Get unlimited infrastructure observability context via MCP

Ground your agent engineering
in structured observability context

Get the latest observability docs and guidance for your infrastructure via MCP

PostgreSQL

PostgreSQL is a powerful, open source object-relational database system with over 35 years of active development that has earned it a strong reputation for reliability, feature robustness, and performance. Official PostgreSQL docs: PostgreSQL 18.1 Documentation

Native PostgreSQLOfficial PostgreSQL metrics docs:

PostgreSQL 18 Documentation - The Cumulative Statistics System

DatadogOfficial Datadog PostgreSQL metrics docs: GitHub

Datadog PostgreSQL Integration

OpentelemetryOfficial Opentelemetry PostgreSQL metrics docs: GitHub

OpenTelemetry PostgreSQL Receiver Documentation

PrometheusOfficial Prometheus PostgreSQL metrics docs: GitHub

Prometheus PostgreSQL Exporter

Get context for PostgreSQL metrics



active_time NCountUnits:millisecondTime spent executing SQL statements (ms)
analyze NGaugeNumber of automatic analyzes
apply_errorN GaugeNumber of errors that occurred while applying changes. Extracted from pg_stat_subscription_stats. Only available on PostgreSQL 15 or higher. This metric is tagged with subscription_name.
archived NCountarchived_count statistic from pg_stat_archiver
backend_xid NGaugeUnits:transactionbackend_xid statistic from pg_stat_activity
backend_xmin NGaugebackend_xmin statistic from pg_stat_replication
backup_streamedNGaugebackup_streamed statistic from pg_stat_progress_basebackup
bitsNGaugebits statistic from pg_stat_ssl
blk_read_time NCountUnits:millisecondread_time statistic from pg_stat_io
blk_write_time NCountUnits:nanosecondTime spent writing WAL to disk (ms)
block_distance NGaugeUnits:blockblock_distance statistic from pg_stat_recovery_prefetch
block_doneNGaugeblocks_done statistic from pg_stat_progress_create_index
block_exists NCountUnits:blockblks_exists statistic from pg_stat_slru
block_hit NCountUnits:blockNumber of buffer hits (blocks found in cache)
block_write NCountUnits:blockblks_written statistic from pg_stat_slru
block_zeroed NCountUnits:blockblks_zeroed statistic from pg_stat_slru
buffer_backend NGaugeNumber of buffers written directly by backends
buffer_backend_fsync NGaugeNumber of fsync calls by backends
buffer_checkpoint NGaugeNumber of buffers written during checkpoints
buffer_clean NGaugeNumber of buffers written by background writer
buffercache_summary_buffer_dirty GaugeUnits:bufferNumber of dirty shared buffers
buffercache_summary_buffer_pinnedGaugeNumber of pinned shared buffers
buffercache_summary_buffer_unused GaugeUnits:bufferNumber of unused shared buffers
buffercache_summary_buffer_used GaugeUnits:bufferNumber of used shared buffers
buffercache_summary_usagecount_avgGaugeAverage usage count of used shared buffers

Full metrics context for PostgreSQL is available with an account. Request access

Understanding PostgreSQL observability

PostgreSQL's observability model is notably comprehensive due to its extensive statistics collector system that tracks detailed runtime information across multiple dimensions. Unlike many commercial databases that require proprietary agents or expensive enterprise editions for deep visibility, PostgreSQL exposes rich telemetry through system catalogs and statistics views like pg_stat_activity, pg_stat_database, and pg_stat_bgwriter. The database provides over 286 distinct metrics covering everything from buffer pool activity to background writer performance, making it particularly well-suited for integration with modern observability platforms. Common collection patterns leverage the postgres_exporter for Prometheus, OpenTelemetry's PostgreSQL receiver, or native integrations like Datadog's PostgreSQL check to transform these statistics into actionable monitoring data.