Backup and Restore playbook is not 100% idempotent when running in snapshot mode

[Juan Larriba]

If the user cancels the playbook while it was taking a rear backup of a node, the next playbook execution fails due to pacemaker having stopped the mysql-galera database.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/openstack/tripleo-ansible/+/799829

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/tripleo-ansible/+/799829
Committed: https://opendev.org/openstack/tripleo-ansible/commit/aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2
Submitter: "Zuul (22348)"
Branch: master

commit aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2
Author: Juan Larriba <email address hidden>
Date: Wed Jul 7 15:30:04 2021 +0200

    Backup and Restore playbook is not 100% idempotent when running in snapshot mode

    This fix a small issue that can happen when the user directly cancels
    the execution of the playbook when the rear backup is happening. After
    that, the node remains disconnected from pacemaker and it is necessary
    to add it back to the cluster so the mysql-galera container is running
    when we do the mysql dump.

    As the pcs cluster start is an idempotent command in itself, we do not
    need to perform a verification check before running it nor it will
    generate errors (return 1) when ran if it was already running.

    Closes-Bug: #1934897

    Change-Id: I9a425665f4714751d8b1a0bd381533e4c383df9d

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/wallaby
Review: https://review.opendev.org/c/openstack/tripleo-ansible/+/799994

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/tripleo-ansible/+/799994
Committed: https://opendev.org/openstack/tripleo-ansible/commit/a22d8d9817e07b76ccd83211659fea55796b735b
Submitter: "Zuul (22348)"
Branch: stable/wallaby

commit a22d8d9817e07b76ccd83211659fea55796b735b
Author: Juan Larriba <email address hidden>
Date: Wed Jul 7 15:30:04 2021 +0200

    Backup and Restore playbook is not 100% idempotent when running in snapshot mode

    This fix a small issue that can happen when the user directly cancels
    the execution of the playbook when the rear backup is happening. After
    that, the node remains disconnected from pacemaker and it is necessary
    to add it back to the cluster so the mysql-galera container is running
    when we do the mysql dump.

    As the pcs cluster start is an idempotent command in itself, we do not
    need to perform a verification check before running it nor it will
    generate errors (return 1) when ran if it was already running.

    Closes-Bug: #1934897

    Change-Id: I9a425665f4714751d8b1a0bd381533e4c383df9d
    (cherry picked from commit aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2)

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/victoria
Review: https://review.opendev.org/c/openstack/tripleo-ansible/+/801244

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ussuri
Review: https://review.opendev.org/c/openstack/tripleo-ansible/+/801245

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/tripleo-ansible/+/801244
Committed: https://opendev.org/openstack/tripleo-ansible/commit/66ffc5f52671e4308c97d5db83ee06f992d07855
Submitter: "Zuul (22348)"
Branch: stable/victoria

commit 66ffc5f52671e4308c97d5db83ee06f992d07855
Author: Juan Larriba <email address hidden>
Date: Wed Jul 7 15:30:04 2021 +0200

    Backup and Restore playbook is not 100% idempotent when running in snapshot mode

    This fix a small issue that can happen when the user directly cancels
    the execution of the playbook when the rear backup is happening. After
    that, the node remains disconnected from pacemaker and it is necessary
    to add it back to the cluster so the mysql-galera container is running
    when we do the mysql dump.

    As the pcs cluster start is an idempotent command in itself, we do not
    need to perform a verification check before running it nor it will
    generate errors (return 1) when ran if it was already running.

    Closes-Bug: #1934897

    Change-Id: I9a425665f4714751d8b1a0bd381533e4c383df9d
    (cherry picked from commit aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2)
    (cherry picked from commit a22d8d9817e07b76ccd83211659fea55796b735b)

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/tripleo-ansible/+/801245
Committed: https://opendev.org/openstack/tripleo-ansible/commit/a1ae19060bc60cab8638b5e410cf54dc7aa349b9
Submitter: "Zuul (22348)"
Branch: stable/ussuri

commit a1ae19060bc60cab8638b5e410cf54dc7aa349b9
Author: Juan Larriba <email address hidden>
Date: Wed Jul 7 15:30:04 2021 +0200

    Backup and Restore playbook is not 100% idempotent when running in snapshot mode

    This fix a small issue that can happen when the user directly cancels
    the execution of the playbook when the rear backup is happening. After
    that, the node remains disconnected from pacemaker and it is necessary
    to add it back to the cluster so the mysql-galera container is running
    when we do the mysql dump.

    As the pcs cluster start is an idempotent command in itself, we do not
    need to perform a verification check before running it nor it will
    generate errors (return 1) when ran if it was already running.

    Closes-Bug: #1934897

    Change-Id: I9a425665f4714751d8b1a0bd381533e4c383df9d
    (cherry picked from commit aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2)
    (cherry picked from commit a22d8d9817e07b76ccd83211659fea55796b735b)

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/train
Review: https://review.opendev.org/c/openstack/tripleo-ansible/+/801725

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/tripleo-ansible/+/801725
Committed: https://opendev.org/openstack/tripleo-ansible/commit/6b0c098586cb418d97d05fce720fc0fecd3df511
Submitter: "Zuul (22348)"
Branch: stable/train

commit 6b0c098586cb418d97d05fce720fc0fecd3df511
Author: Juan Larriba <email address hidden>
Date: Wed Jul 7 15:30:04 2021 +0200

    Backup and Restore playbook is not 100% idempotent when running in snapshot mode

    This fix a small issue that can happen when the user directly cancels
    the execution of the playbook when the rear backup is happening. After
    that, the node remains disconnected from pacemaker and it is necessary
    to add it back to the cluster so the mysql-galera container is running
    when we do the mysql dump.

    As the pcs cluster start is an idempotent command in itself, we do not
    need to perform a verification check before running it nor it will
    generate errors (return 1) when ran if it was already running.

    Not a clean cherry pick due to different naming of the role directory.

    Closes-Bug: #1934897

    Change-Id: I9a425665f4714751d8b1a0bd381533e4c383df9d
    (cherry picked from commit aa93f9d7a05d33bdf0aa6177da10f51a1e784bc2)
    (cherry picked from commit a22d8d9817e07b76ccd83211659fea55796b735b)
    (cherry picked from commit a1ae19060bc60cab8638b5e410cf54dc7aa349b9)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tripleo-ansible 1.5.5 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tripleo-ansible 3.3.0 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tripleo-ansible 2.5.0 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tripleo-ansible 4.1.0 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tripleo-ansible train-eol release.