Description of problem:
CU is facing the upstream bug https://bugs.launchpad.net/nova/+bug/1960230 and he is asking backport for OSP 16.2 of the following patch:
https://opendev.org/openstack/nova/commit/9111b99f739d41c092db8d01712a5aa72388b5fb
after each upgrade they are running playbook to reboot all of the nodes including cold migration of vgpu instances, however it's possible to migrate instance to different compute node only once due to bug mentioned above.
Version-Release number of selected component (if applicable):
OSP 16.2
Upstream stable/wallaby fix was already merged, so RHOSP 17.1 will get it automatically. During triage call we have agreed that we are unlikely to deliver fix in foreseeable future by merging it in upstream stable/train first, so downstream-only cherry-pick will be used.
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 16.2.6 (Train) 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/RHBA-2023:6307