Cloudflare Outage History

Cloudflare reported Stream Upload Delays that prevented or slowed video uploads to Cloudflare Stream, impacting users attempting to publish videos. The incident began at 3:35 PM UTC and was resolved by 7:35 PM UTC, lasting about 4 hours. Cloudflare issued two official updates during the event, first noting an investigation into the delays and later confirming the service had returned to normal.

Cloudflare reported an increased level of errors for customers running Cloudflare Workers scripts, leading to degraded performance for apps relying on Workers. The issue affected Cloudflare Workers within Cloudflare Sites and Services. The outage began on 8:00 AM UTC and was resolved by 9:55 AM UTC, lasting about 1 hour and 55 minutes.

Cloudflare reported network performance issues impacting traffic through its Istanbul edge, causing degraded connectivity for users in Istanbul. The issue was localized to the Europe / Istanbul, Turkey (IST) region and did not indicate a global disruption.

The incident began at 10:10 AM UTC and was resolved by 11:15 AM UTC, lasting about 1 hour and 5 minutes. The provider issued 5 status updates during the incident, moving from investigation to monitoring and eventual recovery as the Istanbul region returned to normal operation.

Cloudflare experienced a period of degraded activity affecting Cloudflare Registrar services for customers using .space domains. Affected users encountered issues in domain registration workflows and related features, with some traffic triggering increased security checks after a service module upgrade and intermittent Interactive Challenge failures. The incident began on Sep 24, 2025 2:40 PM UTC and was resolved by Sep 25, 2025 2:35 AM UTC, lasting about 11 hours and 55 minutes. The provider issued four official status updates during the incident as investigation and remediation progressed.

Cloudflare reported a brief issue affecting IP Lists updates via the Cloudflare Dashboard and API. Users attempting to modify IP Lists may have seen rate limit errors or failed requests. The incident began at 6:05 AM UTC and was resolved by 6:35 AM UTC, lasting about 30 minutes. The impact appeared limited to IP Lists management, with no confirmed disruption to other Cloudflare services.

Cloudflare issued two official status updates during the incident to communicate investigation progress and remediation. The updates indicated an ongoing investigation to resolve the rate limit behavior, followed by confirmation that the issue was resolved.

Cloudflare reported a partial outage affecting the Cloudflare Dashboard and the related communication preferences API. Users attempting to manage dashboard settings or call the API for communications might have experienced failed requests or error messages. The Cloudflare CDN edge and other security features were not affected. The incident began at Sep 20, 2025 3:33 PM UTC and the status was raised to a degraded state at Sep 20, 2025 3:35 PM UTC; the issue was resolved by Sep 20, 2025 6:25 PM UTC, lasting about 2 hours and 52 minutes. Cloudflare issued two status updates during the incident.

Cloudflare responded by investigating the Dashboard and communication preferences API, providing updates through the status page and confirming the scope of impact. The CDN edge and security features remained operational throughout the incident.

An incident started at 8:15 PM UTC affecting multiple Cloudflare components. Cloudflare Pages users attempting to create new Pages projects from GitLab repositories could see missing repos, though existing Pages sites were not affected. Separately, users reported issues with bindings in the Workers dashboard, and Cloudflare’s Workers AI models showed degraded availability. The impact spanned four components within Cloudflare Sites and Services, causing workflow friction for developers and operators across Pages and Workers. The incident lasted for several hours and was characterized by degraded functionality rather than a complete service outage.

Cloudflare WARP users and Cloudflare Zero Trust customers experienced degraded connectivity for about 50 minutes. The incident began at Sep 19, 2025 11:05 AM UTC when Cloudflare flagged an investigation into WARP connectivity problems and noted that users may experience a degraded Internet experience. The issue was resolved by Sep 19, 2025 11:55 AM UTC when the status returned to up. Cloudflare issued 2 status updates during the incident as they worked to restore normal service.

Cloudflare experienced regional issues affecting Workers AI for the bge-m3 model, the Gateway HTTP proxy in the EZE region (Buenos Aires), and Bogotá's network performance. The problems led to degraded service for LATAM users and intermittent connectivity, with some users reporting login problems. The incident began at Sep 18, 2025 7:55 PM UTC and progressed through multiple updates over the next several hours.

Cloudflare logged six official status updates between Sep 18, 2025 7:55 PM UTC and Sep 18, 2025 11:50 PM UTC. Bogotá issues were identified and a fix was being implemented by Sep 18, 2025 9:20 PM UTC, and the overall incident was resolved by Sep 18, 2025 11:50 PM UTC. The LATAM-focused impact affected multiple components (Workers AI, Gateway, and Bogotá infrastructure), with user reports rising mid-incident as conditions degraded before stabilization.

The incident began at Sep 18, 2025 2:55 PM UTC when Cloudflare reported issues with the Dashboard and Cloudflare API, causing sporadic slowdowns and limited access to management tools. These problems did not affect the delivery of cached content by Cloudflare's CDN. The issue expanded with an elevated rate of HTTP 5xx errors in the ORD (Chicago) region, prompting investigation and a series of updates through the afternoon. Health Checks briefly degraded, and the CDN/Cache component entered a degraded state before recovering. The CDN/Cache service was fully restored by Sep 18, 2025 6:15 PM UTC, with other components returning to normal as the incident was mitigated. The provider issued five status updates during the incident, and the outage lasted about 3 hours and 20 minutes.