Between 15:46–15:55 UTC, some client connection attempts to databases in us-east-1 timed out due to unexpected high load on a server. The node was recovered at 15:50 UTC, and the updated DNS record propagated by 15:55 UTC. Services are operating normally.