Comment 7 for bug 1065702

Reviewed: https://review.openstack.org/14790
Committed: http://github.com/openstack/cinder/commit/524c7fa6dfade6efcabda19b105d58cab8c434e2
Submitter: Jenkins
Branch: master

commit 524c7fa6dfade6efcabda19b105d58cab8c434e2
Author: John Griffith <email address hidden>
Date: Wed Oct 24 16:28:02 2012 -0600

    Detect and fix issues caused by vol ID migration

    The migration from volume ID to UUID neglected to update the provider_location
    field on the volume. As a result the iqn and volume name no long match and
    existing volumes are no longer able to be attached after an upgrade
    (essex -> folsom and then nova-vol->cinder).

    This patch adds a method to the volume driver that will check for the
    mismatch of volume name in the iqn during service start up. If
    detected it will update the provider_location field in the database
    to include the new ID. Also it will create a symlink to the device backing
    file that also has the correct naming convention.

    Note: We don't disturb an connections that are currently attached.
    For this case we add a check in manager.detach and do any provider_location
    cleanup that's needed at that time. This ensures that connections
    persist on restarts of tgtd and reboot.

    Change-Id: I8224824b793c98a9767c5d8dd741d892be720c4f
    Fixes: bug 1065702