[openstack-dev] [nova][stable] Circular dependancy to resolve

Matt Riedemann mriedem at linux.vnet.ibm.com
Mon Dec 21 17:29:28 UTC 2015



On 12/21/2015 9:26 AM, Matt Riedemann wrote:
>
>
> On 12/20/2015 5:35 PM, Tony Breeds wrote:
>> Hi all (Actually I'mm really looking a Dan, Sean and Matt)
>>
>> We have a 2 changes in stable/liberty
>>
>> https://review.openstack.org/#/c/248505 Add -constraints sections for
>> CI jobs ; and
>> https://review.openstack.org/#/c/248877 Remove the TestRemoteObject class
>>
>> If you grab 248505 and look at the git DAG you get:
>> $ git log --oneline --decorate  -3
>> 83ca84a (HEAD -> review/sachi_king/bp/Requirements-Management) Add
>> -constraints sections for CI jobs
>> 6e2da82 Remove the TestRemoteObject class
>> 94d6b69 (origin/stable/liberty, gerrit/stable/liberty) Omnibus stable
>> fix for upstream requirements breaks
>>
>> so 248877 is based on stable/liberty and 248505 is based on 248877
>>
>> The problem is that 248877 Depends-On 248505[1]
>>
>> I think the correct solution is to remove the Depends-On directive
>> from 248877.
>>
>> I didn't do that thing as:
>> 1. I don't understand whay that there and could be missing something
>> 2. Doing so would loose th 2+2's and +W anyway so wont really help.
>>
>>
>> Yours Tony.
>> [1] Well it Depends on Icbbb78cfcd074b0050e60c54557637af723f9b92 which
>> maps to
>>      the same change in master and stable/liberty ....
>>
>>
>>
>> __________________________________________________________________________
>>
>> 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
>>
>
> https://review.openstack.org/#/c/248505 depends on
> https://review.openstack.org/#/c/248877 because without it the unit
> tests fail in the constraints job, which we wanted to see passing before
> landing 248505, so they are stacked such that we could see the unit
> tests passing in the constraints job in 248877 when it's on top of the
> stack. I'm not sure why the depends-on was added to 248877, Sean added
> that I think.
>
> We could remove the depends-on and I could fast approve, that's not a
> problem, but now the tempest job is failing on both so I'll have to
> check that out first.
>

We should be good now. I removed the bad depends-on and re-approved. 
Dan's change is merged and the -constraints change is rechecked so that 
should go through now.

Thanks for bringing this up Tony.

-- 

Thanks,

Matt Riedemann




More information about the OpenStack-dev mailing list