[opensourcedex]

s-tui is a free, open source monitoring & observability project written in Python and released under GPL-2.0. It has 5,093 GitHub stars, 181 forks and 39 open issues, and was last pushed 3 days ago. On this registry it ranks #89 of 191 tracked projects in Monitoring & Observability, with 5 head-to-head comparisons available.

What is s-tui?

s-tui is a terminal-based CPU stress and monitoring utility, written in Python and released under the GPL-2.0 licence, that renders CPU temperature, frequency, power and utilization as live graphs for anyone who needs to inspect or load a Linux machine from a text console or an SSH session.

What it is

s-tui, short for Stress Terminal UI, is a Python application that plots CPU sensors in a graphical way from the terminal. It monitors CPU temperature, utilization, frequency and power, and it ships a built-in CPU stress test that requires no dependencies and works out of the box, alongside optional integration with the external stress tools stress and stress-ng. It also provides throttle indicators that show performance dips caused by thermal throttling. The project lives in the Linux and POSIX terminal ecosystem, is published on PyPI, and is packaged in the Ubuntu, Debian, Arch Linux, Manjaro, OpenSUSE and Fedora repositories.

The concrete problem it solves is that graphical system monitors require an X-server, which headless servers, remote machines and minimal installations do not have. s-tui renders its graphs with no X-server and works over a plain TTY or SSH connection. It also removes the usual dependency on an external load generator, because its built-in stress test runs without additional packages while still allowing stress and stress-ng to be driven when heavier, more configurable load is wanted. Both things appear in one view, so a load can be applied and its effect on temperature, frequency and throttling observed at the same time.

Key capabilities

  • Live terminal graphs of CPU temperature, frequency, power and utilization, with a UTF-8 button for smoother graphs on systems that support it.
  • Built-in CPU stress test with zero dependencies, plus optional integration with external stress or stress-ng.
  • Throttle indicators that show performance dips caused by thermal throttling.
  • Single-line stats modes via -t/--terminal and -j/--json, so readings can be piped into other tools.
  • CSV logging via -c/--csv, written to the default file s-tui_log_*.csv and redirectable with --csv-file.
  • Debug logging with -d/--debug to _s-tui.log, a custom path with --debug-file, and -dr/--debug_run to run for five seconds and quit.
  • Saved configurations, selectable graph and summary menus, a reset button, and threshold scripts triggered when a reading crosses a configured limit.

Who uses it and how

  • Operators of headless servers and remote machines, where no X-server exists and readings must come over SSH or a bare TTY.
  • People diagnosing thermal throttling on laptops and workstations, who watch temperature, frequency and power while the built-in stress test or stress-ng applies load.
  • Automation authors who consume --json or --csv output and feed it into other scripts, monitoring pipelines or post-run analysis.
  • Distribution users on x86 and ARM, since installation is available through pip as well as the Ubuntu, Debian, Arch, Manjaro, OpenSUSE and Fedora package managers.
  • Users who attach threshold scripts so that something reacts when a sensor crosses a defined limit.

Getting started

The most up-to-date version installs with pip install s-tui --user, which normally places an executable in ~/.local/bin that must be on the PATH; sudo pip install s-tui installs it system-wide. Distribution packages are simpler where available, for example sudo apt install s-tui on Ubuntu 18.10+ and Debian 10+, and the application is then started by running s-tui.

How it compares

The only comparable tools named in the facts are stress and stress-ng, which serve as companions rather than rivals: s-tui can drive both, but those load generators do not monitor temperature, frequency, power or throttling. s-tui therefore fits as the viewing and measurement layer for work that a bare load generator or a psutil-based script would otherwise leave unobserved.

When to use it — and when not to

A self-hoster must provide a Python runtime with psutil available, may need python-dev before installing, and may need root for a system-wide install; the --user path also requires ~/.local/bin to be on the PATH. It is not the right choice for anyone who needs retained historical metrics, dashboards across many hosts, or alerting, because it is a live terminal view that can only export CSV and trigger local threshold scripts. One documented weakness is distribution lag: the PPA for Ubuntu 18.04 and 16.04 is described as not up to date, so those users should install from pip instead.

project readme (upstream, from github) — read inline

The Stress Terminal UI: s-tui

PyPI version Downloads

Stress-Terminal UI, s-tui, monitors CPU temperature, frequency, power and utilization in a graphical way from the terminal.

Screenshot

Table of Contents

What it does

  • Monitoring your CPU temperature/utilization/frequency/power
  • Shows performance dips caused by thermal throttling
  • Requires no X-server
  • Built-in CPU stress test (zero dependencies, works out of the box)
  • Optional integration with external stress tools (stress/stress-ng)

Usage

s-tui

Simple installation

pip (x86 + ARM)

The most up to date version of s-tui is available with pip.

Install with:

pip install s-tui --user

(This usually creates an executable in ~/.local/bin/ dir. Make sure it is in your PATH)

To install as root

sudo pip install s-tui

You might need to install python-dev first

Installation in virtualenv with pipsi:

pipsi install s-tui

More installation methods

Ubuntu (18.10 and newer)

sudo apt install s-tui

Ubuntu (18.04, 16.04)

A PPA is available but is not up to date

sudo add-apt-repository ppa:amanusk/python-s-tui
sudo apt-get update
sudo apt-get install python3-s-tui

Debian (10 and newer)

sudo apt install s-tui

Arch Linux, Manjaro

s-tui is in the Arch repository:

sudo pacman -S s-tui

s-tui-git follows the master branch, maintained by @MauroMombelli

Install it with: yay -S s-tui-git

OpenSUSE

sudo zypper install s-tui

Fedora

s-tui is in the Fedora repository:

sudo dnf install s-tui

Options

TUI interface:

The side bar houses the controls for the displayed graphs.
At the bottom, all sensors reading are presented in text form.

