Huawei driver fails to detach volume after system reboot

[zhangchao]

Huawei T and Dorado drivers fail to detach volumes after the cinder node host sys reboot.
The host group id is created only when attaching volumes. After sys reboot, the host group id will be none. In the case, if detaching the volumes, we will get CLI exceptions. So the driver should get host group id while initializing.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/48293

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/48293
Committed: http://github.com/openstack/cinder/commit/cdcece14267560eb2641a2cab1f187897ed25ed2
Submitter: Jenkins
Branch: master

commit cdcece14267560eb2641a2cab1f187897ed25ed2
Author: zhangchao010 <email address hidden>
Date: Thu Sep 26 00:00:20 2013 +0800

    Get host group id when Huawei driver initializing

    Get host group id when Huawei driver initializing to make sure
    host group id won't be None after cinder node host sys reboot.

    fixes bug 1230372

    Change-Id: Ia3d5de116804930b2bb0035e4c991ebb9801ee76