Recovered and resolved

The Knowledge Bots API experienced a service disruption beginning at approximately 06:38 UTC. The root cause was an internal cache layer becoming saturated under load, which caused request timeouts across the API backend. As a result, the service was unable to process incoming requests and health checks failed.
Our engineering team identified the issue and performed a service restart to restore connectivity. We are deploying a long-term fix that increases cache capacity and throughput to prevent recurrence.

Impact:
- Users were unable to interact with Knowledge Bots (chat, search, file upload) during the affected window
- No data was lost - all conversations, files, and knowledge bases remain intact

Main API ist ausgefallen.