ovirt-engine rebase does not run master upgrade from master

Hi all, Patchset 28 of [1] failed jenkins on some jobs. Then Roy rebased it on master to patchset 29, and jenkins passed, after running only some of the tests. Roy then commented 'ci please build', which again passed. In practice it should have failed, and failed for me when running manually - thus I pushed patchset 30 which fixed the issue. Why doesn't a rebase causes jenkins to run all tests, but only some of them? [1] https://gerrit.ovirt.org/70422 Best, -- Didi

Hey, The two jobs that didn't run after you've re based you patch are: - [1] upgrade from 3.6 to master - [2] upgrade from master to master We've disabled those jobs because [1] is not supported, I've understood that in order to upgrade from 3.6 to master we actually have to upgrade first to 4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything. Sincerely, On Wed, Feb 8, 2017 at 6:02 PM, Yedidyah Bar David <didi@redhat.com> wrote:
Hi all,
Patchset 28 of [1] failed jenkins on some jobs.
Then Roy rebased it on master to patchset 29, and jenkins passed, after running only some of the tests. Roy then commented 'ci please build', which again passed. In practice it should have failed, and failed for me when running manually - thus I pushed patchset 30 which fixed the issue.
Why doesn't a rebase causes jenkins to run all tests, but only some of them?
[1] https://gerrit.ovirt.org/70422
Best, -- Didi _______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- *Daniel Belenky* *RHV DevOps* *Red Hat Israel*

On Thu, Feb 9, 2017 at 9:40 AM, Daniel Belenky <dbelenky@redhat.com> wrote:
Hey,
The two jobs that didn't run after you've re based you patch are:
[1] upgrade from 3.6 to master [2] upgrade from master to master
We've disabled those jobs because [1] is not supported, I've understood that in order to upgrade from 3.6 to master we actually have to upgrade first to
OK.
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does. If it's not terribly wasteful, please restore this. If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks. Best,
Sincerely,
On Wed, Feb 8, 2017 at 6:02 PM, Yedidyah Bar David <didi@redhat.com> wrote:
Hi all,
Patchset 28 of [1] failed jenkins on some jobs.
Then Roy rebased it on master to patchset 29, and jenkins passed, after running only some of the tests. Roy then commented 'ci please build', which again passed. In practice it should have failed, and failed for me when running manually - thus I pushed patchset 30 which fixed the issue.
Why doesn't a rebase causes jenkins to run all tests, but only some of them?
[1] https://gerrit.ovirt.org/70422
Best, -- Didi _______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- Daniel Belenky RHV DevOps Red Hat Israel
-- Didi

Hey,
The two jobs that didn't run after you've re based you patch are:
[1] upgrade from 3.6 to master [2] upgrade from master to master
We've disabled those jobs because [1] is not supported, I've understood
On Thu, Feb 9, 2017 at 9:40 AM, Daniel Belenky <dbelenky@redhat.com> wrote: that
in order to upgrade from 3.6 to master we actually have to upgrade first to
OK.
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does. If it's not terribly wasteful, please restore this. If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks. Best, On Thu, Feb 9, 2017 at 10:40 AM, Yedidyah Bar David <didi@redhat.com> wrote:
Hey,
The two jobs that didn't run after you've re based you patch are:
[1] upgrade from 3.6 to master [2] upgrade from master to master
We've disabled those jobs because [1] is not supported, I've understood
On Thu, Feb 9, 2017 at 9:40 AM, Daniel Belenky <dbelenky@redhat.com> wrote: that
in order to upgrade from 3.6 to master we actually have to upgrade first to
OK.
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does.
If it's not terribly wasteful, please restore this. If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks.
Best,
Sincerely,
On Wed, Feb 8, 2017 at 6:02 PM, Yedidyah Bar David <didi@redhat.com>
wrote:
Hi all,
Patchset 28 of [1] failed jenkins on some jobs.
Then Roy rebased it on master to patchset 29, and jenkins passed, after running only some of the tests. Roy then commented 'ci please build', which again passed. In practice it should have failed, and failed for me when running manually - thus I pushed patchset 30 which fixed the issue.
Why doesn't a rebase causes jenkins to run all tests, but only some of them?
[1] https://gerrit.ovirt.org/70422
Best, -- Didi _______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- Daniel Belenky RHV DevOps Red Hat Israel
-- Didi
-- *Daniel Belenky* *RHV DevOps* *Red Hat Israel*

On Thu, Feb 9, 2017 at 10:40 AM, Yedidyah Bar David <didi@redhat.com>wrote:
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does.
If it's not terribly wasteful, please restore this.
I've restored this job for now
If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks.
Your'e right, we're working on migrating the upgrade jobs to ost. This project is currently on it's final stretch, and as soon as it'll be fully working - we'll replace the current upgrade job with the new ost. Sincerely, On Thu, Feb 9, 2017 at 11:17 AM, Daniel Belenky <dbelenky@redhat.com> wrote:
Hey,
The two jobs that didn't run after you've re based you patch are:
[1] upgrade from 3.6 to master [2] upgrade from master to master
We've disabled those jobs because [1] is not supported, I've understood
On Thu, Feb 9, 2017 at 9:40 AM, Daniel Belenky <dbelenky@redhat.com> wrote: that
in order to upgrade from 3.6 to master we actually have to upgrade first to
OK.
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does.
If it's not terribly wasteful, please restore this. If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks.
Best,
On Thu, Feb 9, 2017 at 10:40 AM, Yedidyah Bar David <didi@redhat.com> wrote:
Hey,
The two jobs that didn't run after you've re based you patch are:
[1] upgrade from 3.6 to master [2] upgrade from master to master
We've disabled those jobs because [1] is not supported, I've understood
On Thu, Feb 9, 2017 at 9:40 AM, Daniel Belenky <dbelenky@redhat.com> wrote: that
in order to upgrade from 3.6 to master we actually have to upgrade first to
OK.
4.0. And [2] was disabled because upgrading from snapshot to latest doesn't really check anything.
Well, it does. Even upgrade from the patched build to itself does.
If it's not terribly wasteful, please restore this. If it is, perhaps add something similar to OST - I understand migrating the upgrade jobs to OST is currently being worked on, please include there this flow. Thanks.
Best,
Sincerely,
On Wed, Feb 8, 2017 at 6:02 PM, Yedidyah Bar David <didi@redhat.com>
wrote:
Hi all,
Patchset 28 of [1] failed jenkins on some jobs.
Then Roy rebased it on master to patchset 29, and jenkins passed, after running only some of the tests. Roy then commented 'ci please build', which again passed. In practice it should have failed, and failed for me when running manually - thus I pushed patchset 30 which fixed the issue.
Why doesn't a rebase causes jenkins to run all tests, but only some of them?
[1] https://gerrit.ovirt.org/70422
Best, -- Didi _______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
-- Daniel Belenky RHV DevOps Red Hat Israel
-- Didi
--
*Daniel Belenky*
*RHV DevOps*
*Red Hat Israel*
-- *Daniel Belenky* *RHV DevOps* *Red Hat Israel*
participants (2)
-
Daniel Belenky
-
Yedidyah Bar David