Monitoring

Hammerspace recommends Grafana as a monitoring tool, with Prometheus as the time series database service to collect metrics from Hammerspace and LSS / Tier 0 nodes. The Prometheus exporters are enabled on a Hammerspace cluster with the cluster-config command as follows:

anvil1> cluster-config --prometheus-exporters-enable
ID: 07106af1-4ae9-4509-8d3c-497614109567
Name: AE9Y4XC9E5X9J5
State: Standalone
Management IPs: [10.200.106.160/20]
Data IPs: [10.200.106.160/20]
Cluster floating IPs: [10.200.106.160/20]
Since: 2025-06-18 19:20:10 UTC
Timezone: UTC
GFS participant max suspected time: 30 minutes
Prometheus exporters: Enabled
Evaluation expiration: 2025-07-18 19:18:47 UTC
Online license activation support: true
NAS volume capacity: [Total: 136.6GB, Used: 1.2GB, Free: 135.3GB]
Share space (quota): [Total: 0B, Used: 0B, Free: 0B]
Metadata servers:
[Object type: Anvil, Node name: anvil1.hammer.space, Role: Primary, Admin state: Up, Oper state: Up]

For more information, reference the following:

For the Hammerspace Grafana dashboards — downloading, installing, and configuring them — see Monitoring Cluster Health in the Hammerspace Administration Guide.
Refer to Monitoring Linux host metrics with the Node Exporter | Prometheus for installation instructions for Prometheus exporters for Linux nodes.
Refer to DCGM Exporter — NVIDIA GPU Telemetry 1.0.0 documentation for using Prometheus and Grafana with Nvidia GPU nodes.