List of main features¶
- Non-invasive setup on PostgreSQL side - no extensions nor superuser rights are required for the base functionality so that even unprivileged users like developers can get a good overview of database activities without any hassle
- Lots of preset metric configurations covering all performance critical PostgreSQL internal Statistics Collector data
- Intuitive metrics presentation using a set of predefined dashboards for the very popular Grafana dashboarding engine with optional alerting support
- Easy extensibility of metrics which are defined in pure SQL, thus they could also be from the business domain
- Many metric data storage options - PostgreSQL, PostgreSQL with the compression enabled TimescaleDB extension, InfluxDB, Graphite or Prometheus scraping
- Multiple deployment options - PostgreSQL configuration DB, YAML or ENV configuration, supporting both “push” and “pull” models
- Possible to monitoring all, single or a subset (list or regex) of databases of a PostgreSQL instance
- Global or per DB configuration of metrics and metric fetching intervals and optionally also times / days
- Kubernetes/OpenShift ready with sample templates and a Helm chart
- PgBouncer, Pgpool2, AWS RDS and Patroni support with automatic member discovery
- Internal health-check API (port 8081 by default) to monitor metrics gathering status / progress remotely
- Built-in security with SSL connections support for all components and passwords encryption for connect strings
- Very low resource requirements for the collector even when monitoring hundreds of instances
- Capabilities to go beyond PostgreSQL metrics gathering with built-in log parsing for error detection and OS level metrics collection via PL/Python “helper” stored procedures
- A Ping mode to test connectivity to all databases under monitoring