Rabbitmq doesn't start during cluster re-installation

[Sergey Novikov]

Detailed bug description:

 Test https://github.com/openstack/fuel-qa/blob/master/fuelweb_test/tests/test_node_reinstallation.py#L355-L386 fails with https://product-ci.infra.mirantis.net/job/9.x.system_test.ubuntu.error_node_reinstallation/42/testReport/(root)/reinstall_failed_primary_controller_deployment/reinstall_failed_primary_controller_deployment/

 In two words, the cluster re-installation after the first failed deployment on primary controller fails on task "primary-rabbitmq". Rabbitmq doesn't start due to the following error:

<30>Aug 29 06:24:38 node-1 lrmd: INFO: p_rabbitmq-server[27889]: get_status(): failed with code 69. Command output: Error: unable to connect to node 'rabbit@messaging-node-1': nodedown

Steps to reproduce:
  1. Create a cluster
  2. Add 3 nodes with controller and mongo roles
  3. Add a node with compute and cinder roles
  4. Provision nodes
  5. Start deployment; fail deployment on primary controller
  6. Reinstall the cluster
  7. Run network verification
  8. Run OSTF

Fuel version http://paste.openstack.org/show/564578/

[Kyrylo Galanov (kgalanov)]

It looks like an environment issue. Please reopen the bug if the issue comes back again.

[Oleksiy Molchanov (omolchanov)]

Marking as Invalid, because of no update for more than a month.