On Thu, Jul 24, 2014 at 12:32 PM, Nikhil Manchanda <nik...@manchanda.me> wrote:
> > > On Wed, Jul 23, 2014 at 7:28 AM, Denis Makogon <dmako...@mirantis.com> > wrote: > > [...] > >> > >> Add Neutron-based configuration for DevStack to let folks try it > > This makes sense to tackle; now that the neutron integration pieces have > merged in Trove (yahoo!). > > However, it looks like the changes you propose in your DevStack patchset > [1] have been copied directly from the trove-integration scripts at > [2]. I have two primary concerns with this: > > a. Most of these values are only required for the trove functional > tests to pass -- they aren't required for a user install of trove with > Neutron. For such values, the trove-integration scripts seem like a > better place for this configuration. > > b. Since the trove functional tests run based on the trove-integration > scripts, what this means is that if this change is merged, this > configuration code will be run twice, once in devstack, and once again > as part of the test-init script from trove-integration. > > [1] https://review.openstack.org/#/c/108966 > [2] > https://github.com/openstack/trove-integration/blob/master/scripts/redstack#L406-427 > > > > >> Implementing/providing new type of testing job that will test on a > regular > >> basis all Trove tests with enabled Neutron to verify that all our > networking > >> preparations for instance are fine. > >> > >> The last thing is the most interesting. And i’d like to discuss it with > all > >> of you, folks. > >> So, i’ve wrote initial job template taking into account specific > >> configuration required by DevStack and Trove-integration, see [4], and > i’d > >> like to receive all possible feedbacks as soon as possible. > > > > So it looks like the test job you propose [3] is based on a current > experimental job template: gate-trove-functional-dsvm-{datastore} > [4]. Since pretty much most of it is an exact copy (except for the > NEUTRON_ENABLED bit) I'd suggest working that in as a parameter to the > current job template instead of duplicating the exact same code as part > of another job. > > Nikhil, i already did lots of refactoring inside trove.yaml (see patchset [1] and its dependent patchset). The same thing i'm going to do I know that there's lots of duplications, just wanted to describe complete template. The actual question is - Is given template is correct? Would it work with trove-integration and with pure devstack in the nearest future? P.S. I've got only basic knowledge about jenkins jobs inside infra. [1] https://review.openstack.org/#/c/100601/ > [3] https://gist.github.com/denismakogon/76d9bd3181781097c39b > [4] > https://github.com/openstack-infra/config/blob/master/modules/openstack_project/files/jenkins_job_builder/config/trove.yaml#L30-63 > > > Thanks, > Nikhil > > _______________________________________________ > OpenStack-dev mailing list > OpenStack-dev@lists.openstack.org > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >
_______________________________________________ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
