Getting docker

Docker is a software suite to run software in a platform independent way akin to virtual machines.

Learn how to download and install it on your system (admin rights required): https://www.docker.com/get-started

CbmRoot docker environment

There are following containers in this package build upon each other in the following order:

• fairsoft
• fairroot
• cbmroot
• jupyter notebook with cbmroot (cbmroot-notebook)
• jupyterhub with cbmroot-notebook enabled (cbmroot-jupyterhub)

To run any of these containers, you need to initialize them first, e.g. run the following commands:

docker-compose up -d cbmroot

The pre-build container images will then be downloaded. The containers will run and keep any files you create inside it until you won't remove them from your system with:

docker-compose down -v

The software is installed in /opt within containers, e.g.:

• /opt/fairsoft
• /opt/fairroot
• /opt/cbmroot

To enter the container environment, run

docker-compose exec cbmroot bash

Running interactive ROOT analysis in JupyterLab

Run the following command in your terminal

docker-compose up cbmroot-jupyterhub

You can navigate to http://localhost:8000 in your browser.

Authorize yourself with user dummy and password dummy.

CbmRoot development environment

If you are a CbmRoot power-user or a developer you might want to check out the cbmroot-dev container, also known as cbmdock.

When creating the container for the first time, run:

export UID
docker-compose up cbmroot-dev

and wait until the development version of CbmRoot will be compiled.

You can change GIT_REPO and GIT_TAG variables in the docker-compose.yml to compile the code from destination and git tag (or commit) of your choise, respectively.

The files will be shared with the container and your computer (located in code/cbmroot), so you can edit them conveniently with your favorite code editor.

You either can enter the container as usual or ssh into it by issuing the command:

ssh -Y cbmdock@localhost -p 2222

Troubleshooting

Visualization

If you have no visualization you need to install the x11 on your host machine.

In macOS it is XQuartz. Run it, go to preferences -> security and enable connections from network clients.

Then from your terminal execute:

xhost + localhost
defaults write org.macosforge.xquartz.X11 enable_iglx -bool true

In linux try to export the environment variable before running visualisations:

export LIBGL_ALWAYS_INDIRECT=1

More info can be found here https://twiki.cern.ch/twiki/bin/view/CMSPublic/WorkBookFireworksHowToFix#OpenGL_and_X11_problems

Image is being built rather than downloaded:

If this happens to any of non-development containers, run for example:

docker-compose pull cbmroot

Installing docker in Linux

curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh

Read more at https://www.docker.com/get-started