[openstack-qa] [Nova] High priority bugs preventing full Tempest gate

Sean Dague sdague at linux.vnet.ibm.com
Wed Dec 12 00:08:42 UTC 2012


On 12/11/2012 06:56 PM, Jay Pipes wrote:
> Hi Vish,
>
> The QA team was hoping to focus the Nova core members' attention to a
> few bugs in Nova that are causing failures in the full Tempest run --
> currently gating only the Tempest project trees, but intended to gate
> Nova, Glance, Keystone, Quantum and Cinder in the future.
>
> The first bug has not been prioritized yet, but we hope it can be, and
> some nova-core resources need to be assigned to identify the source
> (race condition?) of the failures:
>
> https://bugs.launchpad.net/nova/+bug/1079687

https://jenkins.openstack.org/job/periodic-tempest-devstack-vm-check-hourly/696/console 
is a recent fail. Was actually chatting with clarkb just now to make it 
so failed vms can be kept around longer to debug.

> Unfortunately, the above bug will be masked by the following bug:
>
> https://bugs.launchpad.net/nova/+bug/1088584

Actually, I think we got that fixed yesterday after some nova and 
devstack changes. So the fails aren't masked.

The nova race is really hard to reproduce though, I think in reality 
most of our dev machines are just too much faster than the CI 
infrastructure. So running the tests inside a 4 vcpu 4 GB guest (making 
the devstack creates on qemu) is probably needed here.

If anyone manages to get to the bottom of this one, you'll be the hero 
of many.

	-Sean

-- 
Sean Dague
IBM Linux Technology Center
email: sdague at linux.vnet.ibm.com
alt-email: sldague at us.ibm.com




More information about the openstack-qa mailing list