Bug 1833440
| Summary: | NTO cannot remove kernel command line parameters shipped in parent profiles. | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Jiří Mencák <jmencak> |
| Component: | Node Tuning Operator | Assignee: | Jiří Mencák <jmencak> |
| Status: | CLOSED ERRATA | QA Contact: | Simon <skordas> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 4.5 | CC: | sejug, yquinn |
| Target Milestone: | --- | ||
| Target Release: | 4.5.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | No Doc Update | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-07-13 17:36:23 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
> Additional info: > This is a tuned bug which was fixed upstream: > https://github.com/redhat-performance/tuned/pull/265 I would consider this as fixed if all parent profiles contain a unique cmdline name as fixed in https://github.com/redhat-performance/tuned/pull/265. (Probably they are already aligned but this should be part of the criteria for VERIFIED) Retest positive: # oc get clusterversions.config.openshift.io NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.5.0-0.nightly-2020-05-14-021132 True False 145m Cluster version is 4.5.0-0.nightly-2020-05-14-021132 # node=$(oc get nodes --no-headers | grep worker | cut -d ' ' -f 1 | head -1) && echo $node skorda-z5h4h-w-a-xnlfc.c.openshift-qe.internal # pod=$(oc get pods -n openshift-cluster-node-tuning-operator -o wide --no-headers | grep $node | cut -d' ' -f1) && echo $pod tuned-4jcj4 # oc label node $node node-role.kubernetes.io/worker-rt= node/skorda-z5h4h-w-a-xnlfc.c.openshift-qe.internal labeled # oc get nodes NAME STATUS ROLES AGE VERSION skorda-z5h4h-m-0.c.openshift-qe.internal Ready master 5h56m v1.18.2 skorda-z5h4h-m-1.c.openshift-qe.internal Ready master 5h56m v1.18.2 skorda-z5h4h-m-2.c.openshift-qe.internal Ready master 5h56m v1.18.2 skorda-z5h4h-w-a-xnlfc.c.openshift-qe.internal Ready worker,worker-rt 5h44m v1.18.2 skorda-z5h4h-w-b-d5x2q.c.openshift-qe.internal Ready worker 5h45m v1.18.2 skorda-z5h4h-w-c-x2vkf.c.openshift-qe.internal Ready worker 5h44m v1.18.2 after creating tuned and machineconfigpool # oc logs $pod -n openshift-cluster-node-tuning-operator 2020-05-14 19:50:46,077 INFO tuned.daemon.daemon: static tuning from profile 'openshift-realtime' applied I0514 19:50:46.198499 2829 tuned.go:602] updated Profile skorda-z5h4h-w-a-xnlfc.c.openshift-qe.internal with bootcmdline: skew_tick=1 isolcpus=1 nosoftlockup tsc=nowatchdog 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, 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-2020:2409 |
Description of problem: When creating child profiles for NTO, it is impossible to remove kernel command line parameters defined in parent profiles. Version-Release number of selected component (if applicable): All How reproducible: Always Steps to Reproduce: $ oc label node "worker-node" node-role.kubernetes.io/worker-rt= $ oc create -f- <<EOF apiVersion: tuned.openshift.io/v1 kind: Tuned metadata: name: openshift-realtime namespace: openshift-cluster-node-tuning-operator spec: profile: - data: | [main] summary=Custom OpenShift realtime profile include=openshift-node,realtime [variables] # isolated_cores take a list of ranges; e.g. isolated_cores=2,4-7 isolated_cores=1 [bootloader] cmdline_openshift_realtime=-intel_pstate=disable name: openshift-realtime recommend: - machineConfigLabels: machineconfiguration.openshift.io/role: "worker-rt" priority: 30 profile: openshift-realtime EOF $ oc create -f- <<EOF apiVersion: machineconfiguration.openshift.io/v1 kind: MachineConfigPool metadata: name: worker-rt labels: worker-rt: "" spec: machineConfigSelector: matchExpressions: - {key: machineconfiguration.openshift.io/role, operator: In, values: [worker,worker-rt]} nodeSelector: matchLabels: node-role.kubernetes.io/worker-rt: "" EOF $ oc logs "NTO-operator-pod" I0508 15:46:28.543553 1 main.go:24] Go Version: go1.13.4 I0508 15:46:28.543643 1 main.go:25] Go OS/Arch: linux/amd64 I0508 15:46:28.543653 1 main.go:26] node-tuning Version: v4.5.0-202005061719-0-g32eea34-dirty I0508 15:46:28.546784 1 controller.go:780] trying to become a leader I0508 15:46:28.570352 1 controller.go:785] became a leader I0508 15:46:28.576367 1 controller.go:792] starting Tuned controller I0508 15:46:28.676653 1 controller.go:850] started events processor/controller I0508 15:47:39.803593 1 controller.go:406] updated Tuned rendered I0508 15:47:39.817353 1 controller.go:517] updated profile ip-10-0-137-57.eu-west-1.compute.internal [openshift-realtime] I0508 15:47:41.404917 1 controller.go:544] created MachineConfig 50-nto-worker-rt with kernelArguments: [skew_tick=1 isolcpus=1 intel_pstate=disable nosoftlockup tsc=nowatchdog] Actual results: intel_pstate=disable has not been removed. Expected results: intel_pstate=disable has been removed. Additional info: This is a tuned bug which was fixed upstream: https://github.com/redhat-performance/tuned/pull/265