Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-11118

The shutdownGracePeriod configuration are not set by default

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Normal Normal
    • None
    • 4.13
    • Networking / SR-IOV
    • None
    • Quality / Stability / Reliability
    • False
    • Hide

      None

      Show
      None
    • None
    • Moderate
    • No
    • None
    • None
    • None
    • None
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      Description of problem:

      By default, shutdownGracePeriod and shutdownGracePeriodCriticalPods are set to 0. (https://56770--docspreview.netlify.app/openshift-enterprise/latest/nodes/nodes/nodes-nodes-graceful-shutdown.html#nodes-nodes-grace-periods-graceful-shutdown_nodes-nodes-graceful-shutdown) 
      
      But these configuration doesn't exist in /etc/kubernetes/kubelet.conf by default.

       

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

      % oc get clusterversionNAME   VERSION                           AVAILABLE   PROGRESSING   SINCE   STATUSversion   4.13.0-0.nightly-2023-03-28-014156   True     False      98m  Cluster version is 4.13.0-0.nightly-2023-03-28-014156

      How reproducible:

      always

      Steps to Reproduce:

      1.oc debug node and check if shutdownGracePeriod and shutdownGracePeriodCriticalPods are included in the file /etc/kubernetes/kubelet.conf
      2.
      3.
      

      Actual results:

      1 the shutdownGracePeriod and shutdownGracePeriodCriticalPods are not included in /etc/kubernetes/kubelet.conf
      
      oc debug node/worker-0 
      ...
      sh-4.4# chroot /host   
      sh-5.1# cat /etc/kubernetes/kubelet.conf 
      kind: KubeletConfiguration
      apiVersion: kubelet.config.k8s.io/v1beta1
      authentication:
        x509:
          clientCAFile: /etc/kubernetes/kubelet-ca.crt
        anonymous:
          enabled: false
      cgroupDriver: systemd
      cgroupRoot: /
      clusterDNS:
        - 172.30.0.10
      clusterDomain: cluster.local
      containerLogMaxSize: 50Mi
      maxPods: 250
      kubeAPIQPS: 50
      kubeAPIBurst: 100
      podPidsLimit: 4096
      protectKernelDefaults: true
      rotateCertificates: true
      serializeImagePulls: false
      staticPodPath: /etc/kubernetes/manifests
      systemCgroups: /system.slice
      featureGates:
        APIPriorityAndFairness: true
        RotateKubeletServerCertificate: true
        DownwardAPIHugePages