On Nov 23, 2017 17:23, "Nir Soffer" <nsoffer@redhat.com> wrote:


On Thu, Nov 23, 2017 at 4:55 PM Dan Kenigsberg <danken@redhat.com> wrote:
On Thu, Nov 23, 2017 at 4:03 PM, Nir Soffer <nsoffer@redhat.com> wrote:
> On Thu, Nov 23, 2017 at 3:59 PM Dan Kenigsberg <danken@redhat.com> wrote:
>>
>> On Thu, Nov 23, 2017 at 1:56 PM, Nir Soffer <nsoffer@redhat.com> wrote:
>> > On Thu, Nov 23, 2017 at 1:51 PM Edward Haas <ehaas@redhat.com> wrote:
>> >>
>> >> Per what I see, all CI jobs on vdsm/fc27 fail.
>> >> This is the second time this week, please consider reverting.
>> >>
>> >> We should try to avoid such changed before the weekend
>> >
>> >
>> > Some of the failures like
>> >
>> > http://jenkins.ovirt.org/job/vdsm_master_check-patch-fc27-x86_64/100/console
>> >
>> > Should be fixed by:
>> > https://gerrit.ovirt.org/84569/
>> >
>> > I have see some network test failing when running mock_runner locally,
>> > you may need to mark them as broken on fc27.
>>
>> We have a busy couple of weeks until release of ovirt-4.2.0. As much
>> as I like consuming Fedora early, I'm not sure that enabling so close
>> to the release was a good idea. Nothing forces us to do it now (and
>> there are a lot of reasons to do it later). Let's give it another go,
>> but let us not keep it on the red for the weekend.
>
>
> I'm happy with the storage code being tested on current fedora.

I'm more than happy. I'm thrilled to for it to be tested and run.
I am not happy to give a vote to a job that was never ever successful.

The job is successful,

That's an oversimplification. http://jenkins.ovirt.org/job/vdsm_master_check-patch-fc27-x86_64/174/consoleText fails with another dependency

  - nothing provides mom >= 0.5.8 needed by vdsm-4.20.8-1.git383bc1031.fc27.x86_64
  - nothing provides mom >= 0.5.8 needed by vdsm-4.20.8-3.gita9ee9c65f.fc27.x86_64
  - nothing provides python-argparse needed by vdsm-4.18.999-444.git0bb7717.fc27.x86_64

some tests or maybe the code they test need work.
This is why we have skipif/xfail and broken_on_ci.
 

>
> Please make sure the few failing network tests are not breaking the build.

It's probably time to mark the tests that use loseup as broken-on-jenkins

loop devices are usually ok on jenkins. We have several tests in storage using
them and I don't know about any failures. For example storage/blockdev_test.py.

Francesco, do you want to mark them as broken for now?
 

ERROR: Tests mkimage.injectFilesToFs creating an image and checking its
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/tests/testlib.py",
line 143, in wrapper
    return f(self, *args)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/tests/testValidation.py",
line 191, in wrapper
    return f(*args, **kwargs)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/tests/mkimage_test.py",
line 176, in test_injectFilesToFs
    mkimage.injectFilesToFs(floppy, self.files, fstype)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/lib/vdsm/mkimage.py",
line 112, in injectFilesToFs
    m.mount(mntOpts='loop', vfstype=fstype)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/lib/vdsm/storage/mount.py",
line 208, in mount
    cgroup=cgroup)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/lib/vdsm/storage/mount.py",
line 278, in _mount
    _runcmd(cmd)
  File "/home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64/vdsm/lib/vdsm/storage/mount.py",
line 306, in _runcmd
    raise MountError(rc, ";".join((out, err)))
MountError: (32, ';mount:
/tmp/vdsm-mkimage-testspahtvm/images/vmId_inject.7301b08f8c0dc496bedd25577dbddbd8.img:
failed to setup loop device: No such file or directory\n')



>
> Nir
>
>