This repo contains information to setup a dockerized instance with R, Rstudio, Shiny, Radiant, Python, and JupyterLab
To use the docker images you first need to install Docker
- For Mac: https://docs.docker.com/docker-for-mac/
- For Windows: https://docs.docker.com/docker-for-windows/
- For Linux: https://docs.docker.com/engine/installation/
After installing Docker, check that it is running by typing docker --version in a terminal. This should return something like the below:
docker --version
Docker version 18.09.0, build 4d60db4The full rsm-msba setup uses Docker Compose so also check this is available by typing docker-compose --version in a terminal. This should return something like the below:
docker-compose --version
docker-compose version 1.21.1, build 5a3f1a3On windows please install Git Bash:
http://www.techoism.com/how-to-install-git-bash-on-windows/
For detailed install instructions on Windows see install/rsm-msba-windows.md
For detailed install instructions on macOS see install/rsm-msba-macos.md
To jump straight in and run the main application run the command below on macOS:
docker run --rm -p 8080:80 -p 8787:8787 -p 8989:8888 -v ~:/home/rstudio vnijs/rsm-msbaFor Windows run the command below:
docker run --rm -p 8080:80 -p 8787:8787 -p 8989:8888 -v c:/Users/$USERNAME:/home/rstudio vnijs/rsm-msbaPerhaps even easier, you can start the rsm-msba container on macOS using launch-mac.command and on Windows using launch-windows.sh. To get these files download the repo https://github.com/radiant-rstats/docker or clone the repo using git clone https://github.com/radiant-rstats/docker.git is you have git installed. To run the script on Windows you will need Git Bash installed
as referenced above.
Another alternative approach is to use docker-compose and the command below after cloning the repo:
docker-compose -f ./rsm-msba/docker-rsm-msba.yml upNote: For Windows you may need to change the path in the volumes: section to c:/Users/$USERNAME
For more information about running the radiant application see radiant/README.md
For more information about running the rsm-msba application see rsm-msba/README.md
You probably don't want to run this image by itself. It is used in the radiant and rsm-msba application (see below). To build a new container based on r-bionic add the following at the top of your Dockerfile
FROM vnijs:docker-bionic
To build r-bionic yourself use:
docker build -t $USER/r-bionic ./r-bionicPush to docker hub:
sudo docker login
docker push $USER/r-bionicThe second image builds on r-bionic and adds radiant and required R-packages. To build a new container based on radiant add the following at the top of your Dockerfile
FROM vnijs:radiant
To build radiant yourself use:
docker build -t $USER/radiant ./radiantPush to docker hub:
sudo docker login
docker push $USER/radiantAdd the following to .Rprofile in your home directory
options(radiant.ace_vim.keys = FALSE)
options(radiant.maxRequestSize = -1)
# options(radiant.maxRequestSize = 10 * 1024^2)
options(radiant.report = TRUE)
# options(radiant.ace_theme = "cobalt")
options(radiant.ace_theme = "tomorrow")
# options(radiant.ace_showInvisibles = TRUE)The third image builds on the radiant image and adds python and Jupyter. To build a new container based on rsm-msba add the following at the top of your Dockerfile
FROM vnijs:rsm-msba
To build rsm-msba yourself use:
docker build -t $USER/rsm-msba ./rsm-msbaPush to docker hub:
sudo docker login
docker push $USER/rsm-msbaThe rsm-msba directory also contains a docker-compose file that pulls in a postgres image and database admin tool adminer. To run the full application use the command below.
docker-compose -f ./rsm-msba/docker-rsm-msba.yml upIf you want to install an R-package, e.g., fortune, in a way that persists when using the container again, use the command below. This will install the package and create a personal directory for future package installs. You will only need to add the lib = Sys.getenv("R_LIBS_USER") argument once to generate the personal directory.
install.packages("fortunes", lib = Sys.getenv("R_LIBS_USER"))
If you want to install a python package, e.g., redis, in a way that persists when using the container again, use the command below from the Jupyter (or Rstudio) terminal. This will install the package and create a personal directory for future package installs.
pip3 install -U "redis"
To stop (all) running containers use:
docker kill $(docker ps -q)If the build fails for some reason you can access the container through the bash shell using to investigate what went wrong:
docker run -t -i $USER/rsm-msba /bin/bashTo remove an existing image use:
docker rmi --force $USER/rsm-msbaTo remove stop all running containers, remove unused images, and errand docker processes use the dclean.sh script
./dclean.shCheck the disk space used by docker images
docker ps -sdocker system dfOn mac you can use the commands below to push your custom image to docker hub:
sudo docker login
docker push $USER/rsm-msbaAdd the following to .Rprofile in your home directory
options(radiant.ace_vim.keys = FALSE)
options(radiant.maxRequestSize = -1)
# options(radiant.maxRequestSize = 10 * 1024^2)
options(radiant.report = TRUE)
# options(radiant.ace_theme = "cobalt")
options(radiant.ace_theme = "tomorrow")
# options(radiant.ace_showInvisibles = TRUE)Shiny and Shiny Server are registered trademarks of RStudio, Inc. The use of the trademarked terms Shiny and Shiny Server and the distribution of the Shiny Server through the images hosted on hub.docker.com has been granted by explicit permission of RStudio. Please review RStudio's trademark use policy and address inquiries about further distribution or other questions to permissions@rstudio.com.
Jupyter is distributed under the BSD 3-Clause license (Copyright (c) 2017, Project Jupyter Contributors)
Thanks to Ajar Vashisth for helping me get started with Docker and Docker Compose