If the check tests ran prior to the netaddr release and got a +1 any approvals will put them straight into the gate. They will only go through check first (and fail there) if they do not already have a +1 from Jenkins.
When they fail in the gate they reset all the changes behind them. When you have several changes that will fail like this you end up reseting over and over until the queue processes each one at the head of the queue, removes them, and leaves a -2 vote for the failed gate run. On Mon, Aug 31, 2015, at 08:20 AM, Kevin Benton wrote: > Why would they reset the gate? Shouldn't they just fail the check test? > > On Mon, Aug 31, 2015 at 6:18 AM, Ihar Hrachyshka <[email protected]> > wrote: > > > Hi all, > > > > I see neutron folks pushing more neutron patches into the gate. They are > > all doomed to fail until [1] is resolved. So please stop approving patches, > > we only make harm by resetting the gate, with no chance to pass it. > > > > PS: it is the same for all neutron stadium projects, so *aas and > > python-networking-* folks, please also avoid +W until the situation is > > cleared. > > > > [1]: https://bugs.launchpad.net/neutron/+bug/1490380 > > > > Ihar > > __________________________________________________________________________ > > OpenStack Development Mailing List (not for usage questions) > > Unsubscribe: [email protected]?subject:unsubscribe > > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > > > > -- > Kevin Benton > __________________________________________________________________________ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: > [email protected]?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
