Summary: The service experienced a complete outage caused by a configuration error that allowed more concurrent traffic to individual servers than one server could handle, leading to memory exhaustion.

Resolution: We temporarily restricted traffic and restarted the affected servers to resolve the incident.

Next Steps: We have adjusted the server configurations to allow a more appropriate level of concurrent traffic. This adjustment prevents server crashes by enabling us to limit traffic under excessive load more effectively, thereby improving our control over similar situations in the future.