Description of problem:

In the basic situation, the following command shows the list of consumes for rabbitmq cluster.

 $ sudo docker exec $(sudo docker ps -f name=rabbitmq-bundle -q) rabbitmqctl list_consumers

However, after we restart all the controller nodes one by one, we can not get the result by the command
as the command get stacked at some point.

~~~
[heat-admin@controller-0 ~]$ sudo docker exec $(sudo docker ps -f name=rabbitmq-bundle -q) rabbitmqctl list_consumers 
Listing consumers
l3_agent_fanout_73c3e66599a648379e3149a85af6bd8e	<rabbit.5979.0>	3	true	0	[]
q-l3-plugin_fanout_b2e4eb3ef5d641dfb189002c4d6f9133	<rabbit.10747.0>	3	true	0	[]
...
neutron-vo-SubPort-1.0.compute-0.localdomain	<rabbit.6308.0>	2	true	0	[]
q-reports-plugin_fanout_46dab178092e4ac986bb4bbad877ac99	<rabbit.6053.0>	3	true	0	[]
q-reports-plugin_fanout_94b1171459d24ad5b79f483902c03223	<rabbit.5904.0>	3	true	0	[]
trunk_fanout_b5424ee9eeaf47f480394907b4e03e49	<rabbit.6260.0>	3	true	0	[]
~~~

We do not see any specific error in console, and neither in rabbitmq logs.
Also, to recover from this situation, we need to restart whole the controller nodes.


Version-Release number of selected component (if applicable):

~~~
$ sudo docker images | grep rabbitmq
192.168.24.1:8787/rhosp13/openstack-rabbitmq                    2019-01-10.1        766efb5b9b38        3 months ago        635 MB
192.168.24.1:8787/rhosp13/openstack-rabbitmq                    pcmklatest          766efb5b9b38        3 months ago        635 MB
$ sudo docker exec $(sudo docker ps -f name=rabbitmq-bundle -q) rpm -qa | grep rabbitmq
rabbitmq-server-3.6.15-3.el7ost.noarch
puppet-rabbitmq-8.1.1-0.20180216013831.d4b06b7.el7ost.noarch
~~~


RHOSP13

How reproducible:

Always

Steps to Reproduce:
1. Restart all the controllers one by one, by "sudo reboot"
2. run the said command in one of the controller

Actual results:

The command get stuck

Expected results:

The command should return the complete information without any error or stuck.

Additional info: