Functional test intermittent failure: test_volume_snapshot

[Julie Pichon]

Some of the volume functional tests appear to fail intermittently due to deletion issues:

openstackclient.tests.functional.volume.v2.test_volume.VolumeTests.test_volume_snapshot
---------------------------------------------------------------------------------------
Captured traceback:
~~~~~~~~~~~~~~~~~~~
    Traceback (most recent call last):
      File "openstackclient/tests/functional/base.py", line 52, in openstack
        return execute('openstack ' + cmd, fail_ok=fail_ok)
      File "openstackclient/tests/functional/base.py", line 41, in execute
        result_err)
    tempest.lib.exceptions.CommandFailed: Command 'openstack volume delete 8bf70e4fb3c84d3ba4ad9037d2bb92d7' returned non-zero exit status 1.

      stderr:
    Failed to delete volume with name or ID '8bf70e4fb3c84d3ba4ad9037d2bb92d7': Invalid volume: Volume status must be available or error or error_restoring or error_extending or error_managing and must not be migrating, attached, belong to a group or have snapshots. (HTTP 400) (Request-ID: req-833d1731-b2e1-4ae9-8722-8628cbc72329)
     1 of 1 volumes failed to delete.

This appears to be particularly frequent on the stable/ocata branch (example: http://logs.openstack.org/84/493884/1/check/gate-osc-dsvm-functional-ubuntu-xenial/f043616/console.html.gz#_2017-08-21_10_53_39_943107 ).

It also sounds similar to the description in bug 1698085 (for a different test), perhaps the solution will be along the same lines.

[Julie Pichon (jpichon)]

https://review.openstack.org/#/c/466507/ looks promising as a backport that may help to resolve the issue.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ocata
Review: https://review.openstack.org/501191

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/501191
Committed: https://git.openstack.org/cgit/openstack/python-openstackclient/commit/?id=5eb85674668e46fd657f150c77239b6a980eb2f5
Submitter: Jenkins
Branch: stable/ocata

commit 5eb85674668e46fd657f150c77239b6a980eb2f5
Author: Akihiro Motoki <email address hidden>
Date: Sun May 21 03:36:23 2017 +0000

    volume functest: ensure snapshots deleted when volume delete

    Deleting snapshot may take time. The current volume API does not allow
    to delete volumes with snapshots, so if deleting snapshot may take time,
    a delete request for a parent volume will fail.
    This sometimes causes functional test failures in slow environments.

    wait_for_status() checks whether volume status is in error statuses
    but previously the expected error status was wrong. Cinder API uses
    lower case as volume status, so it did not work expectedly.

    Conflicts:
      openstackclient/tests/functional/compute/v2/common.py
      openstackclient/tests/functional/compute/v2/test_server.py
      openstackclient/tests/functional/volume/v1/test_volume.py
      openstackclient/tests/functional/volume/v2/test_volume.py
      openstackclient/tests/functional/volume/v3/common.py

    Change-Id: I095894ba39f23bf81d71351818d24dbb5ca459fb
    (cherry picked from commit eeb614c47759fa9a01e6d886ed07acceb8d9ff61)
    Closes-Bug: #1714977

[Julie Pichon (jpichon)]

Closing as I think this was a stable/ocata issue only (I can't set "Target to series").

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/python-openstackclient 3.8.2 release.