conman-tool 1.2.1

Description:

conmantool 1.2.1

CONtainer MANager : conman

Description
Conman is a tool to manage containers. It is designed to be used with docker, docker-compose et VScode devcontainer tool.

Requirements

Unix (Linux, macOS) based operating system (WSL workaround for Windows)
A container engine ans its composer tool:

Docker 24.0+ and Docker-compose 2.20+
Podman 4.0+ and Podman-compose 1.0.6+


Python 3.8+ and <3.11
Python modules :

pip 21.0.1+ (Need setuptools integration)
pyyaml 5.4.1+




Installation


Install conman using pip:
pip install conman-tool

You can also install conman from source, tcd he current stable realease is 1.2.0:
pip install git+https://github.com/elmokulc/[email protected]




Usage
When starting a new project


Go to you project directory:
mkdir my_project
cd my_project



Create a conman configuration file by running:
conman init

This will create a directory named .conman within the configuration file conman-config.yml.
$ ➜ ~/my_project $ conman init
$ ➜ ~/my_project $ tree -a
.
└── .conman
└── conman-config.yml

1 directory, 1 file



Edit the conman-config.yml configuration file to fit your needs.


As example:
# Images Settings
images:
root:
generate: true
name: BigFoot
tag: latest
from_image:
name: ubuntu
tag: '20.04'
conda_environment:
directory: /opt/conda
env_name: myenv
environment_file: ./environment.yml
extra_instructions:
- "RUN echo 'export $PYTHONPATH=/python_modules:$PYTHONPATH' >> ~/.bashrc"
user:
extra_instructions:
- "RUN sudo mkdir -p /python_modules"
- "RUN sudo chown -R $USER:$USER /python_modules"
- "RUN echo 'export $PYTHONPATH=/python_modules:$PYTHONPATH' >> ~/.bashrc"

# Container Settings
container:
engine: docker # docker or podman
compose:
service_name: main
volumes:
- ..:/workspace
- path_module1/module1:/python_modules/module1
- path_module2/module2:/python_modules/module2
devcontainer:
customizations:
vscode:
settings: {}
extensions:
- ms-python.python
- ms-python.vscode-pylance
- ms-toolsai.jupyter
graphical:
protocol: x11
gpu:
manufacturer: nvidia
count: 1



By running:
conman build

Conman will generate a docker-compose.yml file and a Dockerfile.user, enventually: a .devcontainer directory within a devcontainer.json file, a Dockerfile.root and a conda environment.yml file, all of this according to your configuration file: .conman-config.yml .
Here is an example of the output from the previous configuration file:
$ ➜ ~/my_project $ conman build
Building...
Directory .devcontainer created
Creating devcontainer.json file...
Executing xhost +local: ...
Appending volumes for display configuration...
-> Add volume: /tmp/.X11-unix:/tmp/.X11-unix:rw
-> Add volume: /home/vscode/.Xauthority:/home/vscode/.Xauthority:rw
-> Display forwading activated
-> Add volume: ..:/workspace
-> Add volume: path_module1/module1:/python_modules/module1
-> Add volume: path_module2/module2:/python_modules/module2
-> GPU activated
--- Build user Dockerfile ---
Adding user extra instructions to Dockerfile...
Generated Dockerfile.user at: ~/my_project/.devcontainer/Dockerfile.user
--- Build root Dockerfile ---
Creating conda env file at: ~/my_project/.conman/environment.yml
Adding root extra instructions to Dockerfile...
Generated Dockerfile.root at: ~/my_project/.conman/Dockerfile.root
Project Building done successfully

The project directory will now look like this:
$ ➜ ~/my_project $ tree -a
.
├── .conman
│ ├── conda
│ │ └── environment.yml
│ ├── conman-config.yml
│ └── scripts
│ ├── initializeCommand.sh
│ ├── onCreateCommand.sh
│ ├── postCreateCommand.sh
│ ├── postStartCommand.sh
│ └── updateContentCommand.sh
├── .devcontainer
│ ├── build_root_img.sh
│ ├── devcontainer.json
│ ├── docker-compose.yml
│ ├── Dockerfile.root
│ └── Dockerfile.user
└── .env

4 directories, 13 files



When working on an existing project
In case you want to work on an existing project, you can use the conman update command by running:
conman update

This command will regenerate user dependant files such as Dockerfile.user, docker-compose.yml or devcontainer.json:
$ ➜ ~/my_project $ conman update
Updating conman build...
Creating devcontainer.json file...
[...]
Appending volumes for display configuration...
-> Add volume: /tmp/.X11-unix:/tmp/.X11-unix:rw
-> Add volume: /home/vscode/.Xauthority:/home/vscode/.Xauthority:rw
-> Display forwading activated
-> Add volume: ../:/workspace
--- Build user Dockerfile ---
Adding conda environment to Dockerfile...
No extra instructions in user image
Generated Dockerfile.user at: /workspaces/conman/myproject/.devcontainer/Dockerfile.user

Help
All available commands can be listed by running:
conman --help

if you need some details about a specific command, you can run:
conman <command> --help


Root image generation
In case you need to generate a root image, conman will generate a build script for you name build_root_img.sh in the same folder that Dockerfile.root.
Unfortunalety, you will need to set execution permissions and run this script manually.
To set proper permissions and run the script:
chmod +x build_root_img.sh && ./build_root_img.sh

Troubleshoting


According to your system configuration and yours permissions, the location of the entry point of conman may change.
You may need to update your $PATH variable.
The command:
pip show conman

will give you the following output within the location of the conman package:
Name: conman
Version: [****]
Summary: [****]
Home-page: [****]
Author: [****]
Author-email: [****]
License: [****]
Location: /home/<username>/.local/lib/python3.8/site-packages
Editable project location: [****]
Requires: [****]

In this example, the location of the entrypoint to be add will be /home/<username>/.local/bin.
You can then add the following line to your .bashrc or .zshrc file:
export PATH=$PATH://home/$USER/.local/bin

or by running:
echo 'export PATH=$PATH:/home/'$USER'/.local/bin' >> ~/.bashrc

This will add the conman executable to your $PATH variable.


Contributing
If you want to contribute to this project, please read the CONTRIBUTING.md file.

License

For personal and professional use. You cannot resell or redistribute these repositories in their original state.

Share

• Share on Facebook
• Share on Twitter