2025: The year of the global cloud outage
There were 236 Cloudflare outages since May 2025 which are summarized below, including incident details, duration, and resolution information.
Cloudflare not working for you?
Cloudflare not working for you?
Cloudflare carried out scheduled maintenance in the NRT (Tokyo) region on . During the maintenance window, services in the Tokyo area were listed as degraded. The window lasted about 4 hours and 5 minutes and concluded with the status returning to up at . No wider regional impact was reported.
The Cloudflare incident involved degraded performance across multiple services, including Cloudflare Access, Cloudflare WARP, Magic Firewall, and Workers KV, with issues such as delayed configuration updates, posture reporting problems for WARP devices, elevated error rates in Workers KV, and intermittent 500 errors on Access protected domains. The disruption began around during scheduled maintenance in EWR (Newark). Cloudflare issued a series of status updates as engineers investigated and worked on mitigations.
Over the next period, the incident saw ongoing maintenance actions and cross‑component issues across regions, with multiple updates as problems persisted and then gradually recovered. The status was restored to up by .
Cloudflare experienced degraded performance as edits and creation of Load Balancers were slow to propagate to the edge network. The issue began with a scheduled maintenance window in Quito (UIO) at and expanded to propagation delays across the Load Balancing and Monitoring components. The incident persisted through the next day with a maintenance window in Rome (FCO) at , before full recovery at . The provider issued four official status updates during the incident. The outage lasted roughly 17 hours and 5 minutes.
Cloudflare carried out scheduled maintenance at its Buenos Aires data center (EZE) on 2025-07-21. The maintenance window began around and the status moved to warning at as the work proceeded. Normal operation was restored by .
This was a regional event affecting the EZE location, and Cloudflare published two official status updates during the window.
Cloudflare carried out scheduled maintenance in the FCO (Rome) region on 2025-07-21. The maintenance window caused a regional impact affecting traffic routed through the Rome edge location, with degraded performance during the window. The work began at and ran until , lasting roughly 3 hours and 15 minutes. Cloudflare issued two status updates to communicate the maintenance and its completion, and service was restored by . The impact was localized to the Rome area and did not indicate a broader outage.
Cloudflare reported a brief incident affecting the propagation of DNS record updates. The DNS Updates component entered a degraded state (warn) as Cloudflare investigated delays in propagating DNS changes. The issue could cause DNS changes to take longer to propagate, with users potentially seeing stale information during the window. The incident began at and was resolved by , lasting about 10 minutes, and the provider issued two status updates during the event.
Cloudflare reported a regional issue affecting Durable Objects in the Eastern Europe region, causing elevated error rates when accessing Durable Object resources. Users relying on these objects may have experienced errors or timeouts for regional services. The incident began with an investigation at , was listed as a degraded condition at , and was resolved by , lasting about 1 hour and 4 minutes. The outage was limited to the Eastern Europe region and did not indicate a global disruption.\n\nThe Cloudflare status page issued two updates during the incident: one noting the investigation and increased errors in the Eastern Europe region, and a later one confirming the issue was resolved and service restored at .
Cloudflare reported intermittent 500 errors when customers disable SSL Automatic mode via the API, causing failures for that toggle and, in some cases, requiring retries to succeed. The issue affected the Cloudflare Sites and Services API and did not indicate a broader disruption to other Cloudflare services. The incident began at and was resolved by , lasting about 1 hour and 30 minutes. Cloudflare issued two status updates during the event, noting an investigation and later confirming the resolution.
Cloudflare reported API delays impacting the Lists feature (URL lists, bulk redirect lists, and hostname lists), causing changes made via the dashboard or Cloudflare APIs to propagate to the edge more slowly. Cached content serving and other edge security features were not affected. The incident also saw increased error rates for Durable Objects in the Eastern Europe region, with user reports suggesting DNS-like symptoms. The issue began on , and Cloudflare identified the root cause and applied a fix by with ongoing monitoring. A Curitiba maintenance window started on and completed by , after which the Lists service returned to normal. The provider issued 5 status updates across 2 affected components (Lists and Durable Objects).
Cloudflare faced intermittent degradation across multiple regions during a scheduled Auckland maintenance window and subsequent regional issues. The incident began with Auckland maintenance around and included delayed cache purges in the DFW and San Jose regions, observed around . Later, APAC customers reported R2 storage errors with 400 responses (), followed by additional regional maintenance activity in Europe including London (around ). The issue progressed through several updates and affected multiple components, including CDN Cache Purge and R2 storage, across North America, APAC, and Europe. The service returned to normal by after 6 official status updates.
Looking for more Cloudflare outage stats?
Get real-time insights and historical data on Cloudflare outages.
