This repository contatins the code to set up a virtual machine with the case study workflow for SciLuigi, runnable from within a Jupyter Notebook.

For a pre-made image of this virtual machine, in .ova format, see here.

It does also include the SciLuigi case study workflow, which can be found in this folder.

For a view-only version of the Jupyter Notebook with the full case study workflow, see here.

There are two ways to get and start using this virtual machine. You can either download a pre-made virtual machine image in the open virtualization format (.ova) that can be imported in a virtual machine player such as VMWare player or VirtualBox client.

But you can also build the image from scratch, by using the code in this github repository, given that you have installed the dependencies (Vagrant, Ansible and VirtualBox).

To install and use the pre-made virtual machine, follow the steps below:

Usage:

1. Import the .ova image into a Virtual Machine software such as Virtual box.
2. Start the virtual machine.
3. Log in with "ubuntu" and "changethis..." (including the dots)
4. Optional but highly recommended, for security reasons: Open a terminal and execute the passwd command, to immediately set a new password.
5. Click the "Open Jupyter Notebook" icon on the desktop.
6. Inside Jupyter, click: Cell > Run all cells
7. The workflow will now start.

• Vagrant
• Ansible
• VirtualBox

1. Install Virtualbox:

   sudo apt-get install virtualbox

2. Install a recent version of ansible:

   sudo apt-get install ansible/trusty-backports

   (if you ubuntu version is "trusty", otherwise, replace it with your appropriate version)

3. Install Vagrant, by first downloadng the proper .deb file from vagrantup.com

4. ... and then installing it with:

   sudo dpkg -i <deb-file>

git clone https://github.com/pharmbio/bioimg-sciluigi-casestudy.git
cd bioimg-sciluigi-casestudy

vagrant up

vagrant ssh virtualbox

• When logging in to the virtual machine, a few popups with error messages appear, like this:

This does not affect the functionality though (Reported in #1).