yahoo-eng-team team mailing list archive

[Thierry Carrez <thierry.carrez+lp@xxxxxxxxx>]

** Changed in: nova
       Status: Fix Committed => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1433049

Title:
  libvirt-xen: Instance status in nova may be different than real status

Status in OpenStack Compute (Nova):
  Fix Released

Bug description:
  Tempest test
  ServerActionsTestJSON:test_resize_server_confirm_from_stopped and
  other similaire test from_stopped may fail with libvirt-xen driver due
  to the test timing out on waiting the instance to be SHUTOFF, but nova
  is reporting the instance to be ACTIVE.

  Traceback (most recent call last):
    File "/opt/stack/tempest/tempest/api/compute/servers/test_server_actions.py", line 230, in test_resize_server_confirm_from_stopped
      self._test_resize_server_confirm(stop=True)
    File "/opt/stack/tempest/tempest/api/compute/servers/test_server_actions.py", line 209, in _test_resize_server_confirm
      self.client.wait_for_server_status(self.server_id, expected_status)
    File "/opt/stack/tempest/tempest/services/compute/json/servers_client.py", line 183, in wait_for_server_status
      ready_wait=ready_wait)
    File "/opt/stack/tempest/tempest/common/waiters.py", line 93, in wait_for_server_status
      raise exceptions.TimeoutException(message)
  tempest.exceptions.TimeoutException: Request timed out
  Details: (ServerActionsTestJSON:test_resize_server_confirm_from_stopped) Server a0f07187-4e08-4664-ad48-a03cffb87873 failed to reach SHUTOFF status and task state "None" within the required time (196 s). Current status: ACTIVE. Current task state: None.

  
  From nova log, I could see "VM Started (Lifecycle Event)" being reported while the instance is shutdown and being resized.

  After tracking done this bug, the issue may comes from the Change-Id
  I690d3d700ab4d057554350da143ff77d78b509c6, Delay STOPPED lifecycle
  event for Xen domains.

  A way to reproduce would be to run this script on a Xen machine using a small Cirros instance:
  nova boot --image 'cirros-0.3.2-x86_64-uec' --flavor 42 instance
  nova stop instance
  # wait sometime (around 20s) so we start with SHUTDOWN state
  nova start instance
  nova stop instance
  nova resize instance 84
  nova resize-confirm instance
  # check new state, should be shutoff.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1433049/+subscriptions