Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-36870

NTO operand reloads TuneD unnecessarily twice

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done-Errata
    • Icon: Major Major
    • 4.15.z
    • 4.13.z, 4.12.z, 4.14.z, 4.15.z, 4.16
    • Node Tuning Operator
    • None
    • Important
    • No
    • False
    • Hide

      None

      Show
      None
    • Hide
      * Previously, the TuneD daemon could unnecessarily reload an additional time after a Tuned custom resource (CR) update. With this release, the Tuned object has been removed and the TuneD (daemon) profiles are carried directly in the Tuned Profile Kubernetes objects. As a result, the issue has been resolved. (link:https://issues.redhat.com/browse/OCPBUGS-32469[*OCPBUGS-32469*])
      Show
      * Previously, the TuneD daemon could unnecessarily reload an additional time after a Tuned custom resource (CR) update. With this release, the Tuned object has been removed and the TuneD (daemon) profiles are carried directly in the Tuned Profile Kubernetes objects. As a result, the issue has been resolved. (link: https://issues.redhat.com/browse/OCPBUGS-32469 [* OCPBUGS-32469 *])
    • Bug Fix
    • In Progress

      This is a clone of issue OCPBUGS-32469. The following is the description of the original issue:
      —
      Description of problem:

      
TuneD unnecessarily restarts twice when both current TuneD profile changes and when a new TuneD profile is selected.

      Version-Release number of selected component (if applicable):

      
All NTO versions are affected.

      How reproducible:

      
Depends on the order of k8s object updates (races), but nearly 100% reproducible.

      Steps to Reproduce:

          1. Install SNO 
    2. Label your SNO node with label "profile"
    3. Create the following CR:

apiVersion: tuned.openshift.io/v1
kind: Tuned
metadata:
  name: openshift-profile
  namespace: openshift-cluster-node-tuning-operator
spec:
  profile:
  - data: |
      [main]
      summary=Custom OpenShift profile 1
      include=openshift-node
      [sysctl]
      kernel.pty.max=4096
    name: openshift-profile-1
  - data: |
      [main]
      summary=Custom OpenShift profile 2
      include=openshift-node
      [sysctl]
      kernel.pty.max=8192
    name: openshift-profile-2
  recommend:
  - match:
    - label: profile
    priority: 20
    profile: openshift-profile-1

    4. Apply the following CR:

apiVersion: tuned.openshift.io/v1
kind: Tuned
metadata:
  name: openshift-profile
  namespace: openshift-cluster-node-tuning-operator
spec:
  profile:
  - data: |
      [main]
      summary=Custom OpenShift profile 1
      include=openshift-node
      [sysctl]
      kernel.pty.max=8192
    name: openshift-profile-1
  - data: |
      [main]
      summary=Custom OpenShift profile 2
      include=openshift-node
      [sysctl]
      kernel.pty.max=8192
    name: openshift-profile-2
  recommend:
  - match:
    - label: profile
    priority: 20
    profile: openshift-profile-2

      Actual results:

      
You'll see two restarts/applications of the openshift-profile-1

$ cat tuned-operand.log |grep "profile-1' applied"
2024-04-19 06:10:54,685 INFO     tuned.daemon.daemon: static tuning from profile 'openshift-profile-1' applied
2024-04-19 06:13:23,627 INFO     tuned.daemon.daemon: static tuning from profile 'openshift-profile-1' applied

      Expected results:

      
Only 1 application of openshift-profile-1:

$ cat tuned-operand.log |grep "profile-1' applied"
2024-04-19 07:20:31,600 INFO     tuned.daemon.daemon: static tuning from profile 'openshift-profile-1' applied

      Additional info:

              jmencak Jiri Mencak
              openshift-crt-jira-prow OpenShift Prow Bot
              Liquan Cui Liquan Cui
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Created:
                Updated:
                Resolved: