When deploying or upgrading to a Red Hat OpenStack 10 environment that uses Ceph and dedicated blockstorage nodes for LVM, creating instances with attached volumes will no longer work. This is caused by a bug in the way the director configures the Block Storage service during upgrades. Specifically, the heat templates do not account by default for cases where Ceph and dedicated blockstorage nodes are configured together. As such, the director fails to define some required settings. Note that LVM is not a suitable Block Storage back end in production, particularly in enterprise environments. To work around this add an environment file to your upgrade/deployment that contains the following: parameter_defaults: BlockStorageExtraConfig: tripleo::profile::base::cinder::volume::cinder_enable_iscsi_backend: true tripleo::profile::base::cinder::volume::cinder_enable_rbd_backend: false