Sign inSign up

wolfyuan/dolphin

By wolfyuan

ā€¢Updated over 2 years ago

Multi usage container image for hosting various applications

Image
0

10K+

wolfyuan/dolphin repository overview

ā šŸ¬ Dolphin

ā All-in-one solution for hosting your code

Dolphin is still under heavy development, please create an issue if you face any bugs or request a feature

Dolphin is a cool container image that prepared everything for you!

āœØ Features:

  • Container running with s6-overlay
  • Modular design
  • Support a wide variety of language
  • Multi-base image and CPU architecture
  • Code runs in a rootless process

ā šŸƒ Running image

docker run --name dolphin-box -d \
  -e GIT_URL="https://domain.com" \
  --restart unless-stopped \
  docker.io/wolfyuan/dolphin:node

Command breakdown: -e GIT_URL="https://domain.com" - Environment variables for controlling container, see hereā  for a full list

You can add multiple variables, like:

-e GIT_URL="https://domain.com" -e SYSTEM_PACKAGES="automake"

So the whole command will be

docker run --name dolphin-box -d \
 -e GIT_URL="https://domain.com" \
 -e SYSTEM_PACKAGES="automake" \
 --restart unless-stopped \
 docker.io/wolfyuan/dolphin:node

--restart unless-stopped - Restart your container when it fails or you reboot your machine unless you stop it by yourself

docker.io/wolfyuan/dolphin:node - Container image, where node represents to image that's designed for this language, see hereā  for full list

ā šŸ“¦ļø Container images

Most image supports arm based CPU unless mentioned in the note.

Image nameDescriptionVariantsTagArchitecture
Node.jsImage with necessary software to run Node.js softwarealpine, ubuntunodeamd64, arm64 (ubuntu), arm/v7 (ubuntu)
BaseBase image, you should NOT use this image when runningalpine, ubuntubaseamd64, arm64,arm/v7
ā Tagging format

Images are tagged with this format: [Image Tag]-[Image Variants]

For example, the node image with ubuntu base will be: node-ubuntu

If you don't specify image variants, it will pull the ubuntu base

For example, pulling the node tag will pull the node-ubuntu image

ā šŸŒŽļø Environment variables and workflow

ā Applies to every image:

Environment variables:

Environment variables nameDescription
SYSTEM_PACKAGESAdditional packages to install when the container starts
PACKAGES_LOCATIONSystem packages cache location
PACKAGES_DISABLE_CACHEDisable system package cache (Set to any value to disable)
GIT_URLGit repo to clone when the container starts

Workflow:

  1. Container starts
  2. Prepare runner folders
  3. Add additional system packages
  4. Clone git repo with GIT_URL
šŸ“ Note
You can install additional system packages via the SYSTEM_PACKAGES variable
For example, installing automake on the base system, just set SYSTEM_PACKAGES with the value automake
When installing multiple system packages, separate them by spaces, for example, automake cmake
šŸ“ Note
Alpine-based image will use apk to install packages
Ubuntu-based image will use apt-get to install packages
šŸ“ Note
Downloaded packages will be cached to /home/dolphin/packages if PACKAGES_LOCATION isn't changed
ā Image-specific options
node

Environment variables:

Environment variables nameDescription
NODE_VERSIONNode.js version to install, overrides .nvmrc
NODE_START_SCRIPTScript to run when container finished running init script

Workflow:

  1. Container start
  2. Base image initializeā 
  3. Install Node.js via NODE_VERSION or .nvmrc in your project
  4. Install npm packages via the detected package manager
  5. Run build script via detected package manager if presents
  6. Start the Node.js process via NODE_START_SCRIPT or script in package.json via the detected package manager
šŸ“ Note
Supported package managers: pnpm, yarn, npm
šŸ“ Note
In package manager: pnpm > yarn > npm
In node version: NODE_VERSION > .nvmrc
python

Environment variables:

Environment variables nameDescription
PYTHON_VERSIONPython version to install, overrides .python-version
PYTHON_BUILD_SCRIPTScript to run after installing Python packages
PYTHON_START_SCRIPTScript to run when container finished running init script
PYTHON_START_FILEScript to run when container finished running init script

Workflow:

  1. Container start
  2. Base image initializeā 
  3. Install Python via PYTHON_VERSION or .python-version in your project
  4. Install python packages via pip
  5. Start the Python process via PYTHON_START_SCRIPT and the script in PYTHON_START_FILE with command python3
āš ļø Warning
Due to how pyenv switch Python version, it's recommended to presist your container storage, see Presist data inside containerā  for more info
šŸ“ Note
When PYTHON_START_SCRIPT and PYTHON_START_FILE both exists, it will run PYTHON_START_SCRIPT first, and run PYTHON_START_FILE after previous script has finished
šŸ“ Note
In Python version: PYTHON_VERSION > .python-version

ā šŸ’¾ Persist data inside container

You can mount two types of directory to the Dolphin container:

  • /home/dolphin: Contain your code (space), and other runtime binaries installation like nvm, pyenv

    Use this when you want to save more bandwidth

  • /home/dolphin/space

    Use this when you want to prevent rebuilding your project

To mount the directory, append these arguments when starting the Docker container:

-v /path/to/your/host/folder:/home/dolphin

or

-v /path/to/your/host/folder:/home/dolphin/space
āš ļø Warning
You can't mount two directories at the same time!

ā šŸŽ© RedHat Podman support

It SHOULD work, maybe

ā šŸ“½ļø Graphics card support

I haven't tried it, but your application will be run in a rootless environment, so it might not work.

Tag summary

Content type

Image

Digest

sha256:8af02ab12ā€¦

Size

253.6 MB

Last updated

over 2 years ago

docker pull wolfyuan/dolphin:python