PowerMax Driver - Volume updates for volumes in groups not performed during failover

[Simon O'Donovan]

When performing a failover operation, any volume that is in a group will not have its model update read and saved as part of the failover. Afterwards there will be no update for these volumes in the database. Issue resides in common._populate_volume_and_group_update_lists where the volume list that has been passed in is updated. This update is perfomed against the replicated_vols variable defined in manager.py which is used to determine which volumes should updated via volume.save()

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/742861

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/742861
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=d4236cf8655e8bf0845863929858502e4e3faa24
Submitter: Zuul
Branch: master

commit d4236cf8655e8bf0845863929858502e4e3faa24
Author: odonos12 <email address hidden>
Date: Fri Jul 24 12:43:26 2020 +0100

    PowerMax Driver - Failover group vol update fix

    Replicated volumes that were in groups during failover
    do not have their model_updates saved upon return to
    manager.py as the replicated_vols list that is used to
    determine which volume should be updated is changes during
    _populate_volume_and_group_update_lists.

    Change-Id: I03efee569c7f7efec20a04755d6e7c7a60bf61ea
    Closes-Bug: 1888831

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ussuri
Review: https://review.opendev.org/752470

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/752470
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=3906b51df0b9d4e0aa58ff8d7d53d05da9598cb0
Submitter: Zuul
Branch: stable/ussuri

commit 3906b51df0b9d4e0aa58ff8d7d53d05da9598cb0
Author: odonos12 <email address hidden>
Date: Fri Jul 24 12:43:26 2020 +0100

    PowerMax Driver - Failover group vol update fix

    Replicated volumes that were in groups during failover
    do not have their model_updates saved upon return to
    manager.py as the replicated_vols list that is used to
    determine which volume should be updated is changes during
    _populate_volume_and_group_update_lists.

    Change-Id: I03efee569c7f7efec20a04755d6e7c7a60bf61ea
    Closes-Bug: 1888831
    (cherry picked from commit d4236cf8655e8bf0845863929858502e4e3faa24)