Run-Rstudio-Server-from-Docker

Objective:

Run RStudio Server from a Docker container and mount host directories to Docker home.

Docker is a program that allows to launch and stop multiple operating systems (called containers) on your machine (the host). Docker is designed to enclose environments inside an image / a container, thus solving dependency and reproducibility issues. All you need is an installation of Docker on your computer and the Dockerfile corresponding to the container you wish to utilize.

The container on this repository is built from the following image from DockerHub:

rocker/verse:4.1.0

Run RStudio Server:

1. Add a packages directory on the host filesystem. This folder will be used for installing R packages locally from RStudio Server.

2. From the terminal, go to the folder where the Dockerfile and the .Rprofile files are and build the container:

docker build -t rstudio_server .

The Dockerfile specifies some latex and R packages to be installed when running the container; the list can be expanded with additional packages. If the image was built correctly you should see it in the list of images on disk when you run docker images.

3. Run the container with the following command:

docker run --rm \
-p 8888:8787 \ 
-v ${PWD}/packages:/packages \
-v ${PWD}/R:/home/rstudio/R \
-e PASSWORD=password \
rstudio_server

The .Rprofile file contains the line .libPaths('/packages/'), which tells R to look for packages in packages. When the RStudio Server container is started, a volume is mounted which lets the host filesystem share the folder where the R packages are saved with the container filesystem. That way, all the packages that will be saved in the folder by the container will persist after the container is turned off. Additionally, a volume R for storing the data and the R scripts is mounted (change ${PWD} to your local directory).

3. If the container ran without errors, navigate to http://localhost:8888/, insert username rstudio and password password and you can start using Rstudio Server:

5. In RStudio Server, set the library path to packages, so that installed packages are stored locally and do not need to be re-installed every time:

.libPaths(new = "/packages")

After that, new packages can be installed via the install.packages() command. Remember to set the library path every time you run RStudio Server.

5. At this point, you can use R and Rmarkdown from RStudio Server as you usually would on your machine. You can also write and read files from your local filesystem, using the mounted volumes. Once you are done, stop the running container either via CTRL + C or docker stop.

Run script from command line:

To execute an R script from the command line, add the following line at the end of the Dockerfile

CMD ["Rscript", "${PWD}/script_name.R"]

rebuild the container and run it with:

docker run --rm \
-v ${PWD}/packages:/packages \
-v ${PWD}/R:/R \
-e PASSWORD=password \
rstudio_server

Run with proxy:

To run the container behind a corporate proxy:

1. Go to Docker Desktop under Settings -> Resources -> Proxies and input your proxy with the form

http://user:password@proxy:port

2. Add an environment variable to the Dockerfile

ENV http_proxy="http://user:password@proxy:port"
ENV https_proxy="http://user:password@proxy:port"

3. Build the container with

docker build --network=host -t rstudio_server .

Examples:

This repository contains some example files to use in Rstudio Server.

• The script analysis.R in R trains a Random Forest model using the data from winequality-red.csv and saves a csv file with the ranked variable importance in the same folder; the Rmarkdown file report.R produces a report with some plots.

• To execute the scripts from the command line, uncomment the following line in the Dockerfile:

CMD ["Rscript", "/R/render.R"]

rebuild the container and run it.