mdcms/neuraldb-docs/pages/ops-monitoring.md

title	sort	section-id	keywords	description	language
Monitoring	100	operations	monitoring, Prometheus, Grafana, metrics, alerts, observability, dashboards	Monitoring NeuralDB with Prometheus metrics, Grafana dashboards, and alert configuration	en

Monitoring

Prometheus Metrics

Enable the metrics exporter:

metrics.enabled = true
metrics.port = 9187
metrics.path = /metrics

Key metrics:

Metric	Type	Description
neuraldb_connections_total	Gauge	Current connections by state
neuraldb_query_duration_seconds	Histogram	Query duration percentiles
neuraldb_vector_queries_total	Counter	Vector similarity queries by index
neuraldb_hnsw_index_size_bytes	Gauge	In-memory size of HNSW graphs
neuraldb_replication_lag_seconds	Gauge	Time lag per replica
neuraldb_database_size_bytes	Gauge	Total database size

Grafana Dashboard

Import official dashboard ID 18921 from Grafana.com.

Alerting Rules

groups:
  - name: neuraldb
    rules:
      - alert: NeuralDBConnectionsHigh
        expr: neuraldb_connections_total{state="active"} / neuraldb_connections_max > 0.85
        for: 2m
        labels: { severity: warning }
      - alert: NeuralDBReplicationLagHigh
        expr: neuraldb_replication_lag_seconds > 30
        for: 1m
        labels: { severity: warning }
      - alert: NeuralDBVectorBufferExhausted
        expr: neuraldb_hnsw_index_size_bytes > (neuraldb_vector_buffer_size_bytes * 0.90)
        for: 5m
        labels: { severity: warning }

Built-In Query Statistics

SELECT query, calls, round(mean_exec_time::numeric, 2) AS avg_ms
FROM pg_stat_statements ORDER BY mean_exec_time DESC LIMIT 10;

SELECT sum(blks_hit) * 100.0 / sum(blks_hit + blks_read) AS cache_hit_ratio
FROM pg_stat_database WHERE datname != 'template0';