HikariCP in Production: Five Settings That Are Often Tuned Incorrectly

▤ Summary

This article explains five HikariCP settings that teams often misconfigure in production. It walks through pool sizing, minimum idle connections, max lifetime, keepalive time, leak detection, and connection timeout with PostgreSQL-specific examples. It also shows how to combine HikariCP with PgBouncer, Kubernetes probes, and cloud database idle-timeout limits to avoid connection drops and latency spikes. The article closes with monitoring advice using Grafana, Prometheus, and HikariCP metrics so teams can spot saturation, timeouts, and leaks early.

▦ Linked Companies