Skip to content

scottchiefbaker/dool

Repository files navigation

Dool

Dool is a Python3 compatible fork of Dstat.

After Dag Wieers ceased development of Dstat I forked the project to continue development.

What is Dool?

Dool is a command line tool to monitor many aspects of your system: CPU, Memory, Network, Load Average, etc. It also includes a robust plug-in architecture to allow monitoring other system metrics.

Installation:

  1. Download the latest release archive file or clone the Git repo
  2. Extract the archive to an appropriate temporary directory on your system
  3. Run the install.py script

Usage:

dool [--preset] [--plugin] [delay]

My most common usage of Dool is:

dool --more 15

which uses the --more preset and outputs data every 15 seconds. Available presets are --defaults, --more, or --all. If no delay is specified, Dool will default to outputting every second.

Screenshots:

Dark mode (default) Dool Light

Light mode Dool Light

Plugins:

Dool ships with many plug-ins to configure the output to your taste.

dool --cpu --net --time --full # Show CPU usage, and each network interface
dool --disk -D total,sda,sdd   # Show the total disk IO, and /dev/sda and /dev/sdd
dool --net -N eth0,eth1        # Show the network traffic for eth0 and eth1

A list of available plug-ins are available if you run dool --version

Note:

Dool requires a 256 color compatible terminal. Most modern terminal emulators support this automatically.

Bits vs Bytes:

One of the changes in dool is measurement of network and disk bandwidth in bits instead of bytes. This can be confusing if you're used to seeing the lower numbers in dstat. If you'd rather see bandwidth reported in bytes you can pass the --bytes option.

See also:

Other tools similar to Dool

Pull Requests and Branches

The latest stable release (plus bugfixes) will live on the master branch, and development of new features will occur on the next branch. Please have pull requests target the next branch.

Various feature/bug branches may come and go as we work on more complex functionality, but those can be safely ignored.