Details
-
Bug
-
Resolution: Done
-
Undefined
-
None
-
4.13
-
No
-
OCPNODE Sprint 235 (Blue)
-
1
-
False
-
Description
Description of problem:
After running node-density-heavy workload and cleaning up after the workload runs, kubelet CPU does not settle down to idle after pods have been deleted.
Version-Release number of selected component (if applicable):
4.13.0-0.nightly-2023-04-09-095122
How reproducible:
Saw it at 1050 density. Will report back when 2525 run completes.
Steps to Reproduce:
1. Install 4.13 cluster using crun 2. kube-burner ocp node-density-heavy --gc=true 1050
Actual results:
worker kubelet CPU increases as workload cleanup completes and stays pegged at high rate indefinitely.
Expected results:
worker kubelet CPU settles down after pods are deleted and resources are cleaned up.
Additional info:
