error opening pool 'nova' (libvirt-image-backend from qcow2 to rbd fails)

[Alvaro Uria]

Hi,

OpenStack version is Kilo, using juju 1.24.5.

ii nova-common 1:2015.1.1-0ubuntu1~cloud2 all OpenStack Compute - common files
ii nova-compute 1:2015.1.1-0ubuntu1~cloud2 all OpenStack Compute - compute node base
ii nova-compute-kvm 1:2015.1.1-0ubuntu1~cloud2 all OpenStack Compute - compute node (KVM)
ii nova-compute-libvirt 1:2015.1.1-0ubuntu1~cloud2 all OpenStack Compute - compute node libvirt support
ii python-nova 1:2015.1.1-0ubuntu1~cloud2 all OpenStack Compute Python libraries
ii python-novaclient 1:2.22.0-0ubuntu1~cloud0 all client library for OpenStack Compute API

On a compute node configured by default (local ephemeral storage / qcow2), if I try to change libvirt-image-backend to "rbd", ceph pool 'nova' is not created and, hence, nova-compute restart fails. See http://pastebin.ubuntu.com/12205586/.

Initial deploy with "rbd" backend works fine.

Please let me know if I can help you more.

Cheers,
-Alvaro.

[Edward Hope-Morley (hopem)]

Hi Alvaro, thanks for raising this bug. Could you please provide the following:
  * /etc/ceph/ceph.conf from the compute node in question
  * /var/log/juju/unit-nova-compute-<num>.log from your compute host
  * output of 'juju status nova-compute'
  * output of juju ssh ceph/0 sudo ceph -s

[James Page (james-page)]

A configuration change does not trigger the required changes on the ceph relation to create the appropriate pools via the ceph broker.