
Hi All, Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective. Was that done on purpose? Can that be restored for master patches? Regards, Yevgeny

On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <yzaspits@redhat.com> wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=find-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

My bad, sorry. The findbugs job runs indeed. However look at that patch https://gerrit.ovirt.org/#/c/44604/ . Patch set 1 got Continuous-Integration+1 from Jenkins after running unit tests and then, much later, Jenkins changed its mind and gave -1 after findbugs job has finished. IMHO that's is a misleading behavior. Regards, Yevgeny On Wed, Aug 12, 2015 at 12:12 PM, Sandro Bonazzola <sbonazzo@redhat.com> wrote:
On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <yzaspits@redhat.com> wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=find-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

--svZFHVx8/dhPCe52 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 08/12, Yevgeny Zaspitsky wrote:
My bad, sorry. The findbugs job runs indeed. =20 However look at that patch https://gerrit.ovirt.org/#/c/44604/ . Patch set 1 got Continuous-Integration+1 from Jenkins after running unit tests and then, much later,
That much later is after merge, the findbugs ran once the patch was merged.= You should understand that each event triggers a tests run, and for each tests = run you get only one review. In this case the tests that check the code when you send the patch passed, but the ones that check the code after it's merged failed.
Jenkins changed its mind and gave -1 after findbugs job has finished. IMHO that's is a misleading behavior. =20 Regards, Yevgeny =20 On Wed, Aug 12, 2015 at 12:12 PM, Sandro Bonazzola <sbonazzo@redhat.com> wrote: =20
On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <yzaspits@redhat.com> wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save s=
ome
Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=3Dfind-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboratio= n. See how it works at redhat.com
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
--=20 David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605 --svZFHVx8/dhPCe52 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJVyz8qAAoJEEBxx+HSYmnD4HEH+gP4llJCBrn49KkKD72mmUpT 4v8BOCfbXjr1l+XKmGUQvqRHkHhDK62MnLbdwwh3ZN3O2Yo+aa/ZBd6ew8znlTfY 5mxPaOxmT6vFX9mnpQJyWuLwzLxPgNaxs4BtKTI5k46sNlmdr/p9wBu38aAejmTq cjzGCz8ubK7SkoaRWCOwRDhI81KbIWYLtfNAYkZ1UfWqHbJGxBVuEOB8AmOr8c9x ELhj+R7ochFi9t2G5aEvZT1D/3TGgcUhFxCA6zuvcDXZnearySrF+TrPMl66mRgP VrpZ3FSPkUA0mkBODqKpcnkzLkWfrgwLxw91pS+Bp/ayMA+BtVNPKgaF17N7Ovs= =ULdu -----END PGP SIGNATURE----- --svZFHVx8/dhPCe52--

On Wed, Aug 12, 2015 at 2:42 PM, David Caro <dcaro@redhat.com> wrote:
On 08/12, Yevgeny Zaspitsky wrote:
My bad, sorry. The findbugs job runs indeed.
However look at that patch https://gerrit.ovirt.org/#/c/44604/ . Patch set 1 got Continuous-Integration+1 from Jenkins after running unit tests and then, much later,
That much later is after merge, the findbugs ran once the patch was merged. You should understand that each event triggers a tests run, and for each tests run you get only one review. In this case the tests that check the code when you send the patch passed, but the ones that check the code after it's merged failed.
In this case, I think we need the job which failed after merge to be run also on patch creation.
Jenkins changed its mind and gave -1 after findbugs job has finished. IMHO that's is a misleading behavior.
Regards, Yevgeny
On Wed, Aug 12, 2015 at 12:12 PM, Sandro Bonazzola <sbonazzo@redhat.com> wrote:
On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <
yzaspits@redhat.com>
wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=find-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community
collaboration. > > See how it works at redhat.com > >
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- David Caro
Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D
Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

The findbugs failure is related to patch set 1, which was previously was marked by Jenkins as +1. If that wasn't +1'ed, we would've wait until Jenkins job is finished, but since it has +1 we were under impression that the patch is OK. Looks like findbugs job was triggered by Verified+1, whereas test were run upon patch set upload. IMHO the split between the jobs makes their behavior misleading. BTW, what was merged is patch set 2, which wasn't checked by findbugs at all. On Wed, Aug 12, 2015 at 3:42 PM, David Caro <dcaro@redhat.com> wrote:
On 08/12, Yevgeny Zaspitsky wrote:
My bad, sorry. The findbugs job runs indeed.
However look at that patch https://gerrit.ovirt.org/#/c/44604/ . Patch set 1 got Continuous-Integration+1 from Jenkins after running unit tests and then, much later,
That much later is after merge, the findbugs ran once the patch was merged. You should understand that each event triggers a tests run, and for each tests run you get only one review. In this case the tests that check the code when you send the patch passed, but the ones that check the code after it's merged failed.
Jenkins changed its mind and gave -1 after findbugs job has finished. IMHO that's is a misleading behavior.
Regards, Yevgeny
On Wed, Aug 12, 2015 at 12:12 PM, Sandro Bonazzola <sbonazzo@redhat.com> wrote:
On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <
yzaspits@redhat.com>
wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on the patches before they get merged into master. Although that could save some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=find-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community
collaboration. > > See how it works at redhat.com > >
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- David Caro
Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D
Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605

