Testing process inquiry (Research study)

Hello,

I am a research assistant at the University of
Nebraska-Lincoln. I am working on continuous integration environments. My
current project involves Rails and Travis.

I looked at the Rakefiles, .travis.yml and ci/travis.rb
files and I noticed that some tests are run twice in a single build: once all
together in a single process and then in isolation. For instance, action pack in
this build. I
would like to understand why you are doing this. I suspect you want to catch test
dependency issues. I would appreciate it
if you could confirm my thoughts or give me any detail about this process.

Thank you,

Matias Waterloo