On 9/9/2018 1:11 PM, Ghanshyam Mann wrote:
Yeah, Tempest would not fit as best location for such tagging or whitelist. I
think nova may be better choice if nothing else.
OK I've thrown this on the nova ptg etherpad agenda [1] for a misc item
to talk about.
[1] https://etherpad.openstack.org
On Sat, 08 Sep 2018 08:28:06 +0900 Matt Riedemann
wrote
> On 9/7/2018 10:25 AM, William M Edmonds wrote:
> > The concern that I have with whitelisting in a given CI is that it has
> > to be done independently in every compute driver CI. So while I agree
> > that it won't be e
On 9/7/2018 10:25 AM, William M Edmonds wrote:
The concern that I have with whitelisting in a given CI is that it has
to be done independently in every compute driver CI. So while I agree
that it won't be easy to maintain tagging for compute driver on the
tempest side, at least that's one place
Ghanshyam Mann wrote on 09/07/2018 02:18:13 AM:
snip..
> neutron-tempest-plugin or other service test you can always avoid to
> run with regex. And i do not think compute negative or DB test will
> take much time to run. But still if you want to avoid to run then, I
> think it is easy to mainta
;OpenStack Development Mailing List \\"
Date: 09/07/2018 02:18 PM
Subject: Re: [openstack-dev] [tempest][CI][nova compute] Skipping
non-compute-driver tests
On Fri, 07 Sep 2018 04:41:32 +0900 Eric Fried
wrote
> Jichen-
>
> That pa
d an
> > experimental patch [1] with an effectively
> >
> > From: Eric Fried
> > To: "OpenStack Development Mailing List (not for usage questions)"
> >
> > Date: 09/04/2018 09:35 PM
> > Subject: [openstack-dev] [tempest][CI][nova comput
mental patch [1] with an effectively
>
> From: Eric Fried
> To: "OpenStack Development Mailing List (not for usage questions)"
>
> Date: 09/04/2018 09:35 PM
> Subject: [openstack-dev] [tempest][CI][nova compute] Skipping
> non-compute-driver tests
>
>
2018 09:35 PM
Subject:[openstack-dev] [tempest][CI][nova compute] Skipping
non-compute-driver tests
Folks-
The other day, I posted an experimental patch [1] with an
effectively
empty ComputeDriver (just enough to make n-cpu actually start) to see
how much
Folks-
The other day, I posted an experimental patch [1] with an effectively
empty ComputeDriver (just enough to make n-cpu actually start) to see
how much of our CI would pass. The theory being that any tests that
still pass are tests that don't touch our compute driver, and are
therefore
