PostgresAI – enterprise-grade Postgres observability tool

Expert-level Postgres monitoring tool designed for humans and AI systems

Built for senior DBAs, SREs, and AI systems who need rapid root cause analysis and deep performance insights. This isn't a tool for beginners — it's designed for Postgres experts who need to understand complex performance issues in minutes, not hours.

PostgresAI monitoring helps catch Postgres trouble before it becomes an incident. Prioritized findings guide the next step: investigate the cause, test the proposed change on a copy of your data, and review the evidence before approving production changes.

Live demo

Experience the full monitoring solution: https://demo.postgres.ai (login: demo / password: demo)

Supported Postgres versions

PostgresAI full monitoring, express-mode checkups, and PostgresAI Console checkup analysis support PostgreSQL 14 through PostgreSQL 19.

PostgreSQL 19 is still a pre-release (beta). Use it for compatibility testing rather than production workloads until PostgreSQL 19 reaches general availability. PostgresAI preserves the beta version label in checkup reports, selects PostgreSQL 19-compatible metric SQL, and does not report a PostgreSQL 19 beta server as being behind the latest stable major release.

On PostgreSQL 19, the full monitoring collector uses the native pg_get_multixact_stats() function instead of scanning pg_multixact files. The monitoring role needs pg_read_all_stats privileges (included by pg_monitor) for the function to return values; otherwise the metric is reported as unavailable without interrupting other collection.

PostgresAI Console's separate PostgreSQL cluster-provisioning workflow depends on the PostgreSQL versions supported by its automation provider. Monitoring and checkup compatibility does not make a pre-release version available for provisioning automatically.

Console.Postgres.ai integration

PostgresAI monitoring integrates with Console.Postgres.ai, enabling:

• Automated health checks (checkups) — Comprehensive database health assessments with actionable recommendations, running automatically on schedule
• PostgresAI consulting support — Consulting customers benefit from shared monitoring access, allowing the PostgresAI team to work more efficiently on performance optimization and troubleshooting

Key features

• Open source foundation – PostgresAI core components are Apache 2.0 licensed, ensuring transparency and community-driven development

• Metadata only — Only database metadata is collected (statistics, query patterns, wait events). No actual data or query parameters are accessed. See data privacy details

• Expert-focused design: Assumes deep Postgres knowledge and performance troubleshooting experience

• Universal integration – Works with any type of Postgres, including:

  • All popular cloud platforms (RDS, CloudSQL, Azure Database, Supabase, TigerData, and more)
  • Self-managed Postgres installations
  • Kubernetes deployments

• Top-down troubleshooting methodology: Follows the Four Golden Signals approach (Latency, Traffic, Errors, Saturation)

• Powerful dashboards featuring:

  • Emergency dashboard – "Shallow but wide" analysis of all database components and key metrics to identify problematic areas within 1 minute (ideal for incident troubleshooting)
  • Query analysis – Top-to-bottom examination with comprehensive metrics for complete visibility into performance patterns and bottlenecks
  • Wait event analysis – Similar to AWS RDS Performance Insights and CloudSQL Query Insights, providing deep visibility into database wait events

• Renowned PostgresAI health check reports – Comprehensive health assessments with actionable recommendations based on industry best practices and real-world experience

• AI-powered insights backed by human expertise – From diagnostics to mitigation strategies, combining artificial intelligence with seasoned Postgres expert knowledge for actionable recommendations

Documentation

• Getting started — Installation guides for CLI, Docker, Helm, and cloud platforms
• Dashboards — Complete reference for all 14 Grafana dashboards
• Metrics reference — Detailed metrics documentation
• Configuration — Customization and alerting setup
• Troubleshooting — Common issues and solutions
• Advanced topics — Multi-cluster, custom metrics, API integration

Data privacy: metadata only

PostgresAI monitoring collects only database metadata — no actual data or query parameters are ever accessed or stored.

What is collected

• Database statistics from system views (pg_stat_*)
• Normalized query texts from pg_stat_statements (with parameter values replaced by $1, $2, etc.)
• Wait event information from pg_stat_activity
• Table and index statistics (sizes, access patterns, bloat estimates)

What is NOT collected

• Actual table data
• Query parameter values
• Connection credentials
• Application data

Verify collected metrics

Review exactly what metrics are collected by examining the metric definitions:

• Prometheus sink metrics: metrics.yml (pgwatch-prometheus)
• PostgreSQL sink metrics (including normalized queries): metrics.yml (pgwatch-postgres)

Verify database permissions

The monitoring user has read-only access to metadata only. To review the exact SQL statements used to create the monitoring role:

npx postgresai@latest prepare-db --print-sql

This shows all GRANT statements and confirms the minimal, read-only nature of the permissions.

Get started with Console.Postgres.ai

The easiest way to set up PostgresAI monitoring is through Console.Postgres.ai:

1. Navigate to DB health → Monitoring instances in the left menu
2. Click Choose plan
3. Select the Scale plan (or contact us about Enterprise)

See pricing for plan details and features.