-
Bug
-
Resolution: Unresolved
-
Undefined
-
rhel-9.4.z
-
tuned-2.25.0-0.1.rc1.el9
-
No
-
Important
-
Patch, Upstream
-
2
-
rhel-net-perf
-
ssg_core_services
-
5
-
False
-
False
-
-
None
-
_N&P-Refined_, N&P-25_7
-
None
-
None
-
None
What were you trying to do that didn't work?
Setting the following tuneD configuration:
[net] channels=combined 4
might result in a degraded tuneD profile, as it trying to collect info with ethtool on a device that doesn't exist.
We believe, there is a race condition between the system service responsible for the net devices renaming and the tuned pod which tries to grab the device channels (ethtool -l device). The probability of the bug reproducing increases when we create more VFs.
This has been reproduced on an SNO with DU configuration
How reproducible is this bug?:
Around 50%, on an SNO with DU profile
For more details info PTAL at:
*https://issues.redhat.com/browse/OCPBUGS-41934*
- causes
-
OCPBUGS-41934 tuned profile got degraded after node reboot
-
- POST
-
- is related to
-
RHEL-88238 Please add option which will wait for udev settle during startup
-
- Release Pending
-
- links to
-
RHSA-2025:144994 tuned update