On 03/25/2015 11:18 AM, Jordan Pittier wrote: > On Wed, Mar 25, 2015 at 2:44 PM, Salvatore Orlando <sorla...@nicira.com> > wrote: > >> >> >> On 25 March 2015 at 14:21, Sean Dague <s...@dague.net> wrote: >> >>> On 03/25/2015 09:03 AM, Gary Kotton wrote: >>>> >>>> From: Jordan Pittier <jordan.pitt...@scality.com >>>> <mailto:jordan.pitt...@scality.com>> >>>> Reply-To: OpenStack List <openstack-dev@lists.openstack.org >>>> <mailto:openstack-dev@lists.openstack.org>> >>>> Date: Wednesday, March 25, 2015 at 1:47 PM >>>> To: OpenStack List <openstack-dev@lists.openstack.org >>>> <mailto:openstack-dev@lists.openstack.org>> >>>> Subject: Re: [openstack-dev] [nova] CI report formatting (citrix / >>>> hyperv / vmware ) >>>> >>>> Hi >>>> On Wed, Mar 25, 2015 at 12:39 PM, Sean Dague <s...@dague.net >>>> <mailto:s...@dague.net>> wrote: >>>> >>>> Currently Citrix, HyperV, and VMWare CI systems reporting on Nova >>>> patches have a different formatting than the standard that Jenkins >>> and >>>> other systems are using: >>>> >>>> * test-name-no-spaces http://link.to/result >>>> < >>> https://urldefense.proofpoint.com/v2/url?u=http-3A__link.to_result&d=AwMFaQ&c=Sqcl0Ez6M0X8aeM67LKIiDJAXVeAw-YihVMNtXt-uEs&r=VlZxHpZBmzzkWT5jqz9JYBk8YTeq9N3-diTlNj4GyNc&m=EtHYlIXfK33bYsGf2k8XbFtgWlkcm_VdZCrFHTLEdiE&s=5SS-txUrD3o8KS3QIaCL3XMBbeCYK5CjmzmuxDda7Oc&e= >>>> >>>> : [SUCCESS|FAILURE] some >>>> comment about the test >>>> >>>> I don't want to talk for Citrix, HyperV or VMWare but the "standard" >>>> only work if you use Zuul in your CI. I am using a setup based on a >>>> Jenkins plugin called gerrit-trigger and there's no way to format the >>>> message the way it's expected... >>> >> >> FWIW I help maintain one the VMware CIs (the one voting on neutron and >> network-related patches for devstack and tempest). >> We use gerrit-trigger too (mostly out of lazyness, no other real reason), >> but we're able to format the message posted back to gerrit. >> For posting back votes we use the "gerrit review" command to post the >> message in the standard format. >> > Ok. I managed to find a way. It's possible. For future reference, on the > job configuration, there's a field called "URL to post". The correct value > is literally "* $JOB_NAME $BUILD_URL". Sorry for the noise guys. I can't > find myself an excuse not to report results in the expected format anymore. Thank you, Jordan. Are you able to update the documentation consumed by third party folk using the Jenkins Gerrit trigger plugin? http://ci.openstack.org/third_party.html#the-jenkins-gerrit-trigger-plugin-way
http://git.openstack.org/cgit/openstack-infra/system-config/tree/doc/source/third_party.rst#n212 Since I seem to be the only one doing anything by way of turning systems off and having them turned back on again, I would really be grateful in future of any conversation that doesn't include adding more responsibilities to my list. I'm understanding of the reasoning that unless threatened with being disabled most systems won't take any conversation about their behaviour seriously but I really am at my limit of what I can accomplish here. My list is very heavy as it is, I'm not in favour of adding more to it. Thanks for taking my perspective into consideration, Anita. > >> I think the same process is adopted also by the CI voting on nova. >> However, the job result string is not being posted. I will double check >> with the respective owners. >> >> Salvatore >> >> >>>> >>>> >>>> This means these systems don't show up in the CI rollup block - >>>> http://dl.dropbox.com/u/6514884/screenshot_158.png >>>> < >>> https://urldefense.proofpoint.com/v2/url?u=http-3A__dl.dropbox.com_u_6514884_screenshot-5F158.png&d=AwMFaQ&c=Sqcl0Ez6M0X8aeM67LKIiDJAXVeAw-YihVMNtXt-uEs&r=VlZxHpZBmzzkWT5jqz9JYBk8YTeq9N3-diTlNj4GyNc&m=EtHYlIXfK33bYsGf2k8XbFtgWlkcm_VdZCrFHTLEdiE&s=nIjYwznh1_8aVBSz-XVEJrpNaMsDfqyekOQ2IhiHTo8&e= >>>> >>>> >>>> >>>> Current the Vmware CI will vote +1 iff the patch has passed on the CI. >>>> We can investigate adding this to the CI rollup block. >>>> >>>> >>>> >>>> I'd really like that to change. The CI rollup block has been >>> extremely >>>> useful in getting the test results of a patch above the fold, and >>> the >>>> ability to dig into them clearly. I feel like if any CI system isn't >>>> reporting in standard format that's parsible by that, we should >>> probably >>>> turn it off. >>>> >>>> >>>> I do not think that we should turn this off. They have value. It would >>>> be nice if things were all of the same format, which I guess that this >>>> is the intension of the mail. Lets all try and make an effort to work >>>> towards this goal. >>> >>> Right, honestly, I don't want these turned off, I want them reporting in >>> a more standard format. But I do think if they don't report in a >>> standard format it will cause problems and add to them being ignored. >>> >>> -Sean >>> >>> -- >>> Sean Dague >>> http://dague.net >>> >>> >>> __________________________________________________________________________ >>> OpenStack Development Mailing List (not for usage questions) >>> Unsubscribe: >>> openstack-dev-requ...@lists.openstack.org?subject:unsubscribe >>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >>> >>> >> >> __________________________________________________________________________ >> OpenStack Development Mailing List (not for usage questions) >> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >> >> > > > > __________________________________________________________________________ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev