330_enforce_mitaka_online_migrations blocks on soft-deleted records

Bug #1665719 reported by melanie witt on 2017-02-17
14
This bug affects 2 people
Affects Status Importance Assigned to Milestone
OpenStack Compute (nova)
High
melanie witt
Newton
High
melanie witt
Ocata
High
melanie witt
Ubuntu
Undecided
Unassigned

Bug Description

During an upgrade from mitaka -> newton, the 330_enforce_mitaka_online_migrations blocking migration considers soft-deleted records as unmigrated, but the online migrations only migrate non-deleted records. So even after running all of the online migrations, operators can get stuck until they can hard delete any soft-deleted records they have.

Fix proposed to branch: master
Review: https://review.openstack.org/435546

Changed in nova:
status: New → In Progress
melanie witt (melwitt) on 2017-02-17
tags: added: newton-backport-potential ocata-rc-potential
tags: added: upgrades
Matt Riedemann (mriedem) on 2017-02-17
Changed in nova:
importance: Undecided → High
tags: added: ocata-backport-potential
removed: ocata-rc-potential

Reviewed: https://review.openstack.org/435546
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=6d64b7274410ae45b95bd7ac7f702c16daaa0fcd
Submitter: Jenkins
Branch: master

commit 6d64b7274410ae45b95bd7ac7f702c16daaa0fcd
Author: melanie witt <email address hidden>
Date: Fri Feb 17 17:27:57 2017 +0000

    Skip soft-deleted records in 330_enforce_mitaka_online_migrations

    The 330_enforce_mitaka_online_migrations migration considers
    soft-deleted records as unmigrated (the blocker migration uses the
    select function from sqlalchemy), but the online migrations only
    migrate non-deleted records (the migrations use the model_query
    function which defaults to read_deleted='no'). So even after running
    all of the online migrations, operators can get stuck until they can
    hard delete any soft-deleted compute_nodes, aggregates, and
    pci_devices records they have.

    Closes-Bug: #1665719

    Change-Id: I2285005098b7dab7753366f53667ff8d4532d668

Changed in nova:
status: In Progress → Fix Released

Reviewed: https://review.openstack.org/435620
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=642caf0c58ed0a0e0c06103d2e3ef001c1ae142f
Submitter: Jenkins
Branch: stable/newton

commit 642caf0c58ed0a0e0c06103d2e3ef001c1ae142f
Author: melanie witt <email address hidden>
Date: Fri Feb 17 17:27:57 2017 +0000

    Skip soft-deleted records in 330_enforce_mitaka_online_migrations

    The 330_enforce_mitaka_online_migrations migration considers
    soft-deleted records as unmigrated (the blocker migration uses the
    select function from sqlalchemy), but the online migrations only
    migrate non-deleted records (the migrations use the model_query
    function which defaults to read_deleted='no'). So even after running
    all of the online migrations, operators can get stuck until they can
    hard delete any soft-deleted compute_nodes, aggregates, and
    pci_devices records they have.

     Conflicts:
     nova/tests/unit/db/test_sqlalchemy_migration.py

    NOTE(melwitt): The conflict is due to ocata unit tests that don't
    exist in newton.

    Closes-Bug: #1665719

    Change-Id: I2285005098b7dab7753366f53667ff8d4532d668
    (cherry picked from commit 6d64b7274410ae45b95bd7ac7f702c16daaa0fcd)

Reviewed: https://review.openstack.org/435619
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=10a32ddcf85076266c3ede6e8ca2e9776d64d32e
Submitter: Jenkins
Branch: stable/ocata

commit 10a32ddcf85076266c3ede6e8ca2e9776d64d32e
Author: melanie witt <email address hidden>
Date: Fri Feb 17 17:27:57 2017 +0000

    Skip soft-deleted records in 330_enforce_mitaka_online_migrations

    The 330_enforce_mitaka_online_migrations migration considers
    soft-deleted records as unmigrated (the blocker migration uses the
    select function from sqlalchemy), but the online migrations only
    migrate non-deleted records (the migrations use the model_query
    function which defaults to read_deleted='no'). So even after running
    all of the online migrations, operators can get stuck until they can
    hard delete any soft-deleted compute_nodes, aggregates, and
    pci_devices records they have.

    Closes-Bug: #1665719

    Change-Id: I2285005098b7dab7753366f53667ff8d4532d668
    (cherry picked from commit 6d64b7274410ae45b95bd7ac7f702c16daaa0fcd)

This issue was fixed in the openstack/nova 14.0.4 release.

This issue was fixed in the openstack/nova 15.0.1 release.

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

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Duplicates of this bug

Other bug subscribers