* Use the arrow keys or 'hjkl' to navigate the side bar
* Toggle between stressed and regular operation using the radio buttons in 'Modes'.
* If you wish to alternate stress defaults, you can do it in <Stress options>
* Select graphs to display in the <Graphs> menu
* Select summaries to display in the <Summaries> menu
* Use the <Reset> button to reset graphs and statistics
* If your system supports it, you can use the UTF-8 button to get a smoother graph
* Save your current configuration with the <Save Settings> button
* Press 'q' or the <Quit> button to quit

* Run `s-tui --help` to get this message and additional cli options

optional arguments:
  -h, --help            show this help message and exit
  -d, --debug           Output debug log to _s-tui.log
  --debug-file DEBUG_FILE
                        Use a custom debug file. Default: _s-tui.log
  -dr, --debug_run      Run for 5 seconds and quit
  -c, --csv             Save stats to csv file
  --csv-file CSV_FILE   Use a custom CSV file. Default: s-tui_log_<TIME>.csv
  -t, --terminal        Display a single line of stats without tui
  -j, --json            Display a single line of stats in JSON format
  -nm, --no-mouse       Disable Mouse for TTY systems
  -v, --version         Display version
  -tt T_THRESH, --t_thresh T_THRESH
                        High Temperature threshold. Default: 80

Throttle Indicators

When CPU throttling is detected, s-tui changes the frequency graph and summary text color and appends a reason label. Labels may be combined with / (e.g. T/W).

Intel, with root + msr module (detailed per-core reasons via IA32_THERM_STATUS):

Label Meaning
T Thermal — core temperature exceeded TjMax
H PROCHOT — external thermal signal (VRM, GPU, battery)
C Critical — near emergency shutdown temperature
W Power limit — PL1/PL2 watt budget exceeded
A Current limit — electrical current (amps) limit hit
X Cross-domain — throttled by another domain (e.g. GPU)

AMD, with root + msr module (per-core P-state cap via PStateCurLim):

Label Meaning
Pc P-state cap — the SMU has barred the core from its top P-state

AMD exposes no equivalent of Intel's reason breakdown through MSRs. The thermal/power/current limits themselves (PPT, TDC, EDC, THM, STAPM) live in the SMU power-management table, which needs an out-of-tree kernel module to reach, so PStateCurLim is what remains: it indicates that a cap is in force, not why. It behaves as a severity threshold rather than a reason code — the same chip pinned at the same temperature was observed showing Pc under a load heavy enough to drive it below its rated base clock, and nothing under a lighter one. Expect it to stay clear on a well-cooled part that is boost-limited but never forced that far down.

Without root (sysfs fallback, thermal only):

Label Meaning
Tc Core thermal throttle
Tp Package thermal throttle (affects all cores)

When a temperature sensor exceeds the high value defined by the sensor, or a user configured threshold (default 80°C), the sensor summary text summary and associated graph will change color to indicate this.

Dependencies

s-tui includes a built-in CPU stress test that works out of the box with no extra dependencies. For better stress performance, install numpy:

pip install s-tui[stress]

or install numpy directly:

pip install numpy

When numpy is available, the built-in stresser uses repeated dense matrix multiplication (BLAS) for maximum thermal output. Without numpy, it falls back to hashlib SHA-256 hashing which still provides good CPU load.

For additional stress options (memory, I/O, sync workers), you can optionally install the external stress or stress-ng tool:

sudo apt-get install stress

Configuration

s-tui is a self-contained application that can run out-of-the-box and doesn't need config files to drive its core features. However, additional features like running scripts when a certain threshold has been exceeded (e.g. CPU temperature) does necessitate creating a config directory. This directory will be made in ~/.config/s-tui by default.

Saving a configuration

Selecting \ will save the current configuration to ~/.config/s-tui/s-tui.conf. If you would like to restore defaults, simply remove the file.

Adding threshold scripts

s-tui gives you the ability to run arbitrary shell scripts when a certain threshold is surpassed, like your CPU temperature. You can define this custom behaviour by adding a shell file to the directory ~/.config/s-tui/hooks.d with one of the following names, depending on what threshold you're interested in reacting to:

  • tempsource.sh: triggered when the CPU temperature threshold is exceeded

If s-tui finds a script in the hooks directory with the name of a source it supports, it will run that script every 30 seconds as long as the current value of the source remains above the threshold.

Note that at the moment only CPU temperature threshold hooks are supported.

Run from source code

Start by cloning the repository

git clone https://github.com/amanusk/s-tui.git
cd s-tui

Install required dependencies as [root] or as (local user)

[sudo] pip install urwid (--user)
[sudo] pip install psutil (--user)

Install stress (op

readme truncated — read the full docs on github

Frequently asked questions

Is s-tui free to use?

s-tui is open source under the GPL-2.0 licence. There is no licence fee and no seat count — you can self-host it or, where the project offers one, pay a vendor for a managed version instead.

What does s-tui do?

Terminal-based CPU stress and monitoring utility

What is s-tui written in?

s-tui is primarily written in Python. Its source is publicly available at https://github.com/amanusk/s-tui, and it has 5,093 GitHub stars.