Hi Loïc, Actually they are some steps that might be automated such as: * set the virsh secret * create both glance and cinder pools Please take a look at: * https://github.com/dontalton/puppet-cephdeploy/blob/master/manifests/init.pp#L121 * https://github.com/dontalton/puppet-cephdeploy/blob/master/manifests/osd.pp#L73 For the rest this might be already done but your puppet manifests. Please also note that http://ceph.com/docs/next/rbd/rbd-openstack/ will need some updates for OpenStack Havana. –––– Sébastien Han Cloud Engineer "Always give 100%. Unless you're giving blood.” Phone: +33 (0)1 49 70 99 72 Mail: sebastien.han@xxxxxxxxxxxx Address : 10, rue de la Victoire - 75009 Paris Web : www.enovance.com - Twitter : @enovance On October 8, 2013 at 4:18:00 PM, Loic Dachary (loic@xxxxxxxxxxx) wrote: Hi Ceph, Binding ceph to cinder and glance using puppet requires three steps: * Deploy ceph ( with ceph-deploy, puppet, chef ... ) * Follow the ceph documentation instructions ( valid for both cinder and glance ) http://ceph.com/docs/next/rbd/rbd-openstack/ * Part of the above instructions can be skipped if the following are used https://github.com/stackforge/puppet-glance/blob/stable/grizzly/manifests/backend/rbd.pp https://github.com/stackforge/puppet-cinder/blob/stable/grizzly/manifests/volume/rbd.pp They take care of installing a package on the glance and cinder node and writing the cinder and glance config files. * Upgrading the librbd on the compute hosts to match the version of the cluster ( ubuntu precise has bobtail but you may want at least cuttlefish ) I would be delighted to know if there is a simpler way. If not, would it make sense to provide the puppet master with the IP of the monitors and admin rights so that it can automate http://ceph.com/docs/next/rbd/rbd-openstack/ ? * install ceph-common on cinder hosts and python-ceph on glance hosts * set the monitor addresses * copy the keyring to cinder / glance * create the client.volumes / client.images users ( support <= 0.53 ? ) * upgrade the librbd package on the compute hosts to the version matching the cluster * virsh secret-set-value the volume key on each compute host * reload glance/nova/cinder where appropriate The puppet master could even refresh the list of monitors from time to time and update the cinder/glance nodes accordingly. And it could do the right thing depending on the target openstack version and ceph version. Thoughts ? -- Loïc Dachary, Artisan Logiciel Libre All that is necessary for the triumph of evil is that good people do nothing. -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
