Managed Postgres monitoring
You can monitor your Managed Postgres services with the following methods:
| Section | Description | Setup required |
|---|---|---|
| Dashboard | Built-in cloud console charts for resource usage and database activity | None |
| Query Insights | Per-statement telemetry: every query pattern ranked by impact, with diagnostic counters | None |
| Prometheus endpoint | Scrape metrics into Prometheus, Grafana, Datadog, or any OpenMetrics-compatible collector | API key + scraper config |
| Metrics reference | Full list of metrics exposed by the Prometheus endpoint, with types, labels, and meanings | N/A |
Quick start
Open the cloud console and navigate to the Monitoring tab of any Managed Postgres instance to see live charts for CPU, memory, IOPS, connections, transactions, cache hit ratio, and deadlocks. No configuration required.
For per-query telemetry — latency percentiles, cache vs. disk reads, temp spills, parallel worker utilization, and WAL volume — open the Query Insights tab on the same instance. To pipe host-level metrics into your own observability stack, use the Prometheus endpoint.
