Fix has been deployed and tested successfully.
A recently deployed internal tool generated an unexpectedly high volume of concurrent queries, which saturated available database connections and led to a service outage. We will begin rolling out mitigation actions, including Connection Pooling, and improved monitoring & alerting.