
Hi infra, Build failed today because of unrelated test: http://jenkins.ovirt.org/job/vdsm_unit_tests_gerrit/5902/console AlignmentScanTests test_aligned_image ERROR test_bad_path OK test_help_response OK test_nonaligned_image ERROR This is not the first time I see these tests fail randomly. The issue may be either bad code or bad test failing randomly. I suggest to exclude these tests until the issue is resolved. Nir

On Fri, Nov 29, 2013 at 07:37:16AM -0500, Nir Soffer wrote:
Hi infra,
Build failed today because of unrelated test: http://jenkins.ovirt.org/job/vdsm_unit_tests_gerrit/5902/console
AlignmentScanTests test_aligned_image ERROR test_bad_path OK test_help_response OK test_nonaligned_image ERROR
This is not the first time I see these tests fail randomly. The issue may be either bad code or bad test failing randomly.
I suggest to exclude these tests until the issue is resolved.
Do you know if the random failures are limited to Jenkins slaves? If not, and you cannot find the root cause, you can decorate the test with @brokentest("explanation about what's broken here and why it is not fixed right now").

----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "Nir Soffer" <nsoffer@redhat.com> Cc: infra@ovirt.org Sent: Friday, November 29, 2013 3:17:45 PM Subject: Re: Bad test creates unwanted noise?
On Fri, Nov 29, 2013 at 07:37:16AM -0500, Nir Soffer wrote:
Hi infra,
Build failed today because of unrelated test: http://jenkins.ovirt.org/job/vdsm_unit_tests_gerrit/5902/console
AlignmentScanTests test_aligned_image ERROR test_bad_path OK test_help_response OK test_nonaligned_image ERROR
This is not the first time I see these tests fail randomly. The issue may be either bad code or bad test failing randomly.
I suggest to exclude these tests until the issue is resolved.
Do you know if the random failures are limited to Jenkins slaves?
I happens on my hosts from time to time, and now also on jenkins. It would be nice if jenkins would send email to the author of the failing tests. Either the test is bad and you wont to fix it, or someone else broke your test and you want to "fix" someone :-)

sounds like test is not stable, if it fails something even on your pc. i suggest like danken to currently disable it until fixed so it wont cause noise. afaik this test runs per patch so whoever breaks it will get -1 on his patch. Sent from my Android phone using TouchDown (www.nitrodesk.com) -----Original Message----- From: Nir Soffer [nsoffer@redhat.com] Received: Friday, 29 Nov 2013, 15:47 To: Dan Kenigsberg [danken@redhat.com] CC: infra@ovirt.org Subject: Re: Bad test creates unwanted noise?

On Fri, Nov 29, 2013 at 12:34:12PM -0500, Eyal Edri wrote:
sounds like test is not stable, if it fails something even on your pc. i suggest like danken to currently disable it until fixed so it wont cause noise.
afaik this test runs per patch so whoever breaks it will get -1 on his patch.
Yes, but the failure is not consistent. At the moment we do not know if it's a guestfs, libvirt, vdsm, or test framework issue. Since we are too lazy to dig into this right now, I've posted http://gerrit.ovirt.org/#/c/21871/
participants (3)
-
Dan Kenigsberg
-
Eyal Edri
-
Nir Soffer