Steps to reproduce:
1. Create cluster with the following configuration:
- CentOS
- Multi-node with HA
- nova-network (flat DHCP)
- no Savanna/Murano/Celiometer
- no Ceph
2. Add the following nodes:
- 3 controllers
- 2 computes
3. Deploy cluster.
Expected result:
Deployment finishes with success.
Actual result:
Cluster deployment fails. There are following errors in OpenStack logs:
2013-12-10 09:53:40 ERROR
nova.openstack.common.rpc.common [-] AMQP server on 10.108.100.2:5672 is unreachable: [Errno 111] ECONNREFUSED. Trying again in 5 seconds.
2013-12-10 09:53:40 ERROR
nova.openstack.common.rpc.common [-] AMQP server on 10.108.100.2:5672 is unreachable: [Errno 111] ECONNREFUSED. Trying again in 3 seconds.
2013-12-10 09:53:40 ERROR
nova.openstack.common.rpc.common [-] AMQP server on 10.108.100.2:5672 is unreachable: [Errno 111] ECONNREFUSED. Trying again in 1 seconds.
2013-12-10 09:53:40 ERROR
nova.openstack.common.rpc.common [-] Failed to consume message from queue: Socket closed
2013-12-10 09:35:29 ERROR
nova.servicegroup.drivers.db [-] Recovered model server connection!
This errors may happen due to connectivity issues. Anyway, current implementation of RPC reconnect patches recover connection to AMQP server easily. Thus, lowering the priority of this bug.