+++ This bug was initially created as a clone of Bug #2013321 +++ Description of problem: The fix for rhbz#1979352 introduced the [scheduler] plug-in as a standard part of openshift TuneD profiles. Unfortunately, the [scheduler] plug-in can be very CPU intensive, especially on the OpenShift platform. The bug for this issue is tracked by rhbz#1921738. The CPU utilization of the TuneD process can be around 1% of one core. Version-Release number of selected component (if applicable): 4.8->4.10 How reproducible: Always. Steps to Reproduce: 1. Install OCP 2. Watch tuned process utilization either via top -p <pid> or just by querying /proc/<pid>/status Actual results: ~1% of CPU Expected results: ~0% of CPU Additional info: https://github.com/openshift/cluster-node-tuning-operator/pull/278
$ oc get clusterversions.config.openshift.io NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.9.0-0.nightly-2021-10-14-063636 True False 103m Cluster version is 4.9.0-0.nightly-2021-10-14-063636 $ oc get nodes NAME STATUS ROLES AGE VERSION skordas1013a-5jlfb-master-0.c.openshift-qe.internal Ready master 122m v1.22.0-rc.0+894a78b skordas1013a-5jlfb-master-1.c.openshift-qe.internal Ready master 122m v1.22.0-rc.0+894a78b skordas1013a-5jlfb-master-2.c.openshift-qe.internal Ready master 122m v1.22.0-rc.0+894a78b skordas1013a-5jlfb-worker-a-llhdj.c.openshift-qe.internal Ready worker 111m v1.22.0-rc.0+894a78b skordas1013a-5jlfb-worker-b-rzn2b.c.openshift-qe.internal Ready worker 111m v1.22.0-rc.0+894a78b skordas1013a-5jlfb-worker-c-l4knk.c.openshift-qe.internal Ready worker 112m v1.22.0-rc.0+894a78b # master node $ oc debug node/skordas1013a-5jlfb-master-0.c.openshift-qe.internal Starting pod/skordas1013a-5jlfb-master-0copenshift-qeinternal-debug ... To use host binaries, run `chroot /host` Pod IP: 10.0.0.4 If you don't see a command prompt, try pressing enter. sh-4.4# chroot /host sh-4.4# pidof openshift-tuned 16892 sh-4.4# top -p 16892 PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 16892 root 20 0 1495812 45028 26756 S 0.0 0.3 0:00.93 openshift-tuned # worker node $ oc debug node/skordas1013a-5jlfb-worker-a-llhdj.c.openshift-qe.internal Starting pod/skordas1013a-5jlfb-worker-a-llhdjcopenshift-qeinternal-debug ... To use host binaries, run `chroot /host` Pod IP: 10.0.128.2 If you don't see a command prompt, try pressing enter. sh-4.4# chroot /host sh-4.4# pidof openshift-tuned 2504 sh-4.4# top -p 2504 PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 2504 root 20 0 1569800 44908 25452 S 0.0 0.3 0:00.70 openshift-tuned # %CUP = 0
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (OpenShift Container Platform 4.9.4 bug fix update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2021:3935