Open Stack

On 5/14/2015 9:50 AM, Brant Knudson wrote:
>
>
>
> On Thu, May 14, 2015 at 9:41 AM, Matt Riedemann
> <mriedem at linux.vnet.ibm.com <mailto:mriedem at linux.vnet.ibm.com>> wrote:
>
>
>
>     On 5/14/2015 5:46 AM, Sean Dague wrote:
>
>         On 05/14/2015 04:16 AM, Robert Collins wrote:
>
>             Tox 2.0 just came out, and it isolates environment variables
>             - which
>             is good, except if you use them (which we do). So everything is
>             broken.
>
>             https://review.openstack.org/182966
>
>             Should fix it until projects have had time to fix up their local
>             tox.ini's to let through the needed variables.
>
>             As an aside it might be nice to get this specifier from
>             global-requirements, so that its managed in the same place
>             as all our
>             other specifiers.
>
>
>         This will only apply to tempest jobs, and I see lots of tempest jobs
>         passing without it. Do we have a bug with some failures linked
>         because
>         of it?
>
>         If this is impacting unit tests, that has to be directly fixed
>         there.
>
>                  -Sean
>
>
>     python-novaclient, neutron and python-manilaclient are being tracked
>     against bug https://bugs.launchpad.net/neutron/+bug/1455102.
>
>     Heat is being tracked against bug
>     https://bugs.launchpad.net/heat/+bug/1455065.
>
>     --
>
>     Thanks,
>
>     Matt Riedemann
>
>
>
> Here's the fix in keystoneclient if you need an example:
> https://review.openstack.org/#/c/182900/
>
> It just added passenv =OS_*
>
> If you're seeing jobs pass without the workaround then those jobs are
> probably not running with tox>=2.0.
>
> - Brant
>
>
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>

There is a similar issue with horizon:

https://bugs.launchpad.net/horizon/+bug/1458928

It's specifically busted on stable/kilo.  I think it's not hitting on 
master because the jshint stuff has been cleaned up a bit on master and 
it's less strict in the run, which .jshintrc was handling before and why 
it fails on stable/kilo.

I'll be pushing a change soon.  We might think about capping tox<2.0 on 
stable branches though...

-- 

Thanks,

Matt Riedemann