[openstack-dev] [all][infra] Constraints are ready to be used for tox.ini

Andreas Jaeger aj at suse.com
Fri Aug 12 11:54:21 UTC 2016


On 08/12/2016 01:44 PM, Ihar Hrachyshka wrote:
> Andreas Jaeger <aj at suse.com> wrote:
> 
>> On 08/12/2016 12:13 PM, Ihar Hrachyshka wrote:
>>> Jeremy Stanley <fungi at yuggoth.org> wrote:
>>>
>>>> On 2016-08-11 20:19:51 +0200 (+0200), Ihar Hrachyshka wrote:
>>>>> Do I read it right that we can now use constraints for post queue too
>>>>> (releasenotes, cover, venv targets)?
>>>>
>>>> Yes, that was the hardest part to get working, but thanks to
>>>> tireless efforts on the part of a number of people that has now been
>>>> fixed and tested.
>>>
>>> Yay! Kudos to everyone who made it happen!
>>>
>>> I posted a bunch of patches for neutron repos to validate it works:
>>> https://review.openstack.org/#/q/I02b28d3b354c3b175147c5be36eea4dc7e05f2a3,n,z
>>>
>>
>> The real validation is checking the results of post, tag, and release
>> queue...
>>
>> I don't expect surprises in check/gate queue…
>>
> 
> Yeah, we will need to land those to check post queue. But at least we
> can validate that constraints are correctly applied for where we execute
> the targets in check queue (f.e. tox -e releasenotes). I checked
> releasenotes target in all neutron repos with those patches, and at
> least os-client-config is now not pulling the latest version, but the
> one in current upper-constraints.txt file, so it’s already something.

glad to hear.

>> One caveat: Constraints cannot be used for jobs running on long-lived
>> nodes unless special care is taken - like I did for translation jobs,
> 
> I probably lack some knowledge here. What are those jobs, and how do I
> detect them?

Sorry, that was more for completeness for the rest of the infra team.

The long-lived nodes have special purposes and you're not messing with
them directly,

Andreas
-- 
 Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
  SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
   GF: Felix Imendörffer, Jane Smithard, Graham Norton,
       HRB 21284 (AG Nürnberg)
    GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126




More information about the OpenStack-dev mailing list