[Gluster-devel] [Gluster-infra] Smoke results voting

Prasanna Kalever pkalever at redhat.com
Tue Apr 5 13:25:01 UTC 2016


On Tue, Apr 5, 2016 at 4:24 PM, Kaushal M <kshlmster at gmail.com> wrote:
>
> On 5 Apr 2016 3:43 p.m., "Kaushal M" <kshlmster at gmail.com> wrote:
>>
>> So gerrit voting seems to be working again. It required all the jobs
>> to have the same trigger configuration.
>>
>> Smoke vote was given for [1] a test change that I posted for review.
>> Now just need to check if it works for changes which already have
>> regressions run on them.
>> I'll be following [2] to see it works.
>
> Working now. If anyone notices anything amiss, please let gluster-infra know
> of it.

Kaushal,

Since now we have collated smoke report-back, it would be really
helpful If we can make "event based on regexp match with comment"
(<commentAddedCommentContains/>)available for other jobs such as :

compare-bug-version-and-git-branch
netbsd6-smoke
freebsd-smoke


Once a user fix the cause for failure then he can re-trigger via 'recheck *'

Thanks,
--
Prasanna




>
>>
>> [1] https://review.gluster.org/13898
>> [2] https://review.gluster.org/13869
>>
>> On Tue, Apr 5, 2016 at 12:46 PM, Prasanna Kalever <pkalever at redhat.com>
>> wrote:
>> > On Tue, Apr 5, 2016 at 12:39 PM, Kaushal M <kshlmster at gmail.com> wrote:
>> >> On Tue, Apr 5, 2016 at 11:26 AM, Kaushal M <kshlmster at gmail.com> wrote:
>> >>> On Tue, Apr 5, 2016 at 11:10 AM, Atin Mukherjee <amukherj at redhat.com>
>> >>> wrote:
>> >>>>
>> >>>>
>> >>>> On 04/05/2016 11:06 AM, Kaushal M wrote:
>> >>>>> I did some changes so that all smoke jobs (linux, *bsd smoke jobs,
>> >>>>> devrpm jobs etc) are triggered for `recheck smoke`.
>> >>>>>
>> >>>>> The collated results are being reported back and the 'Smoke' flag is
>> >>>>> being set. But sometimes, if regression jobs have been already run
>> >>>>> on
>> >>>>> the patchset, jenkins is collating those results as well.
>> >>>>> When this happens, a '-Verified' flag is being set.
>> >>>>>
>> >>>>> Jenkins and its gerrit plugin collate results for jobs launched by
>> >>>>> the
>> >>>>> same event. The regression and smoke jobs should be triggered for
>> >>>>> different events,
>> >>>>> but for some reason jenkins is assuming that they're being triggered
>> >>>>> by the same event and collating all of them together.
>> >>>>>
>> >>>>> I need some time to figure out why this is happening, and fix it.
>> >>>> Is there a way that this doesn't impact the merging as until we get
>> >>>> all
>> >>>> the positive votes, web interface doesn't provide a submit button.
>> >>>
>> >>> This would require manually running the gerrit ssh command as the
>> >>> build user to set the flag, which requires sudo access on
>> >>> build.gluster.org.
>> >>
>> >> Alternatively, administrators can spoof other users. Administrators can
>> >> do
>> >> `ssh <username>@review.gluster.org  suexec --as
>> >> jenkins at build.gluster.org -- gerrit review --label Smoke=+1
>> >> <patch>,<patchset>`
>> >
>> > As Kaushal mentioned above
>> >
>> > 1. users are free to comment "recheck smoke" (which will  trigger smoke)
>> > 2. only after success on step 1, administrators will get +1 done with
>> > 'ssh ...'
>> >
>> >
>> > Thanks,
>> > --
>> > Prasanna
>> >
>> >
>> >
>> >>
>> >> I'm still figuring out how to solve it properly though.
>> >>
>> >>>
>> >>>
>> >>>>>
>> >>>>> ~kaushal
>> >>>>>
>> >>>>>
>> >>>>> On Mon, Apr 4, 2016 at 10:39 PM, Prasanna Kalever
>> >>>>> <pkalever at redhat.com> wrote:
>> >>>>>> On Mon, Apr 4, 2016 at 9:58 PM, Atin Mukherjee
>> >>>>>> <atin.mukherjee83 at gmail.com> wrote:
>> >>>>>>> Did anyone notice that for few of the patches smoke results are
>> >>>>>>> not voted
>> >>>>>>> back?
>> >>>>>>>
>> >>>>>>> http://review.gluster.org/#/c/13869 is one of them.
>> >>>>>>
>> >>>>>> +1
>> >>>>>>
>> >>>>>> Here is another one http://review.gluster.org/#/c/11083/
>> >>>>>> I have also re-triggered it with "recheck smoke", after which
>> >>>>>> it return success "http://build.gluster.org/job/smoke/26373/ :
>> >>>>>> SUCCESS"
>> >>>>>>
>> >>>>>> but again failed to report back ...
>> >>>>>>
>> >>>>>> --
>> >>>>>> Prasanna
>> >>>>>>
>> >>>>>>
>> >>>>>>>
>> >>>>>>> -Atin
>> >>>>>>> Sent from one plus one
>> >>>>>>>
>> >>>>>>>
>> >>>>>>> _______________________________________________
>> >>>>>>> Gluster-infra mailing list
>> >>>>>>> Gluster-infra at gluster.org
>> >>>>>>> http://www.gluster.org/mailman/listinfo/gluster-infra
>> >>>>>> _______________________________________________
>> >>>>>> Gluster-devel mailing list
>> >>>>>> Gluster-devel at gluster.org
>> >>>>>> http://www.gluster.org/mailman/listinfo/gluster-devel
>> >>>>> _______________________________________________
>> >>>>> Gluster-devel mailing list
>> >>>>> Gluster-devel at gluster.org
>> >>>>> http://www.gluster.org/mailman/listinfo/gluster-devel
>> >>>>>


More information about the Gluster-devel mailing list