CLI Reference¶

Core Commands¶

Command	Description
`nomad init`	Setup wizard
`nomad collect`	Start data collection
`nomad collectors`	Which collectors run here, why, and how their runs went
`nomad dashboard`	Launch web interface
`nomad console launch`	Open the Console in your browser through an SSH tunnel (Console access)
`nomad demo`	Demo mode with synthetic data
`nomad status`	System status
`nomad syscheck`	Verify requirements
`nomad config check`	Does nomad.toml read? If not, the line and what is wrong (Configuration)

Educational Analytics¶

Command	Description
`nomad edu explain <job_id>`	Analyze job with recommendations
`nomad edu trajectory <user>`	User proficiency over time
`nomad edu report <group>`	Group/course report

Options for all edu commands: - --db PATH — Database path - --json — JSON output - --days N — Lookback period (trajectory/report)

Analysis¶

Command	Description
`nomad disk <path>`	Filesystem trends
`nomad jobs`	Job history
`nomad similarity`	Network analysis
`nomad alerts`	View alerts

ML & Prediction¶

Command	Description
`nomad train`	Train prediction models
`nomad predict`	Run predictions
`nomad report`	Generate ML report

Community Dataset¶

Command	Description
`nomad community export`	Export anonymized data
`nomad community preview`	Preview export
`nomad community verify`	Verify export integrity

Data Collection¶

# Continuous collection
nomad collect

# Single cycle
nomad collect --once

# Specific collectors
nomad collect -C disk,slurm,groups

# Which collectors run here, why, and whether they collect anything
nomad collectors

# On the hub: how every site's collectors ran in the last 7 days
nomad collectors --db ~/.local/share/nomad/combined.db

# Jobs from before nomad (or not collected): an sacct export, or sacct itself
nomad import sacct EXPORT [--apply]
nomad import sacct --from 2025-10-01 [--to 2026-10-01] [--apply]

See Collectors (and "Jobs from before nomad" there).

Usage Report¶

# The administrators' report for a period (Markdown and JSON)
nomad usage-report --from 2025-10-01 --to 2026-10-07 --cluster c1 [--out DIR]

# From an sacct export instead of nomad's job records
nomad usage-report --sacct EXPORT --from 2025-10-01 --to 2026-10-07 --cluster c1

# Draft the site settings (tiers, partitions) from nomad's node samples
nomad usage-report init --cluster c1

# The period's usernames, for a department/school map (--user-map)
nomad usage-report people --cluster c1 --out users.csv

Options: --config (report.toml), --db, --user-map, --exclude-users, --teaching SITE, --format md|json|both, --min-cell N, --guard-details FILE, --allow WORD. See Usage report.

Actions¶

# The catalog of commands the Console may start, on the hub or a site
nomad actions list [--json]
nomad actions run NAME [--site SITE | --here] [-p NAME=VALUE ...] [--json]

# The hub's key for reaching sites, and on each site, letting it run the agent only
nomad actions key
nomad agent install-key 'ssh-ed25519 AAAA... nomad-actions@hub' [--from PATTERN] [--apply]

See Actions.

Global Options¶

Option	Description
`--config PATH`	Config file path
`--db PATH`	Database path
`--verbose` / `-v`	Verbose output
`--quiet` / `-q`	Suppress output
`--help`	Show help

Insight Engine¶

Command	Description
`nomad insights brief`	Executive summary
`nomad insights detail`	Comprehensive report
`nomad insights json`	Machine-readable output
`nomad insights slack`	Slack-formatted message
`nomad insights digest`	Periodic digest

Options: --hours N (lookback window), --db PATH, --cluster NAME

System Dynamics¶

Command	Description
`nomad dyn summary`	Full dynamics narrative
`nomad dyn diversity`	Simpson/Shannon diversity indices
`nomad dyn niche`	Pianka niche overlap matrix
`nomad dyn capacity`	Multi-dimensional carrying capacity
`nomad dyn resilience`	Disturbance detection and recovery time
`nomad dyn externality`	Cross-group impact scoring

Options: --hours N, --db PATH, --by {user,group,partition}, --json

Cloud Monitoring¶

Command	Description
`nomad cloud collect`	Collect metrics from cloud providers
`nomad cloud status`	Current resource utilization
`nomad cloud diag`	Cost and performance diagnostics
`nomad cloud edu`	Usage recommendations

Options: --provider {aws,azure,gcp}, --db PATH

Reference¶

Command	Description
`nomad ref`	Browse all topics
`nomad ref <topic>`	Look up a specific topic
`nomad ref search <query>`	Full-text search

Topics cover all commands, configuration, concepts, and mathematical foundations.

Issue Reporting¶

Command	Description
`nomad issue report`	Interactive bug/feature/question form
`nomad issue search <keywords>`	Search existing GitHub issues
`nomad issue info`	Preview auto-collected system info

Options for report:

-c, --category {bug,feature,question} — Issue category
-m, --component {collectors,alerts,dashboard,...} — Affected component
-t, --title TEXT — Issue title
--email — Send via email instead of GitHub
--json — Output formatted issue as JSON
--no-duplicate-check — Skip duplicate search
--db PATH — Database path for system info

When a GitHub token is configured in nomad.toml under [issue_reporting], issues are submitted directly via the GitHub API with auto-labeling. Without a token, opens a pre-filled GitHub issue form in the browser.

`nomad diag gpu`¶

GPU diagnostic report: utilization summary, workload distribution, temperature status, and hardware health.

nomad diag gpu                    # All nodes, last 24h
nomad diag gpu --node node51      # Specific node
nomad diag gpu --hours 48         # Extended window
nomad diag gpu --health           # Hardware health only (PCIe, ECC, row remap)

Options:

Option	Description
`--node TEXT`	Limit to a specific node
`--hours INT`	History window in hours (default: 24)
`--health`	Show hardware health report only
`--db PATH`	Database path override

`nomad insights gpu`¶

Narrative summary of GPU usage patterns and health concerns. Highlights workload classification, the utilization gap between nvidia-smi and Real Util, and any DCGM health alerts.

nomad insights gpu
nomad insights gpu --hours 12
nomad insights gpu --node spdr16