I realized the weird unrelated failure in#1020 (seehttps://travis-ci.org/python-gitlab/python-gitlab/jobs/652064210) was a flaky test for rate limits. I'm getting it locally sometimes but only rarely.

I haven't had the time to investigate but I'll try to have a look at some point. Maybe something in the environment slows down the requests sometimes and it does not trigger 429 or something.