A manual update on a feature switch caused a partial outage to our application. After investigating and identifying the root cause, we resolved the incident by 18:48 UTC. We are taking measures to prevent such an issue from occurring in the future. We sincerely apologize for the inconvenience.