https://www.bleepingcomputer.com/news/security/cloudflare-says-it-lost-55-percent-of-logs-pushed-to-customers-for-35-hours

Cloudflare, a major internet infrastructure company, experienced a significant outage in its log collection service on November 14, 2024. This outage resulted in the loss of approximately 55% of customer logs over a 3.5-hour period.

The Root Cause:

The incident was triggered by a misconfiguration in the Logfwdr system, a key component responsible for forwarding logs to downstream systems. This misconfiguration led to a cascade of failures:

  1. Blank Configuration: A bug in the configuration update caused Logfwdr to believe there were no customers configured for log forwarding, leading to the discarding of logs.
  2. Failsafe Overload: The failsafe mechanism designed to prevent data loss was overwhelmed by the sudden influx of logs, leading to its failure.
  3. Buftee Outage: The Buftee system, responsible for buffering logs, was unable to handle the increased load and shut down, further exacerbating the issue.

Impact on Customers:

The loss of logs can have significant consequences for customers who rely on these logs for security analysis, troubleshooting, and performance optimization. While Cloudflare has taken steps to mitigate future incidents, the impact of this outage highlights the importance of robust logging and monitoring systems.

Lessons Learned and Future Improvements:

Cloudflare has implemented several measures to prevent similar incidents in the future:

  • Misconfiguration Detection: A new system will monitor for anomalies in log forwarding configurations.
  • Buftee Configuration: Buftee will be configured to handle unexpected spikes in log volume.
  • Regular Overload Testing: Cloudflare will conduct regular tests to ensure the resilience of its systems.

This incident underscores the critical role that reliable logging plays in modern cybersecurity and highlights the need for robust fail-safe mechanisms to prevent data loss.