constantly failing http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el
The job on $subject constantly reports failure, regardless of successful output. http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14286/console Ran 801 tests in 94.766s OK (SKIP=57) ... Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to FAILURE Finished: FAILURE Can you look into it?
--WIIRZ1HQ6FgrlPgb Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable I vaguely recall that we already talked about this, and someone started working on it. The issue is that it's trying to use the tests output, but the run does not generate any (that was a feature change in the python test module that it uses). """ Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to FAILU= RE """ On 01/19, Dan Kenigsberg wrote:
The job on $subject constantly reports failure, regardless of successful output. =20 http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14286/conso= le =20 Ran 801 tests in 94.766s =20 OK (SKIP=3D57) ... Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to FAI= LURE Finished: FAILURE =20 =20 Can you look into it? _______________________________________________ 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 --WIIRZ1HQ6FgrlPgb Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJUvhs4AAoJEEBxx+HSYmnDp0YH/RqXsMxH2HUgT7PNebvY60bb sGUJfV7AoHc4PNW93qs0zbwCLRLx/nXCqL8uf8DYU8/DD0RO1MJpNfoYMpeNKC6g Q6nqQnZRZV+WpZbPyLWtih+I55Q43pJsX4Tjijjz0+4vCzt0gzQ18WPFSrLz3mH4 /dOOQ6ukgevLJCCfO76uTTIaMsEANy6lTkkpx0VA1Oxv4EUZ/8/6l5tMYR8bBzft 1S4jA3RM0wKXU9ti5UE9SXndGWj721d4nqmmx3Qcm0c3EdPeRw8dQMI2scziOth3 iGQUULeevdHQ9jCRnyUNelhI53wEB3vmkd34JoqwfcQX1Z6vR5qI8BUsbNJDAu8= =cLgf -----END PGP SIGNATURE----- --WIIRZ1HQ6FgrlPgb--
On Tue, Jan 20, 2015 at 10:09:12AM +0100, David Caro wrote:
I vaguely recall that we already talked about this, and someone started working on it.
The issue is that it's trying to use the tests output, but the run does not generate any (that was a feature change in the python test module that it uses).
""" Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to FAILURE """
I suspect that this is because the installed python-nose does not have xunit. In the old days, we had http://dl.fedoraproject.org/pub/epel/6/i386/python-nose1.1-1.1.2-4.el6.noarc... on el6 hosts to overcome this. Could you install this nose version on the job's el6 slaves? Dan.
On Tue, Jan 20, 2015 at 01:14:03PM +0000, Dan Kenigsberg wrote:
On Tue, Jan 20, 2015 at 10:09:12AM +0100, David Caro wrote:
I vaguely recall that we already talked about this, and someone started working on it.
The issue is that it's trying to use the tests output, but the run does not generate any (that was a feature change in the python test module that it uses).
""" Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to FAILURE """
I suspect that this is because the installed python-nose does not have xunit.
In the old days, we had http://dl.fedoraproject.org/pub/epel/6/i386/python-nose1.1-1.1.2-4.el6.noarc... on el6 hosts to overcome this.
Could you install this nose version on the job's el6 slaves?
https://fedorahosted.org/ovirt/ticket/289 opened to track this request.
--/WwmFnJnmDyWGHa4 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable I've retaken the work on that job, I've got it mostly working but there are= a few tests failing: http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/435/con= sole Can someone from your team take a look? I'll update the tracker also (I thi= nk yeela is cc'd there). On 01/28, Dan Kenigsberg wrote:
On Tue, Jan 20, 2015 at 01:14:03PM +0000, Dan Kenigsberg wrote:
On Tue, Jan 20, 2015 at 10:09:12AM +0100, David Caro wrote:
=20 I vaguely recall that we already talked about this, and someone started working on it. =20 The issue is that it's trying to use the tests output, but the run does not generate any (that was a feature change in the python test module that it uses). =20 """ Recording test results No test report files were found. Configuration error? Build step 'Publish JUnit test result report' changed build result to= FAILURE """ =20 I suspect that this is because the installed python-nose does not have xunit. =20 In the old days, we had http://dl.fedoraproject.org/pub/epel/6/i386/python-nose1.1-1.1.2-4.el6.= noarch.rpm on el6 hosts to overcome this. =20 Could you install this nose version on the job's el6 slaves? =20 https://fedorahosted.org/ovirt/ticket/289 opened to track this request.
