[openstack-dev] [QA][Tempest] Drop javelin off tempest

Daniel Mellado daniel.mellado.es at ieee.org
Thu Dec 10 18:32:07 UTC 2015


Hi All,

comments inline

El 10/12/15 a las 17:17, Matthew Treinish escribió:
> On Thu, Dec 10, 2015 at 11:15:06AM +0100, Daniel Mellado wrote:
>> Hi All,
>>
>> In today's QA meeting we were discussing about dropping Javelin off
>> tempest if it's not being used anymore in grenade, as sdague pointed
>> out. We were thinking about this as a part of the work for [1], where we
>> hit issue on Javelin script testing where gate did not detect the
>> service clients changes in this script.
> So the reason we didn't remove this from tempest when we stopped using it as
> part of grenade is at the time there were external users. They still wanted to
> keep the tooling around. This is why the unit tests were grown in an effort to
> maintain some semblance of testing after the grenade removal. (for a long time
> it was mostly self testing through the grenade job)
About the unit tests, then the option would be for them to use real
clients instead of the Mock ones. Would that work for you?
>
>> Our intention it's to drop the following files off tempest:
>>
>>   * tempest/cmd/javelin.py
>>     <https://review.openstack.org/#/c/254274/3/tempest/cmd/javelin.py>
>>   * tempest/cmd/resources.yaml
>>     <https://review.openstack.org/#/c/254274/3/tempest/cmd/resources.yaml>
>>   * tempest/tests/cmd/test_javelin.py
>>     <https://review.openstack.org/#/c/254274/3/tempest/tests/cmd/test_javelin.py>
>>
>> Before doing so, we'd like to get some feedback about out planned move,
>> so if you have any questions, comments or feedback, please reply to this
>> thread.
> You should not just delete these files, there were real users of it in the past
> and there might still be. If you're saying that javelin isn't something we can
> realistically maintain anymore (which I'm not sure I buy, but whatever) we 
> should first mark it for deprecation and have a warning print saying it will be
> removed in the future. This gives people a chance to stop using it and migrate
> to something else. (using ansible would be a good alternative)
Then should we just mark it as deprecate, and if so, how? I'll put a new
commit marking it as deprecate with a different commit message

Thanks
>
>
> -Matt Treinish
>
>> ---
>> [1]
>> https://review.openstack.org/#/q/status:open+project:openstack/tempest+branch:master+topic:bp/consistent-service-method-names,n,z
>>
>>
>>
>> __________________________________________________________________________
>> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20151210/2092dd9d/attachment.html>


More information about the OpenStack-dev mailing list