allocations not immediately removed when instance is deleted

[Chris Friesen]

Based on code inspection and a discussion with mriedem on IRC, it appears that when deleting an instance in a pure-Pike cloud the allocations are not removed until the update_available_resource() periodic task calls ResourceTracker._update_usage_from_instances(), which calls _remove_deleted_instances_allocations().

In a mixed Ocata/Pike cloud the allocation will be freed up immediately when _update_usage_from_instance() calls self.reportclient.update_instance_allocation().

In the ServerMovingTests functional test we bypass this by forcing the periodic task to run before checking that the allocations have been removed.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/496942

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/pike
Review: https://review.openstack.org/496997

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/496942
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=d6de0e57c21472b99cbe71ec62b3967ef7282dec
Submitter: Jenkins
Branch: master

commit d6de0e57c21472b99cbe71ec62b3967ef7282dec
Author: Matt Riedemann <email address hidden>
Date: Wed Aug 23 17:39:51 2017 -0400

    Delete instance allocations when the instance is deleted

    In a deployment that still has Ocata computes, the resource
    tracker will delete the allocations for an instance when it
    is deleted, but in an all-Pike+ computes deployment, the
    allocations will only be removed for a deleted instance
    during the update_available_resource periodic task.

    The capacity on the node should be freed as soon as the
    instance is deleted, like when we shelve offload an
    instance, so this does that.

    The ServerMovingTests functional tests were glossing over
    this by forcing the run of the update_available_resource
    periodic task code after the instance was deleted but
    before checking the allocations were gone from Placement.
    So this change removes that part of the tests to show
    this fixes the problem.

    Change-Id: I0f57b4ba8e4c3931b70bd737de5b7c8ae47f033d
    Closes-Bug: #1712684

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/496997
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=1b2f3c7bb25667527e7e1d2afd84d5d5533e7751
Submitter: Jenkins
Branch: stable/pike

commit 1b2f3c7bb25667527e7e1d2afd84d5d5533e7751
Author: Matt Riedemann <email address hidden>
Date: Wed Aug 23 17:39:51 2017 -0400

    Delete instance allocations when the instance is deleted

    In a deployment that still has Ocata computes, the resource
    tracker will delete the allocations for an instance when it
    is deleted, but in an all-Pike+ computes deployment, the
    allocations will only be removed for a deleted instance
    during the update_available_resource periodic task.

    The capacity on the node should be freed as soon as the
    instance is deleted, like when we shelve offload an
    instance, so this does that.

    The ServerMovingTests functional tests were glossing over
    this by forcing the run of the update_available_resource
    periodic task code after the instance was deleted but
    before checking the allocations were gone from Placement.
    So this change removes that part of the tests to show
    this fixes the problem.

    Change-Id: I0f57b4ba8e4c3931b70bd737de5b7c8ae47f033d
    Closes-Bug: #1712684
    (cherry picked from commit d6de0e57c21472b99cbe71ec62b3967ef7282dec)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/nova 16.0.0.0rc2 release candidate.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/nova 17.0.0.0b1 development milestone.