--2nTeH+t2PBomgucg Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 08/12, Yevgeny Zaspitsky wrote:
The findbugs failure is related to patch set 1, which was previously was marked by Jenkins as +1. If that wasn't +1'ed, we would've wait until Jenkins job is finished, but since it has +1 we were under impression that the patch is OK.
You are right, the findbugs was triggered by the verified +1 on patch 1.
Looks like findbugs job was triggered by Verified+1, whereas test were run upon patch set upload. IMHO the split between the jobs makes their behavi= or misleading.
BTW, what was merged is patch set 2, which wasn't checked by findbugs at all. =20 On Wed, Aug 12, 2015 at 3:42 PM, David Caro <dcaro@redhat.com> wrote: =20
On 08/12, Yevgeny Zaspitsky wrote:
My bad, sorry. The findbugs job runs indeed.
However look at that patch https://gerrit.ovirt.org/#/c/44604/ . Patch set 1 got Continuous-Integration+1 from Jenkins after running u= nit tests and then, much later,
That much later is after merge, the findbugs ran once the patch was merged. You should understand that each event triggers a tests run, and for each te= sts run you get only one review. In this case the tests that check the code when you send the patch passed, but the ones that check the code after it's merg= ed failed.
Jenkins changed its mind and gave -1 after findbugs job has finished. IMHO that's is a misleading behavior.
Regards, Yevgeny
On Wed, Aug 12, 2015 at 12:12 PM, Sandro Bonazzola <sbonazzo@redhat.c= om> wrote:
On Tue, Aug 11, 2015 at 6:13 PM, Yevgeny Zaspitsky <
yzaspits@redhat.com>
wrote:
Hi All,
Seems like recently Jenkins has stopped running findbugs check on =
Agree, but so far there's no agreement on the flow for it, and there's no effort being done to reach that agreement (it's on hold) so it will not cha= nge any time soon. I think we should drop the triggering by flag, or make it generic but not mix both. That's why I wanted to introduce the workflow fla= g, to trigger only by that flag on demand of the reviewer. Feel free to try and start that discussion again the
patches before they get merged into master. Although that could sa= ve some Jenkins resources and make the its reaction faster, that makes the integration (with master) check less effective.
Was that done on purpose? Can that be restored for master patches?
AFAIK findbugs is still up and running: http://jenkins.ovirt.org/search/?q=3Dfind-bugs Have you any specific patch as example?
Regards, Yevgeny
_______________________________________________ Devel mailing list Devel@ovirt.org http://lists.ovirt.org/mailman/listinfo/devel
-- Sandro Bonazzola Better technology. Faster innovation. Powered by community
collaboration. > > See how it works at redhat.com > >
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- David Caro
Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D
Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605
--=20 David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605 --2nTeH+t2PBomgucg Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJVy2ZXAAoJEEBxx+HSYmnDvZQH/39QZEs6RcZiHC/2ITcHFChm z/RixDW7q8qm32FYaDQAsSToqOo5Vg9aCmQIOlN8/FpAt1eecmIgSFXRQzs5LeIX nT/oI4avrPZ8ENxyae7uWbA3BEqSVlOVmigE4XUG8uy3u6yO++olovaMionthPVB LleNH9/nlNOCxyrv5z5DLJO7p59X3TPB2PV8ebnxsbt34ijwZaZkkJvh7LHkOC5S oFk4bUG3ZOiMqSuTHw3AltZ0JzaA+dxvqhQa6coD8+aOHozELv3FrhVXcAc3Sqmx DaXQ6rngY+ZjC/fjrH2Iud3nnlfyHF/s5uNmuOhJq6UQOKNsMTdm8aWS8INMYE4= =Pt9S -----END PGP SIGNATURE----- --2nTeH+t2PBomgucg--
participants (3)
-
David Caro
-
Sandro Bonazzola
-
Yevgeny Zaspitsky