Skip to main content

Key Features

PMP4PG provides a comprehensive set of features designed around the daily needs of PostgreSQL database administrators and engineers.


Global Dashboard

A single-pane-of-glass view across your entire PostgreSQL fleet.

  • KPI Cards — Total monitored servers, active servers, servers in error, active alerts
  • Servers Grid — All servers with their status, environment, version and agent health at a glance
  • Alert Summary — Recent and unresolved alerts surfaced directly on the dashboard
  • Environment Awareness — Distinguish RUN (production) from BUILD (development/staging) environments

{screenshot: global-dashboard-overview}


Server Detail

Drill into any monitored PostgreSQL server for a comprehensive view of its current state.

  • Real-time server metrics
  • Agent status and last heartbeat
  • Detailed connection and activity information
  • Quick navigation to ASH Viewer, AWR Viewer and Trends for that server

{screenshot: server-detail-page}


ASH Viewer — Real-Time Activity

Inspired by Oracle's Active Session History, the ASH Viewer gives you a live window into what your PostgreSQL instance is doing right now and over the recent past.

Time Windows

  • Live refresh: every 5s, 10s, 30s or 60s
  • Sliding windows: last 15 min, 30 min, 1h, 2h, 4h or 8h
  • Custom zoom on any specific interval

Analysis Panels

  • Activity Timeline — Interactive chart of active sessions over time, colored by wait event type
  • Top 10 SQL — Most resource-consuming queries ranked by execution count, total time or average time
  • Top Wait Events — Distribution of wait event types during the selected period
  • Session States — Breakdown of active, idle, idle in transaction sessions
  • Lock Contention — Identify blocking and blocked sessions

Export

  • Generate and export a full ASH report as a standalone HTML file

{screenshot: ash-viewer-timeline}


AWR Viewer — Proactive Workload Analysis

The AWR Viewer lets you generate Oracle AWR-style reports by comparing any two historical snapshots.

Report Sections

  • Load Profile — Key throughput indicators (transactions/sec, reads/writes, SQL executions)
  • Time Model Statistics — How database time is distributed across activity types
  • Instance Efficiency — Cache hit ratios and efficiency indicators
  • Top SQL — Top queries ranked by elapsed time, CPU time, executions, rows processed and I/O
  • Wait Events Statistics — Ranked wait events for the snapshot period
  • Lock Contention Analysis — Lock wait frequency and duration
  • Bloat Analysis — Table and index bloat estimation
  • Vacuum Statistics — Vacuum and autovacuum activity
  • WAL Activity — WAL generation rates and checkpoint statistics
  • Segment I/O — Per-table and per-index I/O statistics

Export

  • Full report exportable as a standalone HTML file — shareable and archivable

{screenshot: awr-viewer-report}


Visualize the long-term evolution of your PostgreSQL instances with chart-based trend analysis.

  • Database Size — Track how your databases grow over time
  • Activity & Throughput — Reads, writes, commits, rollbacks trends
  • WAL Generation — Monitor WAL volume trends for I/O and replication planning
  • Configurable periods — Last 7 days, 30 days, 90 days or custom date range
  • Organized by thematic tabs for easy navigation

{screenshot: trends-page}


Lightweight Go Agent

Each monitored PostgreSQL server runs a small, self-contained agent written in Go.

What it collects

Metric SourceFrequencyPurpose
pg_stat_activityEvery 2 secondsASH samples (real-time)
pg_stat_statementsEvery minute (delta)Query-level performance
pg_stat_databaseEvery minute (delta)Database-level throughput
pg_database_sizeConfigurableDatabase size tracking
OS MetricsEvery minuteSystem-level context
Server SnapshotEvery 30 minutesAWR snapshot base

Agent characteristics

  • Single binary, no runtime dependencies
  • Configurable via a simple YAML file
  • Self-registration with the central platform
  • Gzip-compressed payloads for minimal network footprint
  • Automatic retry with exponential backoff on network failure

Alerts

  • Centralized list of alerts across all monitored servers
  • Alert deletion / acknowledgment
  • Visible from both the Global Dashboard and the dedicated Alerts page

User Management

  • Multi-user access to the platform
  • User list administration
  • Secure login to the web interface

Next Steps