RABBITMQ cluster crash on CentOS+GRE+525

[Viktoria Efimova]

Rabbit logs reports about cluster crash under rally tests:
logs on node-8 (controller_0)
=INFO REPORT==== 21-Jun-2015::16:14:41 ===
rabbit on node 'rabbit@node-15' down
..
=INFO REPORT==== 21-Jun-2015::16:14:42 ===
node 'rabbit@node-15' down: connection_closed
..
=INFO REPORT==== 21-Jun-2015::16:15:52 ===
node 'rabbit@node-15' down: connection_closed
...
=INFO REPORT==== 21-Jun-2015::16:15:52 ===
node 'rabbit@node-15' down: connection_closed

from node-8
 atop -r /var/log/atop/atop_current -b "16:14"
http://paste.openstack.org/show/311046/

from node-13
 atop -r /var/log/atop/atop_current -b "16:14"
http://paste.openstack.org/show/311047/

from node-15
 atop -r /var/log/atop/atop_current -b "16:14"
http://paste.openstack.org/show/311048/

According to the time crash happened during nova/boot_and_delete_server_with_secgroups.yaml
(use for reference for rally test running sequence http://mos-scale.vm.mirantis.net:8080/job/13_env_run_rally/33/console)

Topology:
UBUNTU+VLAN+525(RC3)
Controllers: 3
Computes: 17

Network configuration
eth0 - public
eth2 - Admin, Storage, Mgmt
eth3 - private

Snapsot at http://mos-scale-share.mirantis.com/fuel-snapshot-2015-06-21_17-39-12.tar.xz

[Bogdan Dobrelya (bogdando)]

Was the cluster recovered after the crash?

[Bogdan Dobrelya (bogdando)]

According to the logs, the failure of rabbit node-15 was recovered at 2015-06-21T16:16:01.440535+00:00, less than in 2 minutes.
Other nodes , node-8, node-13 remained running available. Marking this bug as invalid as it has nothing to be addressed and contains no issue