Description of problem: An issue in the lshw package can cause memory reporting from inspection to fail. A full description is available in the original fix to work around this issue, located at https://review.opendev.org/c/openstack/ironic-python-agent/+/788588. Version-Release number of selected component (if applicable): The issue in lshw is present in lshw = B.02.19.2-5.el8 and ultimately code to handle this issue in ironic-python-agent needs to be backported to a Train based release to handle the lshw package issue. How reproducible: Unknown. This was apparently found upstream by the OpenShift IPI Metal team, the fix was just not backported to the stable/train branch. Running a stable/train branch version on RHEL 8.4 exposes OSP to this potential failure case, and the most appropriate action is to backport the upstream fix into our builds.
I actually have this issue 100% of the time in my lab when building OSP16.2 (including 16.2.1 that was released late yesterday). Although we have a workaround [1], I would like to request that this fix be backported to OSP16.2.1. [1] https://access.redhat.com/solutions/6568411
The fix is on track for being released in 16.2.2.
@sbaker Thanks. I am aware of that. I am asking that it be backported to OSP16.2.1.
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 (Release of components for Red Hat OpenStack Platform 16.2.2), 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-2022:1001