Back to bug 2116979

Who When What Removed Added
Artom Lifshitz 2022-08-09 16:35:18 UTC Target Milestone --- ga
Keywords TechPreview
Red Hat One Jira (issues.redhat.com) 2022-08-09 16:35:27 UTC Link ID Red Hat Issue Tracker OSP-18130
Artom Lifshitz 2022-08-09 17:20:27 UTC Assignee nova-maint rhos-docs
Keywords Triaged
QA Contact nova-maint rhos-docs
Component openstack-nova documentation
Summary [RHOS-17 GA Tech Preview] vGPU mdev instances are not being cleaned up after guest deletion [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up after guest deletion
Artom Lifshitz 2022-08-09 17:20:41 UTC Priority unspecified medium
Severity unspecified medium
James Smith 2022-08-09 18:38:48 UTC Assignee rhos-docs jamsmith
CC jamsmith
James Smith 2022-08-09 18:39:34 UTC Assignee jamsmith rhos-docs
Artom Lifshitz 2022-08-19 19:14:29 UTC Doc Type If docs needed, set a value Technology Preview
Artom Lifshitz 2022-08-22 19:54:42 UTC Comment 0 updated
Artom Lifshitz 2022-08-22 19:57:30 UTC Summary [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up after guest deletion [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up after guest deletion + multiple GPU types
Artom Lifshitz 2022-08-22 20:01:14 UTC Summary [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up after guest deletion + multiple GPU types [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up + multiple GPU types + vGPU cold migration
Artom Lifshitz 2022-08-22 20:01:36 UTC Comment 0 updated
Irina 2022-08-23 13:34:27 UTC Doc Text If you launch a vGPU instance in RHOSP 17 you cannot delete it, stop it, or move it. When an instance with a vGPU is deleted, migrated off its compute host, or stopped, the vGPU’s underlying mdev device is not cleaned up. If this happens to enough instances, all available mdev devices will be consumed, and no further instances with vGPUs can be created on that compute host.
CC igallagh
Doc Type Technology Preview Known Issue
Artom Lifshitz 2022-08-23 13:49:26 UTC Flags needinfo?(igallagh)
Artom Lifshitz 2022-08-23 15:59:18 UTC Doc Type Known Issue Technology Preview
Summary [17.0 ga Tech Preview] vGPU mdev instances are not being cleaned up + multiple GPU types + vGPU cold migration [17.0 ga Tech Preview] vGPU support
Artom Lifshitz 2022-08-23 16:57:28 UTC Summary [17.0 ga Tech Preview] vGPU support [17.0 ga tech preview] vGPU support
Roger Heslop 2022-09-09 17:08:24 UTC CC rheslop
Doc Text If you launch a vGPU instance in RHOSP 17 you cannot delete it, stop it, or move it. When an instance with a vGPU is deleted, migrated off its compute host, or stopped, the vGPU’s underlying mdev device is not cleaned up. If this happens to enough instances, all available mdev devices will be consumed, and no further instances with vGPUs can be created on that compute host. There is currently a known issue that prevents `mdev` devices from being cleaned up when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host.
Artom Lifshitz 2022-09-12 15:21:42 UTC Status NEW CLOSED
Keywords TechPreview
Resolution --- CURRENTRELEASE
Last Closed 2022-09-12 15:21:42 UTC
Irina 2022-09-22 14:59:12 UTC Flags needinfo?(igallagh)
James Smith 2022-09-22 18:34:23 UTC Doc Text There is currently a known issue that prevents `mdev` devices from being cleaned up when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host. A known issue prevents `mdev` devices from being cleaned up when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually, all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host.
Flags needinfo?(alifshit) needinfo?(igallagh) needinfo?(mgeary)
CC mgeary
Artom Lifshitz 2022-09-23 15:25:05 UTC Flags needinfo?(alifshit)
Doc Text A known issue prevents `mdev` devices from being cleaned up when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually, all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host. A known issue prevents `mdev` devices from being freed when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually, all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host.
Irina 2022-09-26 13:14:05 UTC Flags needinfo?(igallagh) needinfo?(jamsmith) needinfo?(astillma)
CC astillma
James Smith 2022-11-09 21:45:08 UTC Doc Text A known issue prevents `mdev` devices from being freed when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually, all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host. When using the Technology Preview vGPU support features, a known issue prevents `mdev` devices from being freed when stopping, moving, or deleting vGPU instances in RHOSP 17. Eventually, all `mdev` devices become consumed, and additional instances with vGPUs cannot be created on the compute host.
Doc Type Technology Preview Known Issue
Flags needinfo?(jamsmith)

Back to bug 2116979