-
Notifications
You must be signed in to change notification settings - Fork 115
Robottelo Setup
- How to setup Robottelo
dnf -y install gcc python3-devel libxml2-devel
To ensure a clean and isolated environment for robottelo, it's highly recommended to use a Python virtual environment (venv). This prevents conflicts with system-wide Python packages and allows you to manage project dependencies independently.
There are multiple tools that you can use to create venv with. For example uv, virtualenvwrapper, venv, and many more.
Use tool of your choice to create a venv. This guide demonstrates how to set
it up using virtualenvwrapper
.
Install (Fedora/RHEL)
dnf install python-virtualenvwrapper
# or using pip
pip install virtualenvwrapper
or install it using brew if you use MacOS
brew install virtualenvwrapper
Reload shell to get access to virtualenvwrapper
scripts or run
source /etc/profile
Create and activate the venv
mkvirtualenv robottelo
After running the command, new directory containing the virtual environment
gets created in $HOME/.virtualenvs
. This is where virtualenvwrapper
stores venvs.
Essential virtualenvwrapper
commands for venv management are:
-
mkvirtualenv
to create the venv -
workon <venv-name>
to activate the venv or to show what venv is currently activated -
rmvirtualenv
to remove the virtualenv. -
lsvirtualenv
list all virtual environments See the documentation for more.
Clone the upstream repository from GitHub and enter the robottelo directory.
git clone --origin upstream [email protected]:SatelliteQE/robottelo.git
cd robottelo/
Optional: Fork the repository if you haven't done already and add it as a remote.
git remote add origin [email protected]:<your-github-username>/robottelo.git
Tip
Optional: Install uv package manager which
is blazingly fast - 10-100x faster than pip
.
pip install uv
Install requirements. If you do not want to use uv
, start the command with
pip
. Make sure the virtual environment is activated.
uv pip install -r requirements.txt -r requirements-optional.txt
Robottelo is using broker to perform remote execution on the remote hosts.
First, we need to ensure that broker_settings.yaml
exists and contains the
correct credentials for remote execution. To do that, run broker --version
to
generate it and to display the path where the config file is stored. Once the
config file is obtained, an editor is opened.
broker --version
Broker settings file not found at /home/franta/.broker/broker_settings.yaml.
Get example file from
https://raw.githubusercontent.com/SatelliteQE/broker/master/broker_settings.yaml.example?
https://raw.githubusercontent.com/SatelliteQE/broker/master/broker_settings.yaml.example
(y, n) [y]:
Downloading example file from:
https://raw.githubusercontent.com/SatelliteQE/broker/master/broker_settings.yaml.example
Version: 0.5.3
Broker Directory: /home/franta/.broker
Settings File: /home/franta/.broker/broker_settings.yaml
Inventory File: /home/franta/.broker/inventory.yaml
Log File: /home/franta/.broker/logs/broker.log
That let's you modify the settings. It is recommended to change the following settings to have remote execution set up correctly.
host_username: <remote-user-you-want-to-use-for-remote-exec>
host_password: "<password>"
# IMPORTATNT: keep the following setting only if you want to use key-based
# authentication, comment/remove it otherwise
host_ssh_key_filename: "</path/to/the/ssh-key>"
Robottelo uses Dynaconf configuration engine which is
set to load configuration files from the conf/
directory.
Transform the template configuration files shipped with robottelo into an actual configuration files.
for conffile in conf/*.yaml.template; do
cp -- "$conffile" "${conffile%.yaml.template}.yaml"
done;
You can either modify config files in conf/
or create settings.local.yaml
file in the root of the repository to override the settings from conf/
.
cp settings.sample.yaml settings.local.yaml
To test the initial setup of robottelo, it is recommended to run a test.
Robottelo spins Satellite instances dynamically, to suppress this behavior, a
hostname of an existing Satellite/Foreman instance needs to be specified. To do
that put the following into your settings.local.yaml
or similarly edit
conf/server.yaml
.
dynaconf_merge: true
server:
hostnames: ['the.fqdn.of.your.satellite.acme.com']
ssh_username: <remote-user-name>
ssh_password: <remote-user-password>
Please keep in mind that the above setup is a minimal one. If you or your
organization has an existing robottelo configuration stored elsewhere, it might
be desired to not copy the template configuration files, but copy the existing
one into conf/
. Depending on what is desired, configuration can be also
symbolically linked to robottelo's conf/
directory and these settings can be
overridden by settings.local.yaml
.
