Open Stack

Le 24/04/2015 23:19, Matt Riedemann a écrit :
>
>
> On 4/21/2015 2:00 PM, Andrew Laski wrote:
>> It's been a long road but due to the hard work of bauzas and melwitt the
>> cells Tempest check job should now be green for changes that don't break
>> cells.  The job has been red for a long time so it's likely that people
>> don't think about it much.  I would ask that until we can get the
>> confidence to make it voting please take notice when it's red and
>> investigate or bring it to the attention of one of us in 
>> #openstack-nova.
>>
>> __________________________________________________________________________ 
>>
>> 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
>>
>
> I opened a couple of new bugs today since I had a cells job failure in 
> a change that wasn't related to cells (was just adding some debug 
> logging somewhere else).
>
> 1. https://bugs.launchpad.net/nova/+bug/1448316
>
> That looks like a legit race failure in the cells job only.

As it is related to an instance destroy method in a negative patch, I 
think it's related to the race condition we discovered and that Andrew 
is trying to fix in https://review.openstack.org/#/c/177356/

More to investigate tho.

>
> 2. https://bugs.launchpad.net/nova/+bug/1448302
>
> This is more cosmetic than anything, it doesn't appear to be related 
> to anything functionally breaking.  We should get the trace cleaned up 
> though since it makes debugging the cells job failures harder.
>

Again, we need to check if all of that is not just due to the race 
condition I said above.

-Sylvain