As one of the internet’s most critical infrastructure providers, Cloudflare powers websites, APIs, and applications for millions of businesses around the world. From DDoS protection to performance optimization and DNS management, its global network is integral to the modern web. But like any large-scale system, Cloudflare is not immune to outages.
This article tracks a detailed history of major Cloudflare outages. We examine incidents that disrupted services across Cloudflare regions and products.
From global DNS failures to performance degradation in specific countries, we break down what happened, where it happened, how long it lasted, and what Cloudflare reported as the root cause.
Whether you’re an engineer, IT manager, or site reliability professional, this timeline offers insights into the impact of Cloudflare downtime. Also, it helps you understand the patterns behind these service disruptions.
Let’s dive into the most significant Cloudflare incidents in the past 5 years.
April 2025: Widespread Intermittent Issues Across Cloudflare Services
In April 2025, Cloudflare experienced a wave of service degradation affecting multiple core systems. This affected its Dashboard, DNS, Workers, Zero Trust services, and regional network infrastructure.
It was not a single, massive outage, but a series of intermittent failures, degraded performance, and service disruptions spanning several days, highlighting systemic instability across Cloudflare’s platform.
- Two-Factor Authentication (2FA) login errors on Dashboard
Started on April 20, 2025, at 00:30 UTC. Users attempting to log in via 2FA on the Cloudflare Dashboard encountered intermittent errors.
- DNS Root Server instability in Scotland
On April 22, 2025, at 09:53 UTC, issues with DNS Root Servers and DNS Updates affecting users in Scotland.
Cloudflare acknowledged the incident and launched an investigation. The precise scope of the regional impact was not disclosed, but root-level DNS degradation could have caused resolution delays.
- Zero Trust Dashboard failures
On April 22, 2025, at 11:34 UTC, operations on the Zero Trust & Gateway Dashboards failed intermittently, resulting in error messages and failed requests for customers.
Additional warnings and similar symptoms were reported at 13:35 UTC, indicating persistent instability.
- Cloudflare Workers build failures
On April 22, 2025, at 17:21 UTC, a build system problem disrupted Cloudflare Workers, preventing the initialization of new deployments.
The issue extended into scheduled maintenance in Bangalore (BLR).
- Browser Isolation outages
On April 22, 2025, at 23:49 UTC, users reported intermittent failures to acquire remote browsers, a key component of Cloudflare’s Browser Isolation service.
Fixes were implemented and monitoring began shortly after midnight UTC on April 23.
- Dashboard and Support Case errors
On April 23, 2025, 04:27 UTC, some users were unable to view support cases via the dashboard (ticket creation via email remained functional).
- Follow-up incident
At 17:03 UTC, users experienced 403 errors when accessing the Cloudflare Dashboard.
Cloudflare acknowledged and worked on a fix during active maintenance windows.
- Network-level errors in Taipei
On April 24, 2025, at 07:52 UTC, an increase in HTTP 500 errors was observed at the Taipei (TPE) data center. The network team began active investigation and mitigation efforts.
March 18, 2024: Performance Degradation in China
On March 18, 2024, Cloudflare reported a performance-related outage impacting multiple regions across China. The issue began at approximately 04:30 UTC, and the Cloudflare status changed from UP to WARN across numerous data centers.
More than 25 Chinese cities experienced degraded performance, including major metropolitan areas. Users in these locations likely faced elevated error rates and connection timeouts, particularly when accessing resources routed through specific Cloudflare IPs.
The downtime was mitigated by 06:20 UTC. The duration of the incident was nearly 2 hours.
Though a billing system upgrade was simultaneously in progress (running from March 15–22), there is no direct confirmation that it contributed to this incident.
February 2023: Major Multi-Service Outage
Cloudflare continued with a series of maintenance across the globe. Services experienced degraded performance during these times. However, on February 21, 2023, a widespread outage occurred, affecting multiple core services and global regions.
The incident was declared at around 06:38 UTC, with status degradations affecting services like Stream, DNS Updates, Bot Management, and SSL Certificate Provisioning.
The disruption also extended to Cloudflare’s Ethereum Gateway and impacted many data centers across India and the Middle East, and Asia.
December 2022: Series of Outages Across 25+ Regions
Cloudflare experienced a minor outage starting late on December 11, 2022. It temporarily impacted service availability across a broad range of global regions – Asia, Latin America & the Caribbean, the Middle East, North America, and Europe. The issue was marked by regions transitioning from UP to WARN status and gradually recovering within hours.
On December 12, 2022, at 04:35 UTC, Cloudflare marked a number of global regions as experiencing a WARN status, primarily affecting cities in South America, the Middle East, Africa, and Eastern Europe. The status change coincided with scheduled maintenance at Cloudflare’s OTP (Bucharest) facility.
