On Wed, Jul 22, 2015 at 09:23:46AM +0300, Yedidyah Bar David wrote:
On Wed, Jul 22, 2015 at 2:34 AM, Alan Murrell
<lists(a)murrell.ca> wrote:
> Hi Dan,
>
> Quoting "Dan Kenigsberg" <danken(a)redhat.com>:
>>
>> virsh -r dumpxml vdsm-ovirtmgmt
>
>
> I will see if I can get you that output a little later, along with the
> vdsClient output suggested by Yedidyah.
>
>> your own, you should also define that network. Skipping network
>> definition might have worked in the past, but it would certainly not
>
>
> When you refer to defining the "network definition", are you referring to
a
> "NETWORK=" line in my ifcfg-ovirtmgmt file (which I think was referred to
by
> Yedidyah)?
Dan refers to the below:
Actually, I was referring to placing a file (say /tmp/1.xml) with
<network>
<name>vdsm-ovirtmgmt</name>
<uuid>f661da6b-6dea-8c4d-af7f-c0c9111ea10e</uuid>
<forward mode='bridge'/>
<bridge name='ovirtmgmt'/>
</network>
and defining it in libvirt with
virsh net-define /tmp/1.xml
virsh net-start vdsm-ovirtmgmt
but not that I do not endorse it; the following is what should be followed:
>
> If necessary, I can also blow away my current install and try the deployment
> again without defining the bridge first, if for no other reason than if it
> still fails to do so, it is a bug that that you will be aware of.
Yes, please do. This will be mandatory in 3.6.
Thanks, and please report back with any issue (or success :-) ).
Best,
--
Didi