That sounds exactly what I did, do you mind putting those log files into the BZ saves me from having to find some hardware and replicate it again.

So what i did was I ended up just nuking the whole OS, and getting a clean start. First after doing all your initial prep EXCEPT configuring your 4 NICs run the "hosted-engine --deploy" command twice. Assuming the first run fails always like in my case, else once you get to the "Configure Storage" phase press Ctrl + D to exit.

Now configure your NICs (also configure ovirtmgmt manually as there is another BZ about it not being able to create the bridge) and rerun "hosted-engine --deploy" and you should be back in action. This should get you to a working hosted-engine solution.

P.S. could you add me in the CC when you reply, I would've seen your message sooner.

On Thu, Jan 23, 2014 at 9:55 AM, Brad House <brad@monetra.com> wrote:

On 01/22/2014 05:42 PM, Andrew Lau wrote:

Hi,

This looks like this BZ which I reported https://bugzilla.redhat.com/show_bug.cgi?id=1055153

Did you customize your nics before you tried running hosted-engine --deploy?

Thanks,
Andrew

Yes, I created all the /etc/sysconfig/network-scripts/ifcfg-* for my 4 NICs in the bond,
then also created the ifcfg-bond0, as well as an ifcfg-ovirtmgmt which uses the bond0.
So the ovirt management interface should be fully configured before even running
the hosted-engine --deploy ... when I was testing the all-in-one (non-hosted) I had
to do that, so figured it was a good idea for hosted too.

I'm trying to understand your BZ 1055153, did you actually get it to work? Or did
you get stuck at that point? I have seen the same issue as your BZ 1055129 as well.

-Brad
_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users