It's done


http://dbaxps.blogspot.com/2016/06/attempt-of-rdo-triple0-quickstart-ha.html


Thank you.

Boris.



From: Andrew Beekhof <abeekhof@redhat.com>
Sent: Tuesday, June 7, 2016 9:35 PM
To: Boris Derzhavets
Cc: John Trowbridge; Lars Kellogg-Stedman; rdo-list
Subject: Re: [rdo-list] Tripleo QuickStart HA deployment attempts constantly crash
 
In case you didn't resolve this in the meantime...

On Mon, May 30, 2016 at 6:56 PM, Boris Derzhavets
<bderzhavets@hotmail.com> wrote:
>
> # Define a single controller node and a single compute node.
> overcloud_nodes:
>   - name: control_0
>     flavor: control
>
>   - name: compute_0
>     flavor: compute
>
>   - name: compute_1
>     flavor: compute
>
> # Tell tripleo how we want things done.
> extra_args: >-
>   --neutron-network-type vxlan
>   --neutron-tunnel-types vxlan
>   --ntp-server pool.ntp.org
>
> network_isolation: true
>
>
> Picks up new memory setting but doesn't create second Compute Node.
>
> Every time just Controller && (1)* Compute.

You need to add --control-scale 2 and --compute-scale 2 to 'extra_args'