Incident:
Cloudflare experienced an outage that impacted the Workers KV service, a storage infrastructure service our MCP server relies on. This resulted in increased error rates and failed requests for a portion of our MCP traffic over the span of 3 hours.
Impact:
Approximately 100 users making requests to the MCP server received error responses or were unable to reach the server from 2025-11-18 11:20 UTC through 14:20 UTC.
Moving forward:
We are investing in our MCP Server to improve its resilience to KV store outages, and are investigating ways to reduce the overall impact of Cloudflare outages.
Our metric considers a weighted average of uptime experienced by users at each data center. The number of minutes of downtime shown reflects this weighted average.