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 --/WwmFnJnmDyWGHa4 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEbBAEBAgAGBQJU1PIyAAoJEEBxx+HSYmnDOTkH+MyNtNOj4ZBR5fy8jcZ0820D C/Jld7BOizxG1OQthV3Sp8tbCaKHMzc4Tavj0wnp2GLOz+ZiGrvJ+aJ7X/AC3Tel N/EncZZtssXZ/gdYTTnWaSB/YKU3bCj28kkQcntiQWxjg6r7kJCvEqHA0diRAlWS 2gkfUwhhiTMRVHNhBpDt5VW5ZqAIuJe+C5IhC3OUOInt2OpRTz5gE4u3ra2aVRSR sbYErRPbzG77bvkeBBxrL5VxnGLHGDqg/148Aoo741uokYnOj6xznpFXdkiSZz1f /70qO2HtGvM7zw+BfUhcHbH3jfh0RmprS948neexZj1hokO3OD4C2wKvk6OTLg== =Ckly -----END PGP SIGNATURE----- --/WwmFnJnmDyWGHa4--
On Fri, Feb 06, 2015 at 05:56:18PM +0100, David Caro wrote:
I've retaken the work on that job, I've got it mostly working but there are a few tests failing:
http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/435/cons...
Can someone from your team take a look? I'll update the tracker also (I think yeela is cc'd there).
I think that re-installing python-nose1.1 is a simple stopgap measure for the old job. As for the mock-based errors. They have been gc-ed by now, but I do recall that they had testHostSampleHandlesDisappearingVlanInterfaces testHostSampleReportsNewInterface failing. I hopw Francesco can fix that easily.
----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "David Caro" <dcaro@redhat.com>, fromani@redhat.com Cc: infra@ovirt.org Sent: Monday, February 9, 2015 10:06:14 PM Subject: Re: constantly failing http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el
On Fri, Feb 06, 2015 at 05:56:18PM +0100, David Caro wrote:
I've retaken the work on that job, I've got it mostly working but there are a few tests failing:
http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/435/cons...
Can someone from your team take a look? I'll update the tracker also (I think yeela is cc'd there).
I think that re-installing python-nose1.1 is a simple stopgap measure for the old job.
As for the mock-based errors. They have been gc-ed by now, but I do recall that they had
testHostSampleHandlesDisappearingVlanInterfaces testHostSampleReportsNewInterface
failing. I hopw Francesco can fix that easily.
Not really easily, yet. I see them both fail because they claim a numa-related function seems to be missing, but I can't see anything peculiar about that function (getMemoryStats), like monkeypatching or any hack. The function was added sufficiently long ago, so it should not be related to package versions. Moreover, this doesn't reproduce on my F21 laptop. I'll need to dig a bit more. -- Francesco Romani RedHat Engineering Virtualization R & D Phone: 8261328 IRC: fromani
On Tue, Feb 10, 2015 at 05:55:25AM -0500, Francesco Romani wrote:
----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "David Caro" <dcaro@redhat.com>, fromani@redhat.com Cc: infra@ovirt.org Sent: Monday, February 9, 2015 10:06:14 PM Subject: Re: constantly failing http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el
On Fri, Feb 06, 2015 at 05:56:18PM +0100, David Caro wrote:
I've retaken the work on that job, I've got it mostly working but there are a few tests failing:
http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/435/cons...
Can someone from your team take a look? I'll update the tracker also (I think yeela is cc'd there).
I think that re-installing python-nose1.1 is a simple stopgap measure for the old job.
As for the mock-based errors. They have been gc-ed by now, but I do recall that they had
testHostSampleHandlesDisappearingVlanInterfaces testHostSampleReportsNewInterface
failing. I hopw Francesco can fix that easily.
Not really easily, yet. I see them both fail because they claim a numa-related function seems to be missing, but I can't see anything peculiar about that function (getMemoryStats), like monkeypatching or any hack.
The function was added sufficiently long ago, so it should not be related to package versions.
Moreover, this doesn't reproduce on my F21 laptop.
