Bug 2000997
| Summary: | Change to additionalKernelArgs does not take effect on node | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Ian Miller <imiller> |
| Component: | Performance Addon Operator | Assignee: | Martin Sivák <msivak> |
| Status: | CLOSED DUPLICATE | QA Contact: | Gowrishankar Rajaiyan <grajaiya> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 4.8 | CC: | aos-bugs, jmencak, shajmakh |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-09-08 17:09:08 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: | |||
@Ian, can you double check whether the MachineConfig owned by your Performance Profile got updated? (In reply to Martin Sivák from comment #1) > @Ian, can you double check whether the MachineConfig owned by your > Performance Profile got updated? When I edit the PerformanceProfile to add/remove additionalKernelArgs I do not see any changes made to the in the MC "50-performance-perfprofile-policy" on the node. Looking at the contents of that MC the kernelArguments field is null. After more investigation. Edits to the PAO profile are immediately updating the tuned config file: oc exec -n openshift-cluster-node-tuning-operator tuned-z87df -- cat /etc/tuned/openshift-node-performance-perfprofile-policy/tuned.conf from: cmdline_additionalArg=+ idle=poll to: cmdline_additionalArg=+ idle=poll rcupdate.rcu_normal_after_boot=0 However this change does not take effect. Perhaps this is a duplicate of 1998247? As a workaround deleting the node-tuning operator's tuned-xxxxx pod causes the updated configuration to take effect. Hi Ian, (In reply to Ian Miller from comment #3) > However this change does not take effect. Perhaps this is a duplicate of > 1998247? Probably not. Can I have your cluster version (oc get clusterversion)? Thank you. oc get clusterversion NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.8.9 True False 6d9h Cluster version is 4.8.9 (In reply to Ian Miller from comment #6) > oc get clusterversion > NAME VERSION AVAILABLE PROGRESSING SINCE STATUS > version 4.8.9 True False 6d9h Cluster version is > 4.8.9 Thank you. Not sure why does this BZ say 4.9 then. 1998247 is not fixed in 4.8 yet, but I need to see if this is the same bug. A minimum reproducer (without PAO in the picture) would help. This is a duplicate of 1998247/1999608 Verified in Ian's environment. This is already fixed in 4.9. Fix for 4.8 depends on https://bugzilla.redhat.com/show_bug.cgi?id=1999608 i.e. https://github.com/openshift/cluster-node-tuning-operator/pull/268 merging. Thank you for providing me the access to your environment, Ian. Closing. *** This bug has been marked as a duplicate of bug 1999608 *** |
Description of problem: After the initial PerformanceProfile was applied to the system and all changes had taken effect, I modified the "additionalKernelArgs" setting to add a second argument. This value did not get applied to the system (waited 30 minutes). The initial performance profile: apiVersion: performance.openshift.io/v2 kind: PerformanceProfile metadata: name: perfprofile-policy spec: additionalKernelArgs: - idle=poll cpu: isolated: 2-39,42-79 reserved: 0-1,40-41 globallyDisableIrqLoadBalancing: true hugepages: defaultHugepagesSize: 1G pages: - count: 10 node: 0 size: 1G machineConfigPoolSelector: pools.operator.machineconfiguration.openshift.io/master: "" nodeSelector: node-role.kubernetes.io/master: "" numa: topologyPolicy: restricted realTimeKernel: enabled: false The updated version: apiVersion: performance.openshift.io/v2 kind: PerformanceProfile metadata: name: perfprofile-policy spec: additionalKernelArgs: - idle=poll - rcupdate.rcu_normal_after_boot=0 cpu: isolated: 2-39,42-79 reserved: 0-1,40-41 globallyDisableIrqLoadBalancing: true hugepages: defaultHugepagesSize: 1G pages: - count: 10 node: 0 size: 1G machineConfigPoolSelector: pools.operator.machineconfiguration.openshift.io/master: "" nodeSelector: node-role.kubernetes.io/master: "" numa: topologyPolicy: restricted realTimeKernel: enabled: false Version-Release number of selected component (if applicable): 4.9 How reproducible: Always Steps to Reproduce: 1. Apply initial PerformanceProfile 2. Wait for node to reconcile and all changes to take effect 3. Update additionalKernelArgs Actual results: Kernel command line: BOOT_IMAGE=(hd0,gpt3)/ostree/rhcos-4b83116ea516b3615355459ce879612839cfa52edea6851de9a01b6b9eb4eff4/vmlinuz-4.18.0-305.17.1.rt7.89.el8_4.x86_64 random.trust_cpu=on console=tty0 console=ttyS0,115200n8 ignition.platform.id=metal ostree=/ostree/boot.1/rhcos/4b83116ea516b3615355459ce879612839cfa52edea6851de9a01b6b9eb4eff4/0 ip=eno1:dhcp root=UUID=b91d7af0-b528-45fd-98f0-0a66bea1fd1d rw rootflags=prjquota skew_tick=1 nohz=on rcu_nocbs=2-39,42-79 tuned.non_isolcpus=00000300,00000003 intel_pstate=disable nosoftlockup tsc=nowatchdog intel_iommu=on iommu=pt isolcpus=managed_irq,2-39,42-79 systemd.cpu_affinity=0,1,40,41 default_hugepagesz=1G idle=poll nohz_full=2-39,42-79 Expected results: Kernel command line includes: rcupdate.rcu_normal_after_boot=0