CyberHappenings logo
☰
Home Browse About

Track cybersecurity events as they unfold. Sourced timelines. Filter, sort, and browse. Fast, privacy‑respecting. No invasive ads, no tracking.

Cloudflare's worst outage in 6 years caused by database issues

First reported
Last updated
3 unique sources, 4 articles

Summary

Hide ▲

Cloudflare experienced its worst outage in 6 years on November 19, 2025, caused by a change to database access controls that triggered a cascading failure across its Global Network. The outage lasted for nearly 6 hours, affecting core CDN and security services, Turnstile, Workers KV, dashboard access, email security, and access authentication. The issue was not caused by a cyberattack but by a database permissions change that generated an oversized configuration file, exceeding the system's hardcoded limit and causing widespread service disruptions. Cloudflare's global network, which spans over 120 countries and connects to over 13,000 networks, provides content delivery, security, and performance optimization services. The outage led to widespread 5xx errors and failures in the Cloudflare Dashboard and API. As of November 19, 2025, Cloudflare reported signs of recovery but warned of higher-than-normal error rates during remediation efforts. During the outage, some Cloudflare customers managed to pivot their domains away from Cloudflare, potentially exposing their infrastructure to increased malicious traffic. Security experts suggest that organizations should review their web application firewall (WAF) logs during the outage to identify any malicious activity that may have slipped through. The outage served as an impromptu network penetration test for organizations relying on Cloudflare for security, highlighting potential vulnerabilities in their own defenses.

Timeline

  1. 19.11.2025 12:54 3 articles · 1d ago

    Cloudflare's worst outage in 6 years caused by database issues

    On November 19, 2025, Cloudflare experienced its worst outage in 6 years, caused by a change to database access controls. The outage began at 11:20 UTC and lasted until 17:06 UTC, affecting core CDN and security services, Turnstile, Workers KV, dashboard access, email security, and access authentication. The issue was triggered by a change to one of Cloudflare's database systems' permissions, causing the database to output multiple entries into a feature file used by the Bot Management system. The oversized configuration file exceeded the system's hardcoded 200-feature limit, causing the software to crash while routing traffic across Cloudflare's network. The misconfigured file was generated every five minutes by a query running on a ClickHouse database cluster. Cloudflare resolved the issue by stopping the generation and propagation of the bad feature file and manually inserting a known good file into its distribution queue. During the outage, some Cloudflare customers managed to pivot their domains away from Cloudflare, potentially exposing their infrastructure to increased malicious traffic. Security experts suggest that organizations should review their web application firewall (WAF) logs during the outage to identify any malicious activity that may have slipped through.

    Show sources
    Open in new tab
  2. 18.11.2025 14:24 2 articles · 2d ago

    Cloudflare Global Network Outage Affects Multiple Services

    Cloudflare acknowledged an ongoing outage impacting multiple customers with widespread 500 errors and failures in the Cloudflare Dashboard and API. The outage affects Cloudflare nodes across Europe, including cities like Bucharest, Zurich, Warsaw, and Amsterdam. As of November 18, 07:29 EST, Cloudflare reported signs of recovery but warned of higher-than-normal error rates during remediation efforts. The outage was caused by a change to database access controls, leading to a cascading failure across the Global Network.

    Show sources
    Open in new tab

Information Snippets

Similar Happenings

Microsoft DNS outage impacts Azure and Microsoft 365 services

Microsoft is currently experiencing a DNS outage affecting Azure and Microsoft 365 services worldwide. The incident began around 16:00 UTC and is causing server and website connection issues for tens of thousands of users. Affected services include Intune, Azure portals, Exchange admin center, and Azure Front Door CDN. The outage is preventing users from logging into company networks and accessing online business platforms, impacting various sectors including healthcare. Microsoft has acknowledged the issue and is actively working to restore services. Users may experience intermittent request failures or latency while trying to access some Microsoft services.

Open in new tab

Azure Front Door CDN Outage Affecting Microsoft 365 Services

Microsoft is addressing an outage impacting Azure Front Door CDN, which is disrupting access to Microsoft 365 services and admin portals. The incident began around 07:40 UTC, affecting users in Europe, Africa, and the Middle East. Microsoft has restored approximately 98% of the service, with ongoing efforts to fully resolve the issue. The outage has caused delays and timeouts for users attempting to access Azure and Entra portals. Some users may experience intermittent issues with Microsoft 365 services and cloud PCs. Microsoft is actively monitoring telemetry and initiating failovers to accelerate recovery.

Open in new tab

Cloud Service Outages Impact DevOps and Development Workflows

Cloud service disruptions are increasingly affecting DevOps and development workflows, highlighting the need for resilience in software development. Recent outages in services like Anthropic's Claude.ai and GitHub have underscored the vulnerability of cloud-based tools. These incidents, while mostly minor, have led to calls for better preparedness and redundancy in development practices. Major incidents, such as the Shai-Hulud worm's impact on the npm ecosystem, have shown the potential for significant disruptions. Experts warn that a major outage could cause hours to days of downtime, affecting multiple teams and projects. Development platforms like GitHub, Azure DevOps, and GitLab have experienced various incidents, with some lasting several hours. These outages have led to discussions on improving resilience and preparedness in development workflows.

Open in new tab

Microsoft Exchange Online outage affecting global users

Microsoft confirmed an outage affecting Exchange Online users worldwide, causing access issues to emails and calendars. The incident began early in the morning and persisted for over 16 hours. The root cause involved a specific software build version that led to repeated database dismounts and failovers, resulting in increased CPU utilization and message queue buildup. The outage was mitigated after implementing a configuration change and restoring degraded infrastructure. The issue was initially believed to be region-specific but later confirmed to affect users globally. Microsoft identified a recurring issue with a specific software build version causing repeated database dismounts and failovers, leading to increased CPU utilization and message queue buildup. The company has since mitigated the outage and is monitoring the environment to ensure stability.

Open in new tab