Run one of our simplest tests - the test to test hammer ping
.
pytest tests/foreman/api/test_ping.py
The test should pass without errors.
This section describes how to set up robottelo for UI testing.
There are multiple ways to run UI tests - different browsers and webdrivers. The default browser used by robottelo is chrome
and the webdriver_kaifuku driver.
This guide walks you through the setup using the default settings. They contain a minimal setting that should quickly get you started with UI testing.
A prerequisite is having a Selenium Grid running. You can spin it up in a container if you do not have one.
podman run -d -p 4444:4444 -p 5900:5900 --shm-size="2g" -e SE_VNC_NO_PASSWORD=1 docker.io/selenium/standalone-chrome:latest
If you do not have Grid running on localhost
on port 4444
, change the ui.grid_url
setting. You could use the settings.local.yaml
for that.
# settings.local.yaml
---
ui:
grid_url: http://<fqdn-of-your-grid>:<grid-hub-listen-port>
Try to run a UI test
pytest tests/foreman/ui/test_operatingsystem.py::test_positive_verify_os_name
While the test is running, you should see an active session on your Selenium Grid in the Sessions section.
make robottelo env
git clone --origin upstream [email protected]:SatelliteQE/airgun.git
In robottelo venv, remove the existing installation of airgun.
uv pip uninstall airgun
And install airgun in the development mode - -e, --editable
, passing the path to the airgun directory as an argument.
uv pip install --editable ~/airgun
Warning
Keep in mind that every time you run pip install -r requirements.txt
, airgun is updated to the latest version, overriding the editable install.
Let's do a simple patch in airgun, add sleep
to give us more time to investigate what is on the Operating Systems page.
diff --git a/airgun/entities/os.py b/airgun/entities/os.py
index 827c46a..2752044 100644
--- a/airgun/entities/os.py
+++ b/airgun/entities/os.py
@@ -33,6 +33,8 @@ class OperatingSystemEntity(BaseEntity):
def search(self, value):
"""Search for operating system entity"""
view = self.navigate_to(self, 'All')
+ import time
+ time.sleep(200)
return view.search(value)
def read(self, entity_name, widget_names=None):
Apply the patch:
git apply /tmp/the.patch
From robottelo directory, run the test:
pytest tests/foreman/ui/test_operatingsystem.py::test_positive_verify_os_name
This section guides you through setting up your development environment for robottelo.
Pre-commit hooks help ensure code quality and consistency before code is committed by automatically running checks such as formatting, linting, and security checks.
-
Install Pre-Commit
Activate your virtual environment and run:
uv tool install pre-commit
-
Install Pre-Commit Hooks Run the following command to install the hooks defined in the .pre-commit-config.yaml file:
pre-commit install --install-hooks
-
Run Pre-Commit Manually To check all files manually, run:
pre-commit run --all-files
Ruff is a fast and efficient Python linter that checks your code for style violations and other issues, ensuring consistency across the project. It’s installed automatically when you install the optional dependencies. You can also install it by running the following command.
uv tool install ruff
Tip
In most cases, you won't need to run Ruff manually because it's integrated into the pre-commit hooks. Ruff will automatically apply linting and formatting whenever you commit changes.
Once installed, you can manually run Ruff on your code to check for issues.
To lint a specific file or directory, use:
ruff <file_or_directory>
For example:
ruff tests/
This will lint all Python files in the tests/ directory.
Auto-Fixing with Ruff
Ruff has an auto-fix feature that can automatically correct some common issues. To run Ruff in auto-fix mode, use:
ruff --fix <file_or_directory>
For example, to auto-fix all issues in the current directory:
ruff --fix .
Unsafe Fixes
In rare cases, you may need to apply more aggressive fixes that could potentially change the behavior of your code. To do this, you can run Ruff with the --unsafe-fixes
option:
ruff check --fix --unsafe-fixes <file_or_directory>
Warning
Use this option cautiously, as it can apply changes that might impact the functionality of your code.
Pre-Commit Hooks
Ruff is integrated with the pre-commit hooks in this project, so most of the time, simply running git commit
will automatically apply Ruff linting and formatting. The pre-commit hook will handle both safe linting and formatting, keeping your codebase clean and consistent.
In most cases, you’ll be fine relying on pre-commit to handle Ruff, and there’s no need to run Ruff manually unless you want to apply unsafe fixes.