yahoo-eng-team team mailing list archive

[Adam Gandelman <1438920@xxxxxxxxxxxxxxxxxx>]

** Changed in: nova/juno
       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/1438920

Title:
  specification of a nova network to instance during icehouse->juno
  partial upgrade fails

Status in OpenStack Compute (Nova):
  Fix Committed
Status in OpenStack Compute (nova) juno series:
  Fix Released

Bug description:
  Tempest recently made some changes around specifying fixed network to
  instances during spawn.  When this happens during an icehouse->juno
  partial upgrade, we fail to spawn instances and they are stuck in
  scheduling.

  Conductor logs show:

  oslo.messaging.rpc.dispatcher [req-047d24c2-6477-4838-afde-
  cc3c7371b19e VolumesV2ActionsTest-21337662
  VolumesV2ActionsTest-1578486140] Exception during message handling:
  need more than 2 values to unpack

  It looks like there is code in Nova to handle this case during an icehouse partial upgrade, but only takes into account doing so with Neutron:
  https://git.openstack.org/cgit/openstack/nova/tree/nova/compute/rpcapi.py#n982  This needs some special casing to handle NetworkRequestList objects for nova-network, as well.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1438920/+subscriptions