compute CDM problem

[suzhengwei]

I create a continus audit used vm_workload_consolidation strategy, and find one problem.
Once a compute node is set disabled, it will be removed from self._compute_model, because scoped_model only return enable node. There may be some problem with vm_workload_consolidation strategy. If all enabled compute nodes are overloaded, the instance can't migrate to disabled nodes.

The disabled compute nodes in one availability zone is useful to watcher's strategies. But limited to nova api, we only can get the active compute nodes in the availability zone, so we build an imprecise compute CDM(without disabled compute nodes) in watcher.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/489440
Committed: https://git.openstack.org/cgit/openstack/watcher/commit/?id=783627626c7a1d5156272cb3d61faa0a827e023c
Submitter: Jenkins
Branch: master

commit 783627626c7a1d5156272cb3d61faa0a827e023c
Author: Hidekazu Nakamura <email address hidden>
Date: Tue Aug 1 10:16:26 2017 +0900

    Fix compute CDM to include disabled compute node

    Currently compute CDM excludes disabled compute node.
    This patch set fixes compute CDM to include disabled compute node.

    Change-Id: I8236bb73e0d9bb242251c2abfb59ad5693087afa
    Closes-Bug: #1685787

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/watcher 1.4.0 release.