At the moment, exceptions are logged to exceptions.log file.
A copy of the exception should be sent to the maintainer by email.

Log files are located inside the virtualenv, so each time the deployment is performed they are deleted, together with the virtualenv itself.
Log files should move moved to a safer position.

Ansible gives the error

fatal: [37.247.55.31] => Authentication or permission failure.  In some cases, you may have been able
to authenticate and did not have permissions on the remote directory. Consider changing the remote
temp path in ansible.cfg to a path rooted in "/tmp". 
Failed command was: mkdir -p $HOME/.ansible/tmp/ansible-tmp-1395869881.79-232402984760726 && echo
$HOME/.ansible/tmp/ansible-tmp-1395869881.79-232402984760726, exited with result 2

FATAL: all hosts have already failed -- aborting

When domidoo repository is cloned, the build-devbox.sh builds a development workspace.
Tests won't run because the required var/test/images directory is missing.

In production, no logging is performed.
It seems that gunicorn (currently used to serve the webapp) ignores the logging section of production.ini (see http://stackoverflow.com/questions/21927811/gunicorn-paster-log-messages-not-shown)

Database URLs are reported twice, both in the dev/prod configuration and in the fabfile.

This is a violation of the DRY principle.

An idea is to invert the position of the random value and the stub name

the build-server script should install nginx.

Tests have to be fast. Tests using mocks should be preferred on tests hitting the database.

When domidoo repository is cloned, the build-devbox.sh builds a development workspace. Yet, for the webapp to run, a directory var/images is needed, but it's missing.