Horizon static asset compression failures ignored

[Mark Goddard]

If you run the horizon container with a restart policy of never, and the static asset compression fails, the container fails to start, e.g. (seen in CI)

++ /usr/bin/manage.py collectstatic --noinput --clear
Traceback (most recent call last):
  File "/usr/bin/manage.py", line 18, in <module>
    from django.core.management import execute_from_command_line
ImportError: No module named django.core.management

However, with a restart policy of unless-stopped (the default). the horizon container will fail the first time but will succeed after a restart. This is because the settings bundle hash is updated before the compression is attempted.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/651506

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/651506
Committed: https://git.openstack.org/cgit/openstack/kolla/commit/?id=92c2456851857c45e8de9a3ff8f167579e0588f4
Submitter: Zuul
Branch: master

commit 92c2456851857c45e8de9a3ff8f167579e0588f4
Author: Mark Goddard <email address hidden>
Date: Wed Apr 10 12:44:37 2019 +0100

    Write hash after compressing horizon static assets

    If you run the horizon container with a restart policy of never, and the
    static asset compression fails, the container fails to start, e.g. (seen
    in CI)

    ++ /usr/bin/manage.py collectstatic --noinput --clear
    Traceback (most recent call last):
      File "/usr/bin/manage.py", line 18, in <module>
          from django.core.management import execute_from_command_line
      ImportError: No module named django.core.management

    However, with a restart policy of unless-stopped (the default). the
    horizon container will fail the first time but will succeed after a
    restart. This is because the settings bundle hash is updated before the
    compression is attempted.

    To fix this, update the hash after compression completes successfully.

    Change-Id: I3f1682ed55e7d45c1e97a1b34c7993aa839f0bb2
    Closes-Bug: #1824137

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/stein
Review: https://review.openstack.org/652927

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/652927
Committed: https://git.openstack.org/cgit/openstack/kolla/commit/?id=6e010e25e3e4bf3a26a40a3a0b27055512d8a93e
Submitter: Zuul
Branch: stable/stein

commit 6e010e25e3e4bf3a26a40a3a0b27055512d8a93e
Author: Mark Goddard <email address hidden>
Date: Wed Apr 10 12:44:37 2019 +0100

    Write hash after compressing horizon static assets

    If you run the horizon container with a restart policy of never, and the
    static asset compression fails, the container fails to start, e.g. (seen
    in CI)

    ++ /usr/bin/manage.py collectstatic --noinput --clear
    Traceback (most recent call last):
      File "/usr/bin/manage.py", line 18, in <module>
          from django.core.management import execute_from_command_line
      ImportError: No module named django.core.management

    However, with a restart policy of unless-stopped (the default). the
    horizon container will fail the first time but will succeed after a
    restart. This is because the settings bundle hash is updated before the
    compression is attempted.

    To fix this, update the hash after compression completes successfully.

    Change-Id: I3f1682ed55e7d45c1e97a1b34c7993aa839f0bb2
    Closes-Bug: #1824137
    (cherry picked from commit 92c2456851857c45e8de9a3ff8f167579e0588f4)

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/rocky
Review: https://review.opendev.org/669805

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/669805
Committed: https://git.openstack.org/cgit/openstack/kolla/commit/?id=b0ef7abd54ffdd4fdc9a477ed79f95c20cfc423d
Submitter: Zuul
Branch: stable/rocky

commit b0ef7abd54ffdd4fdc9a477ed79f95c20cfc423d
Author: Mark Goddard <email address hidden>
Date: Wed Apr 10 12:44:37 2019 +0100

    Write hash after compressing horizon static assets

    If you run the horizon container with a restart policy of never, and the
    static asset compression fails, the container fails to start, e.g. (seen
    in CI)

    ++ /usr/bin/manage.py collectstatic --noinput --clear
    Traceback (most recent call last):
      File "/usr/bin/manage.py", line 18, in <module>
          from django.core.management import execute_from_command_line
      ImportError: No module named django.core.management

    However, with a restart policy of unless-stopped (the default). the
    horizon container will fail the first time but will succeed after a
    restart. This is because the settings bundle hash is updated before the
    compression is attempted.

    To fix this, update the hash after compression completes successfully.

    Change-Id: I3f1682ed55e7d45c1e97a1b34c7993aa839f0bb2
    Closes-Bug: #1824137
    (cherry picked from commit 92c2456851857c45e8de9a3ff8f167579e0588f4)
    (cherry picked from commit 6e010e25e3e4bf3a26a40a3a0b27055512d8a93e)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/kolla 8.0.0.0rc2 release candidate.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/kolla 7.0.4 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/kolla 9.0.0.0rc1 release candidate.