When a server is misbehaving, you want the whole picture on one screen - not top for CPU, then df for disk, then something else for network. Glances gives you that single pane. It's a cross-platform monitor that shows CPU, memory, load, disk I/O, filesystem usage, network throughput, and the busiest processes together, updating live, and it intelligently highlights whatever is currently a problem so your eye goes straight to it.
The Whole System at a Glance
Run it locally, or expose it to watch a remote box:
# Live local dashboard
glances
# Run as a server to monitor remotely
glances -s
# Or expose a web dashboard
glances -w
Adaptive and Accessible
Glances color-codes metrics by severity, so a filesystem filling up or a process eating CPU stands out without hunting. It can run as a client/server for remote monitoring, serve a web UI, and export metrics to time-series databases - so the same tool scales from a quick terminal check to a lightweight monitoring feed.
Rough Edges and Limitations
- Python footprint. It's heavier than a single static Go/Rust binary, and optional features need extra packages.
- Not a full monitoring stack. For history, alerting, and dashboards at scale, dedicated systems fit better.
- Density can overwhelm. The all-in-one view is a lot at first glance.
For a fast, complete read on a machine's health from one terminal, Glances is hard to beat.
Internal link suggestion: link "the busiest processes" to the btop or procs post for process-focused views.