replication failback in SolidFire isn't setting portal info properly

[John Griffith]

When failing back to default using SolidFire and Cinder's replication feature the model update isn't setting the updated portal field correctly, making it impossible to attach the volume after a failback to primary. That's not cool.

[Robert Esker (esker)]

Status indicates "triaged" but I'm not clear how. Has this behavior been mitigated or is there some mechanism for workaround available? Or is "triaged" an indication of something else entirely?

[Erlon R. Cruz (sombrafam)]

From launchpad definition[1]: Triaged -> the bug supervisor considers that the bug report contains all information a developer needs to start work on a fix.

[1] https://help.launchpad.net/Bugs/Statuses

[Thiago Correa (thgcorrea)]

Do we have more information on this? Could you, please, provide some extra information about the environment (OS version, whether python 3 was enabled or not) on which this bug happens?

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/671510

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/675630
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=215d6b2b272847d695c5c5d17c9d387e123b1396
Submitter: Zuul
Branch: master

commit 215d6b2b272847d695c5c5d17c9d387e123b1396
Author: Erlon R. Cruz <email address hidden>
Date: Wed Jul 24 17:40:33 2019 -0300

    NetApp SolidFire: Fix replication

    This patch addresses several of the problems currently found in this
    driver:

     - When creating a SolidFire volume with a volume type that has
    replication enabled the replication_status of this volume was not
    changing.
     - During initialization, if a cluster is already paired, the driver
    fails to load.
     - Driver does not create volumes on the replication target on
    several configurations. For example, if the volume is being
    created from a snashot or from another volume, etc.
     - During Failback, the driver doesn't update iSCSI connection
    info to the secondary cluster nor the volume's replication status.

    Closes-bug: #1834013
    Closes-bug: #1751932
    Change-Id: I920f958db1d48e52b548082d852c03f427a279ca
    Co-Authored-By: Fernando Ferraz <email address hidden>
    Co-Authored-By: Thiago Correa <email address hidden>

[OpenStack Infra (hudson-openstack)]

Change abandoned by Thiago Correa (<email address hidden>) on branch: master
Review: https://review.opendev.org/668557
Reason: fix merged on patch: https://review.opendev.org/#/c/675630/

[OpenStack Infra (hudson-openstack)]

Change abandoned by Erlon R. Cruz (<email address hidden>) on branch: master
Review: https://review.opendev.org/671510
Reason: This was fixed in another patch for SolidFire.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/train
Review: https://review.opendev.org/686519

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/686519
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=7b227c647d5d81c1ae9e54b0e5fd76cc07c73257
Submitter: Zuul
Branch: stable/train

commit 7b227c647d5d81c1ae9e54b0e5fd76cc07c73257
Author: Erlon R. Cruz <email address hidden>
Date: Wed Jul 24 17:40:33 2019 -0300

    NetApp SolidFire: Fix replication

    This patch addresses several of the problems currently found in this
    driver:

     - When creating a SolidFire volume with a volume type that has
    replication enabled the replication_status of this volume was not
    changing.
     - During initialization, if a cluster is already paired, the driver
    fails to load.
     - Driver does not create volumes on the replication target on
    several configurations. For example, if the volume is being
    created from a snashot or from another volume, etc.
     - During Failback, the driver doesn't update iSCSI connection
    info to the secondary cluster nor the volume's replication status.

    Closes-bug: #1834013
    Closes-bug: #1751932
    Change-Id: I920f958db1d48e52b548082d852c03f427a279ca
    Co-Authored-By: Fernando Ferraz <email address hidden>
    Co-Authored-By: Thiago Correa <email address hidden>
    (cherry picked from commit 215d6b2b272847d695c5c5d17c9d387e123b1396)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/cinder 15.0.0.0rc2 release candidate.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/cinder 16.0.0.0b1 development milestone.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/stein
Review: https://review.opendev.org/729657

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/729657
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=5fec81daa6b0cd694bb8987abdd88e0e6106bdfc
Submitter: Zuul
Branch: stable/stein

commit 5fec81daa6b0cd694bb8987abdd88e0e6106bdfc
Author: Erlon R. Cruz <email address hidden>
Date: Wed Jul 24 17:40:33 2019 -0300

    NetApp SolidFire: Fix replication

    This patch addresses several of the problems currently found in this
    driver:

     - When creating a SolidFire volume with a volume type that has
    replication enabled the replication_status of this volume was not
    changing.
     - During initialization, if a cluster is already paired, the driver
    fails to load.
     - Driver does not create volumes on the replication target on
    several configurations. For example, if the volume is being
    created from a snashot or from another volume, etc.
     - During Failback, the driver doesn't update iSCSI connection
    info to the secondary cluster nor the volume's replication status.

    Closes-bug: #1834013
    Closes-bug: #1751932
    Change-Id: I920f958db1d48e52b548082d852c03f427a279ca
    Co-Authored-By: Fernando Ferraz <email address hidden>
    Co-Authored-By: Thiago Correa <email address hidden>
    (cherry picked from commit 215d6b2b272847d695c5c5d17c9d387e123b1396)
    (cherry picked from commit 7b227c647d5d81c1ae9e54b0e5fd76cc07c73257)