[oVirt Jenkins] vdsm_master_unit-tests_merged - Build # 4349 - Still Failing!

David Caro dcaro at redhat.com
Thu Dec 11 19:10:14 UTC 2014


On 12/11, Petr Horacek wrote:
> 
> 
> ----- Original Message -----
> > From: "David Caro" <dcaro at redhat.com>
> > To: "Dan Kenigsberg" <danken at redhat.com>
> > Cc: phoracek at redhat.com, "Jenkins ci oVirt Server" <jenkins at ovirt.org>, eedri at redhat.com,
> > vdsm-patches at lists.fedorahosted.org, nsoffer at redhat.com
> > Sent: Thursday, December 11, 2014 11:42:39 AM
> > Subject: Re: [oVirt Jenkins] vdsm_master_unit-tests_merged - Build # 4349 - Still Failing!
> > 
> > On 12/11, Dan Kenigsberg wrote:
> > > On Thu, Dec 11, 2014 at 10:29:52AM +0100, David Caro wrote:
> > > > On 12/11, Dan Kenigsberg wrote:
> > > > > On Wed, Dec 10, 2014 at 11:42:07PM +0000, Jenkins ci oVirt Server
> > > > > wrote:
> > > > > > Project: http://jenkins.ovirt.org/job/vdsm_master_unit-tests_merged/
> > > > > > Build:
> > > > > > http://jenkins.ovirt.org/job/vdsm_master_unit-tests_merged/4349/
> > > > > > Build Number: 4349
> > > > > > Build Status:  Still Failing
> > > > > > Triggered By: Triggered by Gerrit: http://gerrit.ovirt.org/36075
> > > > > 
> > > > > 
> > > > > David,
> > > > > 
> > > > > This junk mail is tiring.
> > > > > 
> > > > > Could you get this job to run on a host with ioprocess>=0.14 ?
> > > > 
> > > > 
> > > > I prefer getting this fixed:
> > > > 
> > > > http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/
> > > > 
> > > > It's just missing a test that hangs (testMonitorIteration) waiting for
> > > > a interface down event. They passed because I manually executed:
> > > > 
> > > > > ip link set $interface_name down
> > > > 
> > > > When the test was stuck. It passed once though without that.
> > > 
> > > Can you pass NOSE_EXCLUDE=testMonitorIteration so that the test is not
> > > attempted?

I've done that, but now it seems to fail on another test (might be
that something got merged in the meantime that breaks it):

00:06:54.192 ERROR: testAddNetworkValidation (configNetworkTests.TestConfigNetwork)
...
00:06:54.192   File "/tmp/run/vdsm/tests/monkeypatch.py", line 133, in wrapper
00:06:54.192     return f(*args, **kw)
00:06:54.192   File "/tmp/run/vdsm/tests/configNetworkTests.py", line 101, in testAddNetworkValidation
00:06:54.192     _netinfo=fakeInfo), errors.ERR_USED_BRIDGE)
00:06:54.192   File "/tmp/run/vdsm/tests/configNetworkTests.py", line 47, in _addNetworkWithExc
00:06:54.192     api.addNetwork(netName, **opts)
00:06:54.192 AttributeError: 'module' object has no attribute 'addNetwork'

Also there's an ugly tracedump in the middle of the logs:

00:06:49.709     testFailuresOnExposedFuncs                                  can't remove configuration of module a
00:06:49.709 Traceback (most recent call last):
00:06:49.709   File "/tmp/run/vdsm/lib/vdsm/tool/configurator.py", line 166, in remove_config
00:06:49.709     c.removeConf()
00:06:49.709   File "/tmp/run/vdsm/tests/toolTests.py", line 76, in removeConf
00:06:49.710     raise Exception('mock invalid remove conf')
00:06:49.711 Exception: mock invalid remove conf
00:06:49.711 can't remove configuration of module b
00:06:49.711 Traceback (most recent call last):
00:06:49.711   File "/tmp/run/vdsm/lib/vdsm/tool/configurator.py", line 166, in remove_config
00:06:49.711     c.removeConf()
00:06:49.711   File "/tmp/run/vdsm/tests/toolTests.py", line 76, in removeConf
00:06:49.711     raise Exception('mock invalid remove conf')
00:06:49.711 Exception: mock invalid remove conf
00:06:49.711 OK

But that one does not fail the job.

> > > 
> > > Petr, since you figured out why the test is blocking (race with starting
> > > ip monitor on terribly slow jenkins slave), would you post a quick hack
> > > so we can disable the old test?
> Sleep hack in ipwrapperTests or in ipwrapper?
> 
> > 
> > I've been able to reproduce on idle slaves, it's not that the slave is
> > loaded, or anything similar. I've reproduced also on my laptop (8GB
> > RAM, 8cores i7...not doing anything else). So it might be a delay when
> > running the monitoring, but it's not due to a slow slave (not in this
> > case at least).
> I reproduced it on a quite fast machine too (not always, but in many and
> many cases).
> 
> > 
> > > 
> > 
> > --
> > David Caro
> > 
> > Red Hat S.L.
> > Continuous Integration Engineer - EMEA ENG Virtualization R&D
> > 
> > Tel.: +420 532 294 605
> > Email: dcaro at redhat.com
> > Web: www.redhat.com
> > RHT Global #: 82-62605
> > 

-- 
David Caro

Red Hat S.L.
Continuous Integration Engineer - EMEA ENG Virtualization R&D

Tel.: +420 532 294 605
Email: dcaro at redhat.com
Web: www.redhat.com
RHT Global #: 82-62605
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: not available
URL: <https://lists.fedorahosted.org/pipermail/vdsm-patches/attachments/20141211/17e3d459/attachment-0001.sig>


More information about the vdsm-patches mailing list