+++ This bug was initially created as a clone of Bug #1710557 +++ Given the following deployment: - central control plane - edge site1 - edge site 2 - edge site 3 Where glance cache [1] is deployed to each edge site, there is a usecase to precache the glance images from the central site to edge sites 1 and 2, but not 3. The precaching operation for image foo can be achieved by running glance and/or nova commands on edge nodes. Running these commands across all edge nodes via Ansible will be useful for operators. This RFE tracks providing this set of playbooks. An appropriate place for them is probably [2]. This RFE depends on having the inventory produced by 1710078. [1] https://blueprints.launchpad.net/tripleo/+spec/split-controlplane-glance-cache [2] https://github.com/openstack/tripleo-ansible --- Additional comment from John Fulton on 2019-05-15 20:00:07 IST --- This playbook (RFE) should be able to pre-cache the image at the site both for the case where a Glance API worker is there, as well as the case where there are only compute nodes, where the image just gets written to disk in the proper place so that Nova thinks it has already cached the image. It might be useful to have tags in the playbook for glance vs nova to support both of the options above.
This has been given conditional approval for 16.1.2 if the patch is landed by Sept 08. If not, we will automatically remove this.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Red Hat OpenStack Platform 16.1 bug fix and enhancement advisory), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2020:4284