(This is still very alpha...)
You will need Vagrant to run and provision the VMs.
Every VMs lives in a subdirectory which contains a Vagrantfile.
The Vagrantfiles only differ by the definition of the base system. Vagrant then uses the ansible playbook in the ansible folder to provision the VMs.
To run one of the VMs you enter the corresponding directory and run vagrant up. You can enter the VM using vagrant ssh. Every VM exposes port 80 on the host port 8080. You can halt the VM using vagrant halt, you can destroy the VM by using vagrant destroy.
The tests should be idempotent, so running vagrant provision after running vagrant up should not give another result.
There is a Makefile that runs vagrant up and vagrant destroy for everyone of the boxes and saves the output to logfiles.
To run the ansible recipies without vagrant directly on the local machine, use:
ansible-playbook --connection=local -i ansible/inventory.yml ansible/playbook.yml
By default the repository tests the unstable repositories (base https://download.opensuse.org/repositories/home:/sebix:/intelmq/:unstable/). Set the environment variable intelmq_vagrant_test_stable=yes to test the stable repositories instead (base https://download.opensuse.org/repositories/home:/sebix:/intelmq/).
Only implemented for CentOS and Debian 10.