stx-openstack apply failed after enabling cpu_dedicated_set
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
StarlingX |
New
|
Undecided
|
Thales Elero Cervi |
Bug Description
STX release 07 Dedicated Storage configured with multiple worker nodes.
One of the worker node with couple of VMs choosed to enable cpu_dedicated_set via the controller-0.
1. Guests/VMs were shutdown then lock the selected worker node.
2. Enabled cpu_dedicate_set on a worker node as per the documentation [1].
3. Unlock the worker
Then noticed that stx-openstack apply process was failed.
Here is the armada log: https:/
There were couple of pods were failing (in the same worker node).
NAME READY STATUS RESTARTS AGE
nova-compute-
pci-irq-
Deleted both pods and check if its recovering but no luck.
Collect Logs available at [2].
pci-irq-
Error from server (BadRequest): container "pci-irq-
[2] https:/
tags: | added: stx.7.0 stx.distro.openstack |
Changed in starlingx: | |
assignee: | nobody → Thales Elero Cervi (tcervi) |
further, on the same worker node Nvidia A40 card has been virtualized (using SR-IOV) and attached to both guest were running.
Both guests were working fine.
CPU_dedicated_set enabled after vGPU creation.