Hi! does anything on a compute node need direct database access? I'm attempting to discover how to setup a separate compute node on ubuntu, grizzly, puppet modules I'm looking at a combination of the puppet modules (which may be focused on folsom) and these instructions: http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_compute.html#basic-install_compute-intro I don't see any setting for telling nova-compute the location of nova-conductor... instead, I see a bunch of sql configuration. Does anyone know if compute nodes require a database connection to any of the services on the controller? Thanks, - Don -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20130613/70af86ab/attachment.html>