Grafana Dashboards

OpenGate IAM includes pre-built Grafana dashboards for monitoring authentication events, session activity, JVM health, and HTTP performance.

Quick Start

# Start Prometheus + Grafana
docker compose -f docker/docker-compose.observability.yml up -d

# Open Grafana
open http://localhost:3002   # admin / admin

Import the dashboard JSON from the observability/dashboards/ directory in the repository.

Available Dashboards

Dashboard	Description
OpenGate Overview	Login rates, active sessions, error rates across all services
Auth Service	Token issuance, MFA usage, grant type breakdown
JVM Health	Heap usage, GC pauses, thread counts per service
HTTP Performance	P50/P95/P99 latency, request rates, error rates
Database	HikariCP pool usage, query duration, connection errors

Key Panels

# Logins per minute (success)
rate(opengate_login_success_total[1m]) * 60

# Logins per minute (failure)
rate(opengate_login_failure_total[1m]) * 60

Active Sessions

opengate_sessions_active_total

HTTP P95 Latency

histogram_quantile(0.95,
  rate(http_server_requests_seconds_bucket{job="opengate-gateway"}[5m])
)

JVM Heap Usage

jvm_memory_used_bytes{area="heap"} / jvm_memory_max_bytes{area="heap"} * 100

Useful PromQL Queries

# Error rate per service (5xx responses)
sum by (job) (
rate(http_server_requests_seconds_count{status=~"5.."}[5m])
)

# DB connection pool saturation
hikaricp_connections_active / hikaricp_connections_max * 100

# MFA challenge rate
rate(opengate_mfa_challenge_total[5m])

# Top slowest endpoints
topk(10, http_server_requests_seconds_max{job="opengate-gateway"})

Pre-built dashboard IDs (Grafana.com)

Import these community dashboards as a starting point:

4701 — JVM Micrometer
6756 — Spring Boot Statistics
11378 — Redis Exporter