Fleet
One page for every server in a workspace — status, resource pressure, open incidents and what changed — sorted so the server that needs you is at the top.
Open a workspace and choose Monitoring in the sidebar. The Fleet tab lists every server in the workspace with its live vitals.
Reading the table
| Column | What it shows |
|---|---|
| Server | Name, provider and status. Click through to the server's monitoring pages. |
| CPU / Memory | The latest value and a one-hour trend line. The line turns amber above 80 % and red above 90 % (memory: 85 % / 95 %). |
| Disk | The fullest mount point on the server, plus a full in ~N d chip when a mount is forecast to fill within 30 days. |
| Network | Current download plus upload throughput across all interfaces. |
| Load | One-minute load average against the core count; amber when load exceeds twice the cores. |
| Incidents | Open alert incidents on that server. |
| Agent | Which agent is installed and its version. |
| Last sample | How long ago the server last reported. |
Status values
- Online — the agent is heartbeating normally.
- Quiet — heartbeats have stopped recently; Devpilot waits a few minutes before calling the server offline.
- Offline — no heartbeat for longer than the grace period. An availability incident is open.
- No agent — the server is connected to Devpilot but the monitoring agent is not installed. Install it from the server's monitoring page.
- Paused — monitoring was switched off for this server in its settings.
Finding the server that needs attention
- Sort by Most stressed (the default — highest of CPU, memory or disk first), Status (offline and quiet first) or Name.
- Filter chips narrow the list to Needs attention (offline, quiet, open incidents, or any resource above 85 %), Online, Offline / quiet or No agent / paused.
- Search matches server names and hosts.
Changes
The Changes panel lists what happened across the fleet in the last 24 hours — deployments, script runs, outages and recoveries, reboots and agent restarts — so a spike on a chart can be matched to the event that caused it.
Links to a specific server from older bookmarks (…/monitoring?resource=server&id=…) open that server's monitoring page directly.
Next steps
Monitoring Overview
See what every server is doing right now and what it did over the last year — live metrics, processes, services, disks, network, availability, alerts, and the activity trail behind every change.
Alerts
Rules that watch any server metric, incidents that capture what the server looked like when they fired, and notifications to the people and channels you choose.