[SWARM 9.2] deploy_neutron_tun_ha_nodegroups test failed

[Dmitry Belyaninov]

Detailed bug description:
There is failed swarm test:

https://product-ci.infra.mirantis.net/job/9.x.system_test.ubuntu.thread_7/119/testReport/(root)/deploy_neutron_tun_ha_nodegroups/deploy_neutron_tun_ha_nodegroups/

Task 'deploy' has incorrect status. error != ready, 'Deployment has failed. All nodes are finished. Failed tasks: Task[openstack-network-server-nova/1] Stopping the deployment process!'

Logs:
File "/usr/lib/python2.7/dist-packages/neutron/db/allowedaddresspairs_db.py", line 78, in _extend_port_dict_allowed_address_pairs
address_pair in port_db.allowed_address_pairs]
AttributeError: 'dict' object has no attribute 'allowed_address_pairs'

http://paste.openstack.org/show/588382/

Steps to reproduce:
run the test
Expected results:
pass
Actual result:
fail
Reproducibility:
 <put your information here>
Workaround:
 <put your information here>
Impact:
 <put your information here>
Description of the environment:
 Operation system: <put your information here>
 Versions of components: <put your information here>
 Reference architecture: <put your information here>
 Network model: <put your information here>
 Related projects installed: <put your information here>
Additional information:
 <put your information here>

[Ann Taraday (akamyshnikova)]

I investigated this problem.

In the fail_error_deploy_neutron_tun_ha_nodegroups-fuel-snapshot-2016-11-08_01-55-43.tar error with "'allowed_address_pairs" is not shown.

I deployed an environment with 9.2 and although neutron server log have a lot of traces like http://paste.openstack.org/show/588382/ environment is completely operable and neutron-server is not stopped. I filed an upstream bug about this trace - https://bugs.launchpad.net/neutron/+bug/1640701, but I'm convinced this can not be cause of failed job.

In fail_error_deploy_neutron_tun_ha_nodegroups-fuel-snapshot-2016-11-08_01-55-43.tar I noticed a lot of traces like http://paste.openstack.org/show/588659/, so I consider that cinder team should take a look on it.

[Ivan Kolodyazhny (e0ne)]

Ann, mentioned logs are not available to cinder. It's a deployment-time issue, when fuel starts cinder before DB setyp

[Ann Taraday (akamyshnikova)]

Ivan, why this is Neutron issue? Can you please show a paste with Neutron error, which I should investigate?

I describe that "AttributeError: 'dict' object has no attribute 'allowed_address_pairs'" can not be related to job failure.

[Ann Taraday (akamyshnikova)]

I remove assignment of this bug for neutron-area.

[Oleksiy Molchanov (omolchanov)]

The problem is that node-1 went offline during environment. Marking this as incomplete, please reopen as soon as we have live env to debug.

[Oleksiy Molchanov (omolchanov)]

Marking as Invalid, because of no reproduce.