Demo 02 — GC Monitoring with Prometheus

Two Spring Boot apps (G1GC on :8080, ZGC on :8081) + Grafana LGTM + standalone Prometheus scraping /actuator/prometheus. Live GC pause histograms and traces simultaneously.

Required configuration

# application.yml
management:
  endpoints:
    web:
      exposure:
        include: health,info,prometheus,metrics
  prometheus:
    metrics:
      export:
        enabled: true
  metrics:
    distribution:
      percentiles-histogram:
        jvm.gc.pause: true
      percentiles:
        jvm.gc.pause: 0.5,0.95,0.99

Key PromQL queries

# GC pause P99 (ms)
histogram_quantile(0.99, rate(jvm_gc_pause_seconds_bucket[1m])) * 1000

# Heap utilisation %
jvm_memory_used_bytes{area="heap"} / jvm_memory_max_bytes{area="heap"} * 100

Fedora / RHEL notes

All bind mounts require :Z for SELinux. Prometheus uses tmpfs + user: root to avoid rootless Podman named volume permission issues. Do not mount custom config into otel-lgtm — it breaks the internal OTel pipeline and produces blank dashboards.

Generate load

# Allocate heap and watch the GC panels react
curl "http://localhost:8080/allocate?mb=100&iterations=10"

# Virtual threads — 500 concurrent tasks
curl "http://localhost:8080/virtual-threads?tasks=500&workMs=5"

Reference

• Demo source
• Grafana JVM dashboard 4701