On 2015-07-14 11:12:03 -0400 (-0400), Doug Hellmann wrote: > At one point Robert proposed that we change the policy for approving > constraint changes like https://review.openstack.org/#/c/201408/ to > allow a single reviewer to approve it, assuming the tests were green. > Is that our current policy, or are we waiting for the integration of > constraints with our tox environments before deciding whether to make > that change?
Since constraints are not being applied at all to tox-based jobs currently (see the thread on dropping Python 2.6 in master branches), approving updates to them has no chance to impact tox-based jobs either positively or negatively. Also, they're really just a manifestation of how pip's resolver interprets our global-requirements.txt, which is what we should be more carefully reviewing. Right now it's serving as a mechanism to keep requirements pinned to a known-passing set in DevStack-based jobs, and that is already determined by the jobs which report on the constraints update itself. -- Jeremy Stanley __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
