ovirt 3.3 nightly on centos

[Ohad Basan]

Alon suggested to use the nightly repository and not the beta one as currently there is a bug in the beta rpm that crashes the all in one plugin
I still missing the required libvirt:
vdsm-4.12.0-103.gitce1e10a.el6.x86_64 requires libvirt >= 0.10.2-18.el6_4.10



----- Original Message -----
> From: "Dan Kenigsberg" <danken at redhat.com>
> To: "Itamar Heim" <iheim at redhat.com>
> Cc: "Ohad Basan" <obasan at redhat.com>, dallan at redhat.com, "infra" <infra at ovirt.org>, "Dave Neary" <dneary at redhat.com>
> Sent: Friday, September 6, 2013 10:31:07 PM
> Subject: Re: ovirt 3.3 nightly on centos
> 
> On Wed, Sep 04, 2013 at 03:41:31PM +0300, Itamar Heim wrote:
> > On 09/03/2013 12:34 AM, Dan Kenigsberg wrote:
> > >On Mon, Sep 02, 2013 at 12:14:44PM -0400, Ohad Basan wrote:
> > >>Hey
> > >>
> > >>during prepping ovirt live
> > >>I am installing ovirt 3.3 nightly on a centos 6.4 machines
> > >>I have discovered that I am missing tons of dependencies not found in the
> > >>standard repo
> > >>1. gluster >=3.4
> > >>2. vdsm requires libvirt >= 0.10.2-18.el6_4.4
> > >
> > >Which reminds me of a problem with libvirt -18:
> > >Bug 895294 - vmUpdateDevice fails when nic has boot order set is to be
> > >fixed only for el6.5. Until then we are stuck with bugs such as
> > >https://bugzilla.redhat.com/show_bug.cgi?id=1001001
> > >
> > >Unless we are willing to hack around the issue with something like the
> > >untested http://gerrit.ovirt.org/#/c/18796/
> > 
> > i don't see a simple solution to this one:
> > if we pick latest stable libvirt/qemu upstream for a
> > "(o)virt-preview" repo for .el6, then we won't get the stable 6.5
> > when its released as it will have a lower version since rhel doesn't
> > tend to rebase.
> 
> The libvirt team has agreed to quickly backport the fix to el6.4.z.
> 
>     Bug 1003934 - vmUpdateDevice fails when nic has boot order set
> 
> is now solved by libvirt-0.10.2-18.el6_4.10 which should be required by
> vdsm. See http://gerrit.ovirt.org/18937
> 
> Thanks!
>