delonix dashboard

Dashboard de resumo/KPIs (TUI estilo htop) — RAM/rede/disco, uptime por-container, JSON e Prometheus.

Summary/KPI dashboard (htop-style TUI) — RAM/network/disk, per-container uptime, JSON and Prometheus.

Vista viva do estado do runtime — containers, VMs, imagens, redes, storage — num só ecrã, sem precisar de correr ls em 5 grupos diferentes. --scope <container|vm|network|volume|image> foca num só grupo — substituiu os cinco <grupo> dash que existiam antes (corte limpo, sem alias): a mesma tela, cinco cascas clap idênticas a menos para manter. KPIs dinâmicos: memória do slice cgroup, tráfego rx/tx acumulado por-container (com contagem explícita de containers --net host/none não medidos, nunca somados como zero em silêncio), uso de disco por área (imagens/volumes/VM-images/containers), e uptime real por container (coluna UP, do pid_starttime). A tecla m alterna o sparkline entre containers a correr e memória usada. --once imprime um snapshot de texto e sai (scripts/CI) — é também o que acontece automaticamente quando o stdout não é um terminal. --json dá o mesmo snapshot em JSON, para scripts ou um datasource do Grafana. Para scrape contínuo, o delonix-mgmt/delonix-cri expõem /metrics (Prometheus — gauges de containers/VMs/memória/rede/disco) e GET /v1/dash (o mesmo DashSummary em JSON); os campos caros (rede/disco) recalculam em background a cada 30s, o scrape em si fica sempre rápido.

A live view of runtime state — containers, VMs, images, networks, storage — in one screen, without running ls across 5 different groups. --scope container|vm|network|volume|image focuses it on one group — it replaced the five <group> dash commands that existed before (clean cut, no alias). Dynamic KPIs: cgroup slice memory, accumulated rx/tx traffic per container (with an explicit count of unmeasured --net host/none containers, never silently summed as zero), disk usage by area (images/volumes/VM-images/containers), and real per-container uptime (the UP column, from pid_starttime). The m key toggles the sparkline between running containers and memory used. --once prints a text snapshot and exits (scripts/CI) — also what happens automatically when stdout isn't a terminal. --json gives the same snapshot as JSON, for scripts or a Grafana datasource. For continuous scraping, delonix-mgmt/delonix-cri expose /metrics (Prometheus — container/VM/memory/network/disk gauges) and GET /v1/dash (the same DashSummary as JSON); the expensive fields (network/disk) recompute in the background every 30s, so the scrape itself always stays fast.

📄 Implementação real em Rust: cmd/dash.rs

Usage: delonix dashboard [OPTIONS]

Options:
      --l18n <en|pt>
          Output language: `en` (default) or `pt` (Portuguese, pt_AO). Also settable via `$DELONIX_L18N`. Global — works before any subcommand

      --scope <SCOPE>
          Focus on one kind of resource instead of the global summary
          
          [possible values: all, container, vm, network, volume, image]

      --once
          Print ONE text snapshot and exit (no TUI) — for scripts/CI; the default when stdout is not a terminal

      --json
          Print ONE snapshot as JSON and exit (no TUI, no ANSI) — for scripts/Grafana JSON datasource

  -h, --help
          Print help (see a summary with '-h')

EXAMPLES:
  # an htop-style live view of containers, VMs, networks and volumes
  delonix dashboard

  # focused on one kind of resource instead of the global summary
  delonix dashboard --scope container

  # one text snapshot and exit — the default when stdout is not a terminal
  delonix dashboard --once

  # one JSON snapshot, for a script or a Grafana JSON datasource
  delonix dashboard --json

SEE ALSO:
  delonix container stats · delonix vm ls · delonix system info · delonix
  serve api

  delonix › dashboard

ExemplosExamples

delonix dashboard

delonix dashboard --scope container

delonix dashboard --once

delonix dashboard --json | jq '.tiles'

LaboratórioLab

delonix dashboard --once
delonix dashboard

delonix dashboard --once
delonix dashboard

DesafioChallenge