This error occurs in U/S CI (CNF tests), and causes the tuned profile to be degraded.
It has been diagnosed that this bug is occurring on a specific hardware: R750 with Intel E810 network device.

When UserLevelNetworking is set to True, tuned attempts to et the combined channel count equal to the reservedCPUs but fails with the following error:
tuned.utils.commands: Executing'ethtool -L ens2f0 combined 1'error: netlink error: Device or resource busy
The error occurs because the ice driver: ens2f0: Cannot change channels when RDMA is active.

This affects all OCP versions down to 4.16

https://testgrid.k8s.io/redhat-openshift-ocp-release-4.19-informing#periodic-ci-openshift-release-master-nightly-4.19-e2e-telco5g-sno-cnftests&width=90&include-filter-by-regex=.It.%20.rfe_id:27368..performance.%20Tuned%20CRs%20generated%20from%20profile%20Tuned%20profile%20shouldn.t%20be%20degraded%20.tier-0.&embed=

As a temporary solution, we will block RDMA to avoid these errors.
This bug has been opened to merge a workaround that will be used.