Observability for Cluster State

Monitor replication lag, connection counts, and transaction throughput to preemptively address availability issues.