poll_rebooting_instances hits DB directly in Xen/VMware virt drivers

[Hans Lindgren]

To hard-reboot any stuck-in-reboot instances, both drivers call compute_api.reboot() on the instances which in turn updates the instance state via calls to self.update() -> self._update() -> self.db.instance_update_and_get_original().

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/29453

[Shawn Hartsock (hartsock)]

This looks like it's just waiting on a core reviewer.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/29453
Committed: http://github.com/openstack/nova/commit/1d4daa8f6179945508767019eaa078e22b1636f6
Submitter: Jenkins
Branch: master

commit 1d4daa8f6179945508767019eaa078e22b1636f6
Author: Melanie Witt <email address hidden>
Date: Thu May 16 22:51:48 2013 +0000

    update xen/vmware virt drivers not to hit db directly

    This change replaces calls to compute api with calls to
    conductor api updated with a 'compute_reboot' function.

    Fix for bug 1177993

    Change-Id: I9aefba95cce29c1486fdfb25008f8da9ece1c19b

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/grizzly
Review: https://review.openstack.org/39036

[Gary Kotton (garyk)]

Removed the grizzly backport potential. The concensus was that it was not a critical issue that should be backported