Given the outstanding results of my recruiting job last week [1] I have been tasked with recruiting one of our glorious and most talented contributors to work on the fast-forward-upgrade script changes needed for the reshape-provider-tree blueprint. The work item is nicely detailed in the spec [2]. A few things to keep in mind: 1. There are currently no virt drivers which run the reshape routine. However, patches are up for review for libvirt [3] and xen [4]. There are also functional tests which exercise the ResourceTracker code with a faked out virt driver interface to test reshaping [5]. 2. The FFU entry point will mimic the reshape routine that will happen on nova-compute service startup in the ResourceTracker [6]. 3. The FFU script will need to run per-compute service rather than globally (or per cell) since it actually needs to call the virt driver's update_provider_tree() interface which might need to inspect the hardware (like for GPUs). Given there is already a model to follow from the ResourceTracker this should not be too hard, the work will likely mostly be writing tests. What do you get if you volunteer? The usual: fame, fortune, the respect of your peers, etc. [1] http://lists.openstack.org/pipermail/openstack-dev/2018-October/136075.html [2] https://specs.openstack.org/openstack/nova-specs/specs/stein/approved/reshape-provider-tree.html#offline-upgrade-script [3] https://review.openstack.org/#/c/599208/ [4] https://review.openstack.org/#/c/521041/ [5] https://github.com/openstack/nova/blob/a0eacbf7f/nova/tests/functional/test_servers.py#L1839 [6] https://github.com/openstack/nova/blob/a0eacbf7f/nova/compute/resource_tracker.py#L917-L940 -- Thanks, Matt