On 08/04/2014 09:46 AM, LeslieWang wrote:
> Dear all,
> Looking at devtest pages at TripleO wiki 
> http://docs.openstack.org/developer/tripleo-incubator/devtest.html. I thought 
> all variables and configurations of devtest are for true bare-metal because I 
> see diskimage-builder options of both overcloud and undercloud doesn't 
> include "vm" option. But I see this configuration in 
> tripleo-incubator/scripts/devtest_testenv.sh, 
> ## #. Set the default bare metal power manager. By default devtest uses
> ##    nova.virt.baremetal.virtual_power_driver.VirtualPowerManager to
> ##    support a fully virtualized TripleO test environment. You may
> ##    optionally customize this setting if you are using real baremetal
> ##    hardware with the devtest scripts. This setting controls the
> ##    power manager used in both the seed VM and undercloud for Nova 
> Baremetal.
> ##    ::
> POWER_MANAGER=${POWER_MANAGER:-'nova.virt.baremetal.virtual_power_driver.VirtualPowerManager'}
> Thus, seems like all setting are for virtual environment, not for true bare 
> metal. So I'm a little confused. Can anyone help clarify it? And what is the 
> right configure of POWER_MANAGER if using real bare metal hardware?
> Best RegardsLeslie                       

Devtest defaults to a virtual environment because that's what most
people working with it have available.  We try to stay as close to the
baremetal use case as possible though, which is why the vm element isn't
included in image builds.  The images are built to be deployed to
baremetal, and the virtual deployment just uses the VirtualPowerManager
to do that.

The correct power manager to use for real baremetal is going to depend
on the hardware you have available.  You'll have to look through the
available power managers and decide which is appropriate for you.


> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

OpenStack-dev mailing list

Reply via email to