dool - Versatile System Resource Statistics Cheatsheet
dool is a versatile system resource monitor and the actively-maintained successor to dstat (whose development ended). It replaces vmstat, iostat, netstat, and ifstat with a single colorized, per-second view that shows CPU, disk, network, memory, paging, and more side by side — making correlations obvious (e.g. a disk spike lining up with a network burst). It keeps full dstat command compatibility and is extensible with Python plugins.
Installation
| Platform | Command |
|---|
| Debian/Ubuntu | sudo apt install dool |
| Fedora | sudo dnf install dool |
| Arch (AUR) | yay -S dool |
| pip | pip install dool |
| From source | clone the repo, run ./dool |
| Verify | dool --version |
Basic Usage
| Command | Description |
|---|
dool | Default: CPU, disk, net, paging, system — every 1s |
dool 2 | Refresh every 2 seconds |
dool 2 10 | 10 updates, 2 seconds apart |
dool -h | Help and full plugin list |
dool --list | List available plugins |
The Classic Combination
# CPU + disk + net + memory + load, timestamped, every 2s
dool -tcdnml 2
| Flag | Shows |
|---|
-t | Time |
-c | CPU stats |
-d | Disk read/write |
-n | Network recv/send |
-m | Memory usage |
-l | Load average |
-g | Paging |
-s | Swap |
-p | Process stats |
-y | System (interrupts, context switches) |
Targeted Monitoring
| Command | Description |
|---|
dool -d -D sda,sdb | Per-disk stats for specific devices |
dool -n -N eth0,eth1 | Per-interface network stats |
dool -c -C 0,1,total | Per-CPU-core stats |
dool --top-cpu | Show the top CPU-consuming process |
dool --top-io | Show the top I/O process |
dool --top-mem | Show the top memory process |
Handy “Top” Plugins
| Plugin | Reveals |
|---|
--top-cpu | Most CPU-hungry process |
--top-io | Heaviest I/O process |
--top-mem | Biggest memory process |
--top-bio | Top block I/O process |
--top-latency | Highest-latency process |
# One-line situational awareness: resources + the worst offenders
dool -tcmndl --top-cpu --top-io 2
Output & Export
| Option | Description |
|---|
--output FILE.csv | Write stats to CSV (great for graphing later) |
--nocolor | Disable colors (logs/pipes) |
--noupdate | Print sequential lines instead of updating in place |
--bits | Show network in bits |
# Capture a soak test to CSV for later analysis
dool -tcdnml --output run1.csv 1 300
| Tool | Scope |
|---|
| dool | All-in-one CPU/disk/net/mem, per-second, CSV |
| dstat | Predecessor (unmaintained) — dool is the drop-in successor |
| iostat | Disk + CPU only |
| vmstat | Memory/CPU/IO summary |
| glances/btop | Full TUI dashboards |
dool’s strength is putting many subsystems on one line so you can correlate — pair it with iostat or iotop when you need to drill into disk specifically.
Resources