David, which version of libvirt is installed by this mock job? My libvirt-0.10.2.8-8.el6.centos.alt.x86_64 has getMemoryStats. 11:01:39 ERROR: testHostSampleHandlesDisappearingVlanInterfaces (samplingTests.InterfaceSampleTests) 11:01:39 ---------------------------------------------------------------------- 11:01:39 Traceback (most recent call last): 11:01:39 File "/tmp/run/vdsm/tests/testValidation.py", line 105, in wrapper 11:01:39 return f(*args, **kwargs) 11:01:39 File "/tmp/run/vdsm/tests/samplingTests.py", line 172, in testHostSampleHandlesDisappearingVlanInterfaces 11:01:39 hs = sampling.HostSample(os.getpid()) 11:01:39 File "/tmp/run/vdsm/vdsm/virt/sampling.py", line 281, in __init__ 11:01:39 self.numaNodeMem = NumaNodeMemorySample() 11:01:39 File "/tmp/run/vdsm/vdsm/virt/sampling.py", line 173, in __init__ 11:01:39 memInfo = caps.getMemoryStatsByNumaCell(int(nodeIndex)) 11:01:39 File "/tmp/run/vdsm/vdsm/caps.py", line 356, in getMemoryStatsByNumaCell 11:01:39 cellMemInfo = libvirtconnection.get().getMemoryStats(cell, 0) 11:01:39 AttributeError: 'virConnect' object has no attribute 'getMemoryStats'
On Tue, Feb 10, 2015 at 11:34:52AM +0000, Dan Kenigsberg wrote:
On Tue, Feb 10, 2015 at 05:55:25AM -0500, Francesco Romani wrote:
----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "David Caro" <dcaro@redhat.com>, fromani@redhat.com Cc: infra@ovirt.org Sent: Monday, February 9, 2015 10:06:14 PM Subject: Re: constantly failing http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el
On Fri, Feb 06, 2015 at 05:56:18PM +0100, David Caro wrote:
I've retaken the work on that job, I've got it mostly working but there are a few tests failing:
http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/435/cons...
Can someone from your team take a look? I'll update the tracker also (I think yeela is cc'd there).
I think that re-installing python-nose1.1 is a simple stopgap measure for the old job.
As for the mock-based errors. They have been gc-ed by now, but I do recall that they had
testHostSampleHandlesDisappearingVlanInterfaces testHostSampleReportsNewInterface
failing. I hopw Francesco can fix that easily.
Not really easily, yet. I see them both fail because they claim a numa-related function seems to be missing, but I can't see anything peculiar about that function (getMemoryStats), like monkeypatching or any hack.
The function was added sufficiently long ago, so it should not be related to package versions.
Moreover, this doesn't reproduce on my F21 laptop.
David, which version of libvirt is installed by this mock job? My libvirt-0.10.2.8-8.el6.centos.alt.x86_64 has getMemoryStats.
11:01:39 ERROR: testHostSampleHandlesDisappearingVlanInterfaces (samplingTests.InterfaceSampleTests) 11:01:39 ---------------------------------------------------------------------- 11:01:39 Traceback (most recent call last): 11:01:39 File "/tmp/run/vdsm/tests/testValidation.py", line 105, in wrapper 11:01:39 return f(*args, **kwargs) 11:01:39 File "/tmp/run/vdsm/tests/samplingTests.py", line 172, in testHostSampleHandlesDisappearingVlanInterfaces 11:01:39 hs = sampling.HostSample(os.getpid()) 11:01:39 File "/tmp/run/vdsm/vdsm/virt/sampling.py", line 281, in __init__ 11:01:39 self.numaNodeMem = NumaNodeMemorySample() 11:01:39 File "/tmp/run/vdsm/vdsm/virt/sampling.py", line 173, in __init__ 11:01:39 memInfo = caps.getMemoryStatsByNumaCell(int(nodeIndex)) 11:01:39 File "/tmp/run/vdsm/vdsm/caps.py", line 356, in getMemoryStatsByNumaCell 11:01:39 cellMemInfo = libvirtconnection.get().getMemoryStats(cell, 0) 11:01:39 AttributeError: 'virConnect' object has no attribute 'getMemoryStats'
Oh, I see that it's libvirt-daemon x86_64 1.2.9.1-2.fc21 which certainly has this method. very odd.
participants (3)
-
Dan Kenigsberg -
David Caro -
Francesco Romani