Monitoring & Metrics
Live CPU, RAM, and disk charts per server, and per-app health and resource usage.
Monitoring gives you a real-time view of your fleet without needing a separate observability stack.
Dashboard
The Dashboard shows KPI stat tiles and per-server metric sparklines, plus a recent-activity feed — the fastest way to glance at "is everything OK" across every connected server.
Per-server charts
The Monitoring page shows detailed CPU, RAM, and disk usage over time per server, as interactive area charts — hover for an exact value at any point via the crosshair tooltip.
Per-application health
Below the server charts, each application gets its own CPU usage chart and memory bar, sampled from the containers it's actually running — so you can tell which specific app is eating resources on a shared server, not just that the server as a whole is busy.
Request metrics
Below an app's CPU/memory panel, click Request metrics for on-demand HTTP-level detail — latency percentiles and the mix of 2xx/3xx/4xx/5xx status codes the app has actually served, read straight from the reverse proxy.
This is Traefik-only: it reads Traefik's own metrics endpoint, so it isn't available for servers running Caddy. If it's not enabled yet, an Enable button reconfigures Traefik to expose metrics — a one-time step per server.
Alerts
Define alert rules (e.g. CPU above a threshold for N minutes) per server. When a rule breaches, DeployOS fires your configured notifications — Slack, Discord, email, or a generic webhook.
Data retention
Recent metrics (last 6 hours) are kept at full resolution; anything older is rolled up into hourly averages, so history stays useful without the metrics store growing without bound.
Next: Logs & Console for drilling into a specific app's behavior, or Running the Doctor for a one-shot health diagnosis.