baremetal driver needs a state between "building" and "deploying"

The code is now different, but I think your proposal is always good to add.

What do you think about to add this states at the just after the call:
https://github.com/openstack/nova/blob/master/nova/virt/baremetal/driver.py#L250

Fix proposed to branch: master
Review: https://review.openstack.org/50348

I took a look at the Ironic PXE driver's handling of this sort of situation, and while I think it's OK and not affected by the precise circumstances described in this bug, I think there may be some similar difficulty in determining why a deploy failed part-way through.

I've tagged the bug as also-affecting and will look into it.

Reviewed: https://review.openstack.org/50348
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=ce2d580106dc04315e71790c98afee062f87351b
Submitter: Jenkins
Branch: master

commit ce2d580106dc04315e71790c98afee062f87351b
Author: Sahid Orentino Ferdjaoui <email address hidden>
Date: Tue Oct 8 13:25:49 2013 +0000

    Adds a PREPARED state after baremetal node power on.

    During a baremetal deployment there is a middle step which
    is not currently represented. If the baremetal node powers on
    but never connects to the deploy-helper, it is impossible to tell
    from the database whether the deploy environment was not created
    or whether the machine is dead.

    Change-Id: I6be3d45fee28970cbb02945c518be34b2bc74689
    Closes-Bug: #1184470

Reviewed: https://review.openstack.org/63037
Committed: https://git.openstack.org/cgit/openstack/ironic/commit/?id=972855e7314c95a07c8483b33138a7a2de8c371c
Submitter: Jenkins
Branch: master

commit 972855e7314c95a07c8483b33138a7a2de8c371c
Author: Devananda van der Veen <email address hidden>
Date: Wed Dec 18 16:57:58 2013 -0800

    Improve error handling in PXE _continue_deploy

    Related to bug 1184470, there was a concern that the PXE driver
    may not be adequately handling errors and informing users when failures
    occur mid-deploy.

    This patch refactors the _continue_deploy() method to handle both errors
    POSTed from the ramdisk and errors that originate within deploy_utils.

    It also fixes an inconsistency in the final provisioning_state:
    ConductorManager.do_node_deploy() will set provisioning_state = ACTIVE,
    however the PXE driver was leaving nodes with state = DEPLOYDONE.

    Change-Id: I29cbff87cbaf85d95687ae094720f8b99f33b65f
    Related-bug: 1184470

Ironic uses a "wait-callback" state as an optional intermediary state if a deploy driver needs to wait for a callback from the node // deploy agent. Closing this bug for Ironic now.