Cloudflare’s recent outage was caused by a failure in its Workers KV storage infrastructure, which was itself due to a third-party cloud provider outage. The incident affected multiple services, including Google Cloud Platform, with no data loss reported. #Cloudflare #WorkersKV
Keypoints
- The outage lasted nearly 2.5 hours and was due to a third-party cloud provider failure.
- Workers KV, a critical part of Cloudflare’s services, experienced over 90% failure rate during the incident.
- Many services, such as Cloudflare’s dashboard, WARP, Gateway, and streaming, faced major disruptions.
- Cloudflare plans to migrate Workers KV storage to its own R2 object storage to increase resilience.
- The company is also developing safeguards and tooling to better handle future storage outages.