Docker Interactive Shell Runner
Docker Interactive Shell Runner
docker-shell command makes it quick and easy to start an interactive shell inside a Docker container, with the following features:
Both Windows and Linux containers are supported.
The current working directory is automatically bind-mounted into the container and set as the container's working directory.
When running Linux containers under Linux host systems, host networking mode is enabled automatically to eliminate the need for exposing individual ports or port ranges.
GPU support is automatically enabled when running GPU-enabled Linux containers under Linux host systems with the NVIDIA binary drivers and the NVIDIA Container Toolkit installed.
Short aliases are provided for running popular shells:
See the developer-images repository for examples of container images designed specifically for use with docker-shell.
One of Docker's supported operating systems (with the additional minimum requirement of version 1809 or newer for Windows systems):
- Linux: 64-bit version of one of Docker's supported Linux distributions (CentOS 7+, Debian 7.7+, Fedora 26+, Ubuntu 14.04+)
- macOS: macOS 10.10.3 Yosemite or newer running on 2010 or newer model Mac hardware
- Windows 10: 64-bit Windows 10 Pro/Enterprise/Education version 1809 or newer, or 64-bit Windows 10 Home version 2004 or newer with WSL2
- Windows Server: Windows Server 2019 / Windows Server version 1809 or newer
An appropriate Docker installation for your operating system:
Python version 3.5 or newer
docker-shell, run the following command:
# This may need to be prefixed with sudo under Linux and macOS pip3 install docker-shell
To start a GNU Bash shell in an official Python container under Windows, macOS or Linux, run either of the following commands:
# Long version docker-shell bash python # Short version dbash python
To start a Windows Command Prompt in a Windows Server Core container under Windows, run either of the following commands:
# Long version docker-shell cmd mcr.microsoft.com/windows/servercore:ltsc2019 # Short version dcmd mcr.microsoft.com/windows/servercore:ltsc2019
To start a PowerShell session in a Windows Server Core container under Windows, run any of the following commands:
# Long version docker-shell powershell mcr.microsoft.com/windows/servercore:ltsc2019 # Short versions dps mcr.microsoft.com/windows/servercore:ltsc2019 dpowershell mcr.microsoft.com/windows/servercore:ltsc2019
Passing additional flags to Docker
Any additional flags that are specified on the command-line will be propagated automatically to the underlying docker run command:
# The flags `-u 1000 --name mycontainer` will be passed directly to Docker docker-shell bash python -u 1000 --name mycontainer
Working with alias tags
When working with lengthy image tags it is often more convenient to use the docker tag command to create concise aliases, for example:
# Add a shorter tag alias for the Windows Server Core image docker tag mcr.microsoft.com/windows/servercore:ltsc2019 windows:latest # Run a Windows Command Prompt using the short tag dcmd windows
Specifying additional options using image labels
If there are bind mounts or command-line flags that you always use when running a particular image then you can specify these using image labels in your Dockerfile and
docker-shell will apply them automatically:
FROM my-base-image # The directory specified by the environment variable `SOMEVAR` will be bind-mounted irrespective of the host platform LABEL docker-shell.mounts.1="\$SOMEVAR:/data" # The named volume specified by the special `VOLUME` variable (see below for details) will be bind-mounted irrespective of the host platform LABEL docker-shell.mounts.2="\$VOLUME:/project-data" # The current user's Desktop directory will be bind-mounted irrespective of the host platform LABEL docker-shell.mounts.3="~/Desktop:/desktop" # This specific user's Desktop directory will be bind-mounted only when running on Linux hosts LABEL docker-shell.linux.mounts.1="/home/user/Desktop:/user-desktop" # This specific user's Desktop directory will be bind-mounted only when running on macOS hosts LABEL docker-shell.mac.mounts.1="/Users/user/Desktop:/user-desktop" # This specific user's Desktop directory will be bind-mounted only when running on Windows hosts LABEL docker-shell.windows.mounts.1="C:/Users/User/Desktop:/user-desktop" # The command-line flags `-u 1000` will be applied irrespective of the host platform LABEL docker-shell.args.1="-u" LABEL docker-shell.args.2="1000" # The command-line flags `--priveleged` will be applied only when running on Linux hosts LABEL docker-shell.linux.args.1="--priveleged" # The command-line flags `-m 1GB` will be applied only when running on macOS hosts LABEL docker-shell.mac.args.1="-m" LABEL docker-shell.mac.args.2="1GB" # The command-line flags `-m 2GB` will be applied only when running on Windows hosts LABEL docker-shell.windows.args.1="-m" LABEL docker-shell.windows.args.2="2GB"
Special variables available for use in labels
When specifying additional options via image labels, the following special variables are supported:
CWD: resolves to the current working directory on the host system
HOSTIP: resolves to the IP address of the host system on the local network, or
127.0.0.1if no network interfaces are available
VOLUME: resolves to the SHA-256 hash of the absolute path to the current working directory on the host system, suitable for use as a volume name when you want to associate data with a particular project but don't want to store files on the host filesystem for performance or compatibility reasons
Automatic Docker daemon selection under Windows 10
When running Docker Desktop under Windows 10, docker-shell can automatically switch between Windows containers mode and Linux containers mode based on the specified shell. (Windows containers mode is used for
powershell, whilst Linux containers mode is used for all other shells.) Because this behaviour alters system state and may be considered intrusive, it is disabled by default. To enable automatic daemon selection, set the environment variable
Copyright © 2020, Adam Rehn. Licensed under the MIT License, see the file LICENSE for details.
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for docker_shell-0.0.9-py3-none-any.whl