On December 12, 2022, at 09:30 UTC, Cloudflare issued a WARN status for several regions, primarily in Brazil, with some effects extending to the Middle East and East Africa. This service degradation occurred during a period of scheduled maintenance at Cloudflare’s ORD (Chicago) data center.
The scheduled maintenance of facilities across the globe continued till the end of February 2023. They caused minor and short-lived downtime worldwide.
April 2022: Cloudflare Zaraz and Spectrum Connectivity Issues
On April 28, 2022, Cloudflare experienced multiple service disruptions affecting both Zaraz and Cloudflare Spectrum, impacting customers globally.
- Cloudflare Zaraz outage
The day began with an issue affecting Cloudflare Zaraz, a service used to manage and optimize third-party tools. The problem was identified at 09:55 UTC, prompting Cloudflare to temporarily disable Zaraz for all users while working on a fix. By 13:10 UTC, the Cloudflare engineering team implemented the resolution and moved into monitoring.
Later in the day, Cloudflare also reported DNS update delays, likely related to the earlier Zaraz incident. These issues were still being monitored after fixes were applied.
- Cloudflare Spectrum connectivity issues
Starting around 23:00 UTC, Cloudflare detected Spectrum connectivity issues across several global data centers, including:
- Lisbon, Portugal (LIS)
- Bangalore, India (BLR)
- Kuala Lumpur, Malaysia (KUL)
- Toronto, Canada (YYZ)
The Cloudflare Spectrum service, which facilitates TCP/UDP connections for non-HTTP traffic, was also directly impacted. Cloudflare quickly identified the root cause and began implementing a fix. The incident timeline ends by 23:20 UTC. The services in the affected regions were restored and returned to normal.
September 2021: Widespread Issues Affect Magic Transit, BYOIP, CDN Purging, and Cloudflare Pages
Between September 29 and 30, 2021, Cloudflare experienced a series of service disruptions that impacted several of its core products, including Magic Transit, Bring Your Own IP (BYOIP), Argo Tunnel, CDN cache purging, and Cloudflare Pages.
These issues occurred in multiple waves, affecting network performance, developer workflows, and caching functionality.
- Initial disruption with Magic Transit and BYOIP
At approximately 23:50 UTC on September 29, both Magic Transit and BYOIP entered a warning state due to network-related issues, with Argo Tunnel soon impacted as well. These products remained degraded for over an hour until services began recovering around 01:00 UTC on September 30. Cloudflare labeled this a minor service outage, but the affected services play a critical role in DDoS mitigation and custom IP routing, meaning the issue likely disrupted enterprise traffic flows and network performance.
- CDN Cache purge delays
Later, at 08:50 UTC, Cloudflare reported delays with CDN single file cache purging, an essential feature for developers and site owners who need precise control over content updates. The issue worsened briefly and was marked as a minor service outage by 09:00 UTC, with recovery following shortly after.
- Cloudflare for Teams and Pages: gateway and build failures
In the afternoon UTC hours of September 30, Cloudflare for Teams encountered Gateway issues starting around 14:40 UTC, potentially affecting zero trust access, DNS filtering, and team-level security policies.
Shortly after, Cloudflare Pages began experiencing build failures at 15:55 UTC, hampering developers’ ability to deploy static sites.
The issue persisted intermittently until 17:35 UTC, when it was downgraded to a minor service outage under maintenance.
Impact on users
- Potential routing issues and increased latency for enterprise traffic
- Delayed cache invalidation for CDN users
- Developer disruptions due to site build failures
- Access issues with secure gateway tools for teams
March 2021: Latency and TLS Failures Impact South America and Oceania
On March 29, 2021, Cloudflare experienced two regional performance issues. They affected users across South America and Oceania, resulting in increased latency and SSL/TLS handshake failures in multiple data centers.
- South America: Latency Spikes in Brazil
At approximately 04:35 UTC, Cloudflare’s São Paulo (GRU) and Rio de Janeiro (GIG) data centers entered a warning state due to network performance issues. The disruption lasted roughly 10 minutes, indicating brief but impactful connectivity degradation.
Cloudflare attributed the event to general network instability in South America, which likely resulted in increased latency and slower page loads for users in the region.
- Oceania: TLS Handshake Failures Across Australia and New Zealand
Later that day, at 09:50 UTC, multiple Cloudflare locations across Australia, New Zealand, and the South Pacific (including Sydney, Melbourne, Auckland, Perth, and Noumea) reported SSL/TLS handshake failures.
These issues indicated problems with the secure connection negotiation process, potentially preventing users from accessing secure websites or services using HTTPS.
The situation began improving gradually, with individual sites like Perth recovering by 09:56 UTC, and others such as Adelaide, Brisbane, and Melbourne returning to normal over the following hours.
Incident resolution was implemented at 13:20 UTC. All impacted sites had returned to the UP status, although some still reported lingering performance degradation.
In event lasted for 8 hours of rolling disruption across both regions.
During this time, users experienced slower load times, possible inability to load secure websites, or interruptions in encrypted connections for users in the affected areas.
April 2020: Widespread Dashboard, API, and Analytics Outage Affects Cloudflare Services
In mid-April 2020, Cloudflare experienced one of the most widespread and prolonged multi-service outages of the year. The incident impacted nearly every major internal service, including the Cloudflare Dashboard, API, Analytics, DNS updates, SSL provisioning, Argo Tunnel, and more.
The first signs of trouble appeared on April 14 at 16:01 UTC, when the Cloudflare Dashboard entered a degraded state, followed shortly by an Analytics outage at 16:46 UTC. Although a fix was implemented and monitored around 17:35 UTC, another Analytics-related issue reappeared around 18:41 UTC.
On April 15, the incident intensified. At 11:35 UTC, DNS update delays were reported, affecting both the Dashboard and API.
Despite a temporary fix being deployed, the underlying problems escalated. By 15:40 UTC, broader issues were observed across the Dashboard and API.
At 17:40 UTC, the situation peaked as multiple Cloudflare services entered full outage or warning status simultaneously, including:
- Cloudflare Dashboard
- Cloudflare API
- Cloudflare Registrar
- Cloudflare Stream
- Argo Tunnel
- Billing
- SSL Certificate Provisioning
- SSL for SaaS Provisioning
- Cloudflare Logs
- CDN Cache Purge
- DNS Updates
- Load Balancing and Monitoring
- Analytics
Cloudflare confirmed that it had restored full connectivity and network redundancy by 20:44 UTC.
Yet, Analytics and Argo Tunnels remained partially degraded into the late evening. Argo Tunnel was fully restored by 22:32 UTC, and Cloudflare Analytics returned to a minor outage state around 23:45 UTC.
While Cloudflare’s public updates pointed to Dashboard and API connectivity issues, specifics of the root cause were not disclosed. However, the broad range of affected services suggests possible internal control plane failures or a critical network infrastructure issue.
This incident was notable not just for its scale, but also for its multi-wave nature. Services temporarily recovered only to degrade again within hours.
It highlighted the interdependence of Cloudflare’s internal systems, where a failure in core components like the Dashboard and API cascaded across DNS, provisioning, billing, and analytics platforms.
January 2020: Persistent Issues Affecting Cloudflare Sites and Dashboard
Cloudflare experienced its first service degradation of 2020 on January 6, affecting its Dashboard and select regional services.
- Cloudflare Dashboard became unavailable
The initial problem involved static assets on the Cloudflare Dashboard becoming unavailable, triggering a minor service outage from 13:20 UTC until a fix was implemented and monitored around 15:59 UTC.
- Network congestion in the Frankfurt data center
Concurrently, the Frankfurt (FRA) data center in Europe reported network congestion and performance issues, contributing to degraded service in the region until resolution later that evening.
- Delays in configuration settings impacting numerous services
Over the next several days, Cloudflare saw a series of recurring incidents related to configuration settings delays, impacting the Cloudflare Dashboard, API, and Access services. These issues primarily caused delays in applying user settings, but did not affect production-level traffic or existing configurations. Nevertheless, the frequent warnings and ongoing minor outages between January 6–11 reflected a prolonged period of instability in Cloudflare’s internal management systems.
- Disruption with the Argo Tunnel
Further compounding the challenges, Argo Tunnel suffered from availability issues globally, with tunnel registration in Cloudflare’s Los Angeles data center.
- Brief outages on January 21 through January 23
The issues continued into the second half of January. On January 21, the Cloudflare Dashboard, API, and Analytics services once again experienced service degradation.
The outages were marked by brief recoveries followed by repeat failures, suggesting ongoing underlying instability. The issues were first reported at 16:25 UTC, re-emerged around 19:20 UTC, and ultimately stabilized after 20:00 UTC.
Just two days later, on January 23, DNS Root Servers were affected, resulting in another minor service outage. Cloudflare acknowledged DNS Root Service issues starting at 19:25 UTC, which were not resolved until later that night.
Conclusion
Cloudflare’s network of products and services plays a vital role in keeping the internet fast, secure, and available. However, as this outage history shows, even the most robust platforms can experience disruptions.
From localized slowdowns to full-scale global outages, there have been times when Cloudflare services became unavailable, affecting everything from DNS and CDN performance to critical Cloudflare products like Bot Management and Stream.
Staying informed during these incidents is essential, especially when your business depends on Cloudflare or other third-party providers. That’s where StatusGator comes in.
StatusGator monitors the status pages of 5,000+ services, including all major Cloudflare products, and delivers unified, real-time alerts whenever outages occur.
By tracking outages across providers and aggregating them in one place, StatusGator helps IT teams respond faster and reduce downtime impact before users even start submitting support tickets.
If you rely on Cloudflare or other cloud platforms, consider using StatusGator to stay ahead of the next global outage.
