Thanks. I didn't realize it was already breaking everything. I thought it
might have been stuck in requirements bump patch somewhere.

>Thats fixed in 1.1.0. So you should be able to unwind that. If you need a
short term workaround, its in mock.mock now.

Yeah, I know it's fixed, I reported the upstream bug quite some time ago.
:) I just felt the need to explain because you implied that the Neutron
reviewers missed something when the patch that introduced that was very
intentional due to the many times that bug has caused non-deterministic
failures.

On Fri, Jul 10, 2015 at 1:28 AM, Robert Collins <[email protected]>
wrote:

> On 10 July 2015 at 20:23, Kevin Benton <[email protected]> wrote:
> > How do we test to see what is failing in each project with the new
> version?
>
> Look at any CI failure in the last 5 hours or so.
>
> Or run tox :).
>
> > Also, I'm responsible for the reference to the private mock method in
> > Neutron. That particular reference is to prevent people from patching the
> > same target twice because mock.patch.stopall() unwinds patches in a
> > non-deterministic order in < py34 which leads to leftover monkey patches.
> > These caused completely unrelated unit tests to randomly and inexplicably
> > explode later. More details here:
> >
> https://github.com/openstack/neutron/commit/1b60df85ba3ad442c2e4e7e52538e1b9a1bf9378
>
> Thats fixed in 1.1.0. So you should be able to unwind that. If you
> need a short term workaround, its in mock.mock now.
>
> -Rob
>
> --
> Robert Collins <[email protected]>
> Distinguished Technologist
> HP Converged Cloud
>
> __________________________________________________________________________
> 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

Reply via email to