rbd: timeout when create volume from snapshot

Error is raised in cinder when I boot VMs from a VM snapshot.
I use rbd as cinder backend, here is the config about ceph:

[ceph1]
volume_driver = cinder.volume.drivers.rbd.RBDDriver
rbd_pool = volumes
rbd_ceph_conf = /etc/ceph/ceph.conf
rbd_flatten_volume_from_snapshot = true
rbd_max_clone_depth = 5
rbd_store_chunk_size = 4
rados_connect_timeout = -1
volume_backend_name = ceph1

Error info in cinder-volume:
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server [-] Can not acknowledge message. Skip processing
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server Traceback (most recent call last):
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/server.py", line 126, in _process_incoming
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server message.acknowledge()
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server File "/usr/lib/python2.7/site-packages/oslo_messaging/_drivers/amqpdriver.py", line 119, in acknowledge
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server self.message.acknowledge()
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server File "/usr/lib/python2.7/site-packages/oslo_messaging/_drivers/impl_rabbit.py", line 251, in acknowledge
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server self._raw_message.ack()
......
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server return self._send_loop(self.fd.send, data, flags)
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server File "/usr/lib/python2.7/site-packages/eventlet/greenio/base.py", line 366, in _send_loop
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server return send_method(data, *args)
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server error: [Errno 32] Broken pipe
2017-01-20 17:09:35.812 14085 ERROR oslo_messaging.rpc.server
2017-01-20 17:09:35.838 14085 DEBUG cinder.volume.flows.manager.create_volume [req-3bbc145a-7453-4f9b-885b-04d6ebfdbf75 1e7c6b8b585e44fb9b2aacc2d0c52f78 fe48a5e0b8e1439abb423fb4a025d064 - default default] Marking volume 898121a2-3251-42bb-afac-4f2d51210a83 as bootable. _enable_bootable_flag /usr/lib/python2.7/site-packages/cinder/volume/flows/manager/create_volume.py:464

This error occuers only when rbd_flatten_volume_from_snapshot is set to true;It always means heartbeat between rabbitmq-server and cinder-volume is missing;

Rbd driver will flatten the volume when rbd_flatten_volume_from_snapshot is set,we shouldn't do this time-consuming operation in greenthread. Use tpool proxy instead may be a good way.

This is similar to https://review.openstack.org/#/c/175555/