Nodes mark as available after changing master

[Egor Kotko]

{"build_id": "2015-06-02_16-28-25", "build_number": "497", "release_versions": {"2014.2.2-6.1": {"VERSION": {"build_id": "2015-06-02_16-28-25", "build_number": "497", "api": "1.0", "fuel-library_sha": "d757cd41e4f8273d36ef85b8207e554e5422c5b0", "nailgun_sha": "3830bdcb28ec050eed399fe782cc3dd5fbf31bde", "feature_groups": ["mirantis"], "openstack_version": "2014.2.2-6.1", "production": "docker", "python-fuelclient_sha": "4fc55db0265bbf39c369df398b9dc7d6469ba13b", "astute_sha": "cbae24e9904be2ff8d1d49c0c48d1bdc33574228", "fuel-ostf_sha": "f899e16c4ce9a60f94e7128ecde1324ea41d09d4", "release": "6.1", "fuelmain_sha": "bcc909ffc5dd5156ba54cae348b6a07c1b607b24"}}}, "auth_required": true, "api": "1.0", "fuel-library_sha": "d757cd41e4f8273d36ef85b8207e554e5422c5b0", "nailgun_sha": "3830bdcb28ec050eed399fe782cc3dd5fbf31bde", "feature_groups": ["mirantis"], "openstack_version": "2014.2.2-6.1", "production": "docker", "python-fuelclient_sha": "4fc55db0265bbf39c369df398b9dc7d6469ba13b", "astute_sha": "cbae24e9904be2ff8d1d49c0c48d1bdc33574228", "fuel-ostf_sha": "f899e16c4ce9a60f94e7128ecde1324ea41d09d4", "release": "6.1", "fuelmain_sha": "bcc909ffc5dd5156ba54cae348b6a07c1b607b24"}

Steps to reproduce:
1. Setup master (Virtual machine)
2. Wait for bootstrap of slaves (Hardware machines)
3. Delete old && Setup new master node

Actual result:
The old slaves alert as available for using on new master. But its were not rebooted
http://paste.openstack.org/show/259125/

If create the environment with this nodes and check network appears the message:
Verification failed.
Method verify_networks. Network verification not avaliable because nodes ["2", "3"] not avaliable via mcollective. Inspect Astute logs for the details

Astute log contains:
http://paste.openstack.org/show/259157/

[Dmitry Pyzhov (dpyzhov)]

Workaround here is to reboot nodes manually. There is no correct way to reboot them automatically because in general case we are not able to do it because our ssh key can differ.

In 7.0 we are going to mark this kind of nodes as 'alien' on UI and recommend to reboot them.

[Vladimir Sharshov (vsharshov)]

Main problem here: Mcollective service run with old credentials. In this case we could not run any usual tasks: reboot, erase, provisioning, network check.

[Dmitry Borodaenko (angdraug)]

Isn't this a duplicate of bug #1431386?