Multi-Node Fleet Dashboard 🔒
Paid feature. Requires a Pro plan or higher. Free tenants still see the single-node dashboard and the classic tenant-summary counts. See Orbit Pro for the tier matrix.
The Multi-Node view renders every node in your fleet as a tile with:
- Live status — green / amber / red based on the last heartbeat (fresher than 5 minutes = online, 5–10 minutes = warning, older = offline).
- CPU / memory / disk sparks — the most recent values reported by the node’s heartbeat ingest.
- 7-day alert badge — the count of active fleet alerts raised against the node in the last week.
- Click-through to the classic node detail page for tags, groups, apps, and metrics history.
Opening the page
Control Panel → Fleet → Grid view.
The backend endpoint is GET /api/fleet/multi-node-summary; it
returns a deduped list of node tiles plus online / warning / offline
rollups. The response is cached per-tenant for 5 seconds so polling at
1 Hz from the dashboard doesn’t hammer the DB.
402 on free plans
If your tenant is on the community tier the endpoint returns
402 upgrade_required with the canonical response shape, and the UI
shows an upgrade card linking to the subscription page.
Known limits
- The tile metrics use the CPU / memory / disk columns written by node heartbeats; there is no per-container breakdown in the grid. Use the Containers page on the node for that.
- The page does not stream; polling at 5 s is good enough for most small-fleet operator workflows. Websocket push is on the roadmap.