On Thu, Oct 19, 2017 at 10:53 AM, Eyal Edri <eedri@redhat.com> wrote:


On Thu, Oct 19, 2017 at 11:51 AM, Barak Korren <bkorren@redhat.com> wrote:


On 19 October 2017 at 11:43, Eyal Edri <eedri@redhat.com> wrote:


On Thu, Oct 19, 2017 at 10:50 AM, Allon Mureinik <amureini@redhat.com> wrote:
The missing deps issue happened again this morning [1]:

Traceback (most recent call last):
  File "/tmp/ovirt-q04eYYi5Ym/pythonlib/otopi/context.py", line 133, in _executeMethod
    method['method']()
  File "/tmp/ovirt-q04eYYi5Ym/otopi-plugins/otopi/packagers/yumpackager.py", line 256, in _packages
    if self._miniyum.buildTransaction():
  File "/tmp/ovirt-q04eYYi5Ym/pythonlib/otopi/miniyum.py", line 920, in buildTransaction
    raise yum.Errors.YumBaseError(msg)
YumBaseError: [u'vdsm-4.20.3-205.git15d3b78.el7.centos.x86_64 requires libvirt-daemon-kvm >= 3.2.0-14.el7_4.3', u'10:qemu-kvm-ev-2.9.0-16.el7_4.5.1.x86_64 requires ipxe-roms-qemu >= 20170123-1.git4e85b27.el7_4.1']
2017-10-19 01:36:37,275-0400 ERROR otopi.context context._executeMethod:152 Failed to execute stage 'Package installation': [u'vdsm-4.20.3-205.git15d3b78.el7.centos.x86_64 requires libvirt-daemon-kvm >= 3.2.0-14.el7_4.3', u'10:qemu-kvm-ev-2.9.0-16.el7_4.5.1.x86_64 requires ipxe-roms-qemu >= 20170123-1.git4e85b27.el7_4.1']
We need to fix the missing packages (broken repo?) issue ASAP, as it would mast any other real problems we may have there

We're looking into it now, it's strange that official qemu-kvm-ev is requiring a version of ipxe-roms-qemu with git sha 20170123-1.git4e85b27.el7_4.1.
It looks like the same pkg is coming from centos base, updateds and kvm-commons, and some repos include older version without the '4.1' suffix.

However, its strange that some jobs does pass, e.g - last finished run from 1.5 hours ago:

 

There is nothing strange about this.
The failure Alon linked to is in OST check-patch for Denial's patch [1] that closes the external repos, so it is expected to fail on issues like this. This is not merged so doesn't cause any issues for "normal" (E.g. change-queues) OST runs.

I can't see that missing pkg error anymore actually, so whatever it was it might be fixed ( it failed also on another patch, not just Daniel's ).
There might be a second issue here with the error Daniel sent on glusterfs-server and firewalld, I think we should focus on investigating that. 

​This is not an issue, which should the reason for jobs to fail (we are ignoring this error during firewalld setup), because AFAIK we are not doing any gluster related tests in basic OST.​


Anyway the discussion about missing gluster firewalld service continues ...
 

--
Barak Korren
RHV DevOps team , RHCE, RHCi
Red Hat EMEA
redhat.com | TRIED. TESTED. TRUSTED. | redhat.com/trusted



--

Eyal edri


MANAGER

RHV DevOps

EMEA VIRTUALIZATION R&D


Red Hat EMEA

TRIED. TESTED. TRUSTED.
phone: +972-9-7692018
irc: eedri (on #tlv #rhev-dev #rhev-integ)

_______________________________________________
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel