os-vif log messages corrupt and pollute test output

[Jay Pipes]

When running unit tests with tox, os-vif logging pollutes the output of the testr runner and on multi-core machines, nearly always corrupts the output stream like so:

http://paste.openstack.org/show/562213/

os-vif logging setup should be examined to ensure it is being processed like all other modules that behave properly with testr.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/370908

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/371785

[OpenStack Infra (hudson-openstack)]

Change abandoned by Pushkar Umaranikar (<email address hidden>) on branch: master
Review: https://review.openstack.org/370908

[Matt Riedemann (mriedem)]

Nova patch: https://review.openstack.org/#/c/374233/

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/374233
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=735f710ae0c0e0037970796beacd3527e2a8ca8a
Submitter: Jenkins
Branch: master

commit 735f710ae0c0e0037970796beacd3527e2a8ca8a
Author: Sean Dague <email address hidden>
Date: Wed Sep 21 11:11:21 2016 -0400

    move os_vif.initialize() to nova-compute start

    os_vif.initialize() was previously executed during module load. This
    means it was entirely possible that it was run before things like
    logging were actually set up in the expected way. Move this back into
    execution time instead of load time to ensure that logging is actually
    setup.

    Changes need to be made to tests which make assumptions about os_vif
    objects to manually initialize os_vif when it will be used. os_vif
    objects can't be created until it is initialized, so some delayed
    object creation is also done in test_vif.py.

    Closes-Bug: #1615676

    Change-Id: I89fe5c5b3d762f3a3238b587685df85d15ee56c4

[OpenStack Infra (hudson-openstack)]

Change abandoned by Pushkar Umaranikar (<email address hidden>) on branch: master
Review: https://review.openstack.org/371785
Reason: The issue is addressed in https://review.openstack.org/#/c/374233/. Hence, abandoning this change.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/nova 15.0.0.0b1 development milestone.