflare experienced a 59-minute outage of its R2 object storage on 2025-02-06 (08:14–09:13 UTC) after an abuse-remediation action mistakenly disabled the R2 Gateway service, causing 100% failures for R2 operations and disruptions across dependent services (Stream, Images, Cache Reserve, Vectorize, Log Delivery, Durable Objects, Cache Purge, Key Transparency Auditor, Workers & Pages); no data was lost, and a brief secondary impact occurred as clients reconnected. Root cause was human error and insufficient validation safeguards in internal tooling; remediation includes immediate guardrails in the Admin API, disabling product-disablement from the abuse UI, restricted access with two-party approvals, improved internal account provisioning/tagging, expanded protections against internal targets, and migration to the new Organizations model.