I deployed RDO Icehouse. I have one controller with eth0 to external network, eth1 to internal network where I plan to add more compute nodes. Controller is basically all-in-one. I am able to create volumes, VM, connect to those VM. Everything perfect.

So I added another machine connected to that private network and configured it as compute node. I followed:

http://docs.openstack.org/icehouse/install-guide/install/yum/content/nova-compute.html
http://docs.openstack.org/icehouse/install-guide/install/yum/content/neutron-ml2-compute-node.html

But my controller does not use that new compute node. On compute node I have:

/var/log/nova/nova-compute.log

full of these lines:

2015-01-23 09:12:21.729 2330 ERROR oslo.messaging._drivers.impl_qpid [-] Unable to connect to AMQP server: [Errno 113] EHOSTUNREACH. Sleeping 5 seconds

The routing itself should be OK as I'm able to:

telnet controller 443

from compute node successfully.

What I see as potentional problem is that on controller there is:

# service qpidd status
Redirecting to /bin/systemctl status  qpidd.service qpidd.service
Loaded: not-found (Reason: No such file or directory)
Active: inactive (dead)

And I could not find any documentation that I should configure qpidd on controller . Well. I find this: https://openstack.redhat.com/Securing... but that is just about hardening. But I could not find anything about initial config.

Did I miss something?