The goal is to write new request specs to test that Workhorse caching works (using Etag) as expected. See https://gitlab.com/gitlab-org/gitlab-workhorse/merge_requests/270#note_84793971. Quoting @jacobvosmaer\-gitlab: > For instance we could make the same request two times in a row. Can we reliably see, perhaps in the logs, that the second request didn't reach gitlab-rails? This could just be new request specs since we're actually using Workhorse in the `gitlab-rails` test suite: For every route at https://gitlab.com/gitlab-org/gitlab-ce/blob/master/lib/gitlab/etag_caching/router.rb: 1. Makes a first request 1. Checks that the request returns a `200` 1. Makes the same request again 1. Checks that the request returns a `304`