PG Monitoring is a paid observability service for PostgreSQL performance, health, security, and capacity. It supports PostgreSQL 13–17 through web access and a hybrid deployment model, with self-hosted setup using the pg_monitor agent and a dashboard token. Query coverage includes analysis, slow queries, wait events, regressions, plan flips, profiling, wait and lock analysis, and execution plans. Health monitoring covers statistics, connections, locks, cache ratio, vacuum, bloat, indexes, replication, WAL, and capacity. Baseline-aware anomaly detection examines metrics and per-query latency. AI Copilot analyzes incidents and suggests workload-grounded actions, while advisors recommend optimizations. Weekly security audits check SSL, superusers, and permissions. Metric collection runs every minute, PostgreSQL statistics every 30 minutes, and Amazon RDS logs can be pulled from CloudWatch every five minutes. Starter costs 59.00 USD per month for up to five instances; Professional costs 179.00 USD per month for up to 25. Enterprise pricing is custom. A 14-day trial is listed.
Who it is for
PG Monitoring suits teams operating PostgreSQL databases, from small teams to scaling operations and critical infrastructure. It is useful when query behavior, database health, security, and capacity need ongoing visibility.
What is good
- Supports PostgreSQL 13–17
- Includes query profiling and execution plans
- Baseline-aware anomaly detection
- AI Copilot and optimization advisors
- 14-day trial available
What to know first
- No free plan
- Starter limited to five instances
- Enterprise pricing is custom
Verdict
PG Monitoring brings PostgreSQL query, health, anomaly, and security monitoring into a web service with a self-hosted component. The trial and tier limits help teams evaluate fit before choosing a paid plan.
PG Monitoring plans and pricing
All plansCompared on database monitoring
- Free plan
- No
- Paid from
- $59/mo
- Query profiling
- Yes
- Wait and lock analysis
- Yes
- Execution plans
- Yes
- Supported databases
- PostgreSQL 13–17
- Deployment model
- hybrid
