Improper behavior for reset environment

[Nastya Urlapova]

iso #225

Deploy simple cluster:
1xController
1xCompute

After successful deploy push "Reset environment".
Add some new option for cluster and add one compute node for example.

Old stuff have discovered status and possible to deploy changes.

[Evgeniy L (rustyrobot)]

We have here several problems

1. dissipated node, somehow this node was assigned to another cluster (see nailgun/api.log)

2. deployment wasn't started on nodes which were reseted, it looks like deploy button was pressed before nodes was moved to discovery state

[Vladimir Sharshov (vsharshov)]

Could not reproduce. Try twice: 1 - add new node after succeed reset env and nodes back to online, 2 - add nodes after succeed and nodes was offline. In both case deploy succeed.

[Evgeniy L (rustyrobot)]

1. we assume that first problem was occured because system tests assigned this node to another cluster

2. problem 2 is in nailgun, in case of inaccessible nodes nailgun doesn't update their statuses

2014-02-28 08:27:16.895 INFO [7fe5d6683700] (receiver) RPC method reset_environment_resp received: {"status": "ready", "progress": 100, "task_uuid": "fe837ea6-4123-4bf1-8bba-2f64611b44a0", "inaccessible_nodes": [{"uid": "6", "error": "Node not answered by RPC."}, {"uid": "7", "error": "Node not answered by RPC."}], "nodes": []}

https://github.com/stackforge/fuel-web/blob/master/nailgun/nailgun/rpc/receiver.py#L528-L533

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/77584
Committed: https://git.openstack.org/cgit/stackforge/fuel-web/commit/?id=507f5085e53f7210b810b238a844efa3fb40f633
Submitter: Jenkins
Branch: master

commit 507f5085e53f7210b810b238a844efa3fb40f633
Author: Nikolay Markov <email address hidden>
Date: Mon Mar 3 17:09:12 2014 +0400

    Inaccessible nodes are also updated on stop/reset

    Closes-Bug: #1286024

    Change-Id: Ic7bcb3d71cebfb0c6d10f02490fa32dc5d4ef909

[Anastasiia Naboikina (anaboikina)]

Verified on iso 235.

[Vladimir Kuklin (vkuklin)]

should be backported to stable/4.1

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/4.1
Review: https://review.openstack.org/84137

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/84137
Committed: https://git.openstack.org/cgit/stackforge/fuel-web/commit/?id=a2ae66f22dd7fff9cb4cfaae87aca4e6c7aad331
Submitter: Jenkins
Branch: stable/4.1

commit a2ae66f22dd7fff9cb4cfaae87aca4e6c7aad331
Author: Nikolay Markov <email address hidden>
Date: Mon Mar 3 17:09:12 2014 +0400

    Inaccessible nodes are also updated on stop/reset

    Closes-Bug: #1286024

    Change-Id: Ic7bcb3d71cebfb0c6d10f02490fa32dc5d4ef909

[Andrey Sledzinskiy (asledzinskiy)]

verified on {"build_id": "2014-04-03_04-17-26", "mirantis": "yes", "build_number": "266", "nailgun_sha": "7a05e365240ab27c492b20585ef8ac8557102cc0", "ostf_sha": "de0222fed646525d248dc6892eeceab139d5c469", "fuelmain_sha": "16637e2ea0ae6fe9a773aceb9d76c6e3a75f6c3b", "astute_sha": "5bcacc84cdaee3b31f1178d92a1c0681dc6ce520", "release": "4.1", "fuellib_sha": "52e7f57695f33bafa5d84d524d77f1bc3a2289b2"}