Resources tracker doesn't free data on instance terminate

[Vish Ishaya]

When an instance is terminated on a compute node, the resource tracker keeps the resources allocated for some time. It should remove the resources as soon as the instance is done being cleaned up.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/47504

[Alan Pevec (apevec)]

Too late for Grizzly, final stable/grizzly 2013.1.5 has been released.

[Joe Gordon (jogo)]

Is this bug still valid? the patch was abandoned.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/289544

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/289544
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=ff8bf96fbb2967ff25e3d42abd40da3e22d00ecf
Submitter: Jenkins
Branch: master

commit ff8bf96fbb2967ff25e3d42abd40da3e22d00ecf
Author: Chris Dent <email address hidden>
Date: Mon Mar 7 19:52:08 2016 +0000

    Insure resource tracker updated for deleted instances

    Instances left in a deleted state at compute manager startup were
    not being explicitly cleaned up in the resoruce tracker. To address
    this _update_resource_tracker is now called from solely the
    _complete_deletion method. That method is called both by the
    _init_instance and _delete_instance code paths, both of which should
    be doing cleanup in the resource tracker when the vm is DELETED.

    Note that this change does mean that if _delete_instance has an
    exception, _update_resource_tracker will not be called. This is
    considered okay because we cannot guarantee at which stage the
    exception happened and the state of the resources involved. For safety
    it is better to err on the side of "in use".

    Change-Id: I586c0305ea0e20d98697bc9173ca0f89dc2b228c
    Related-Bug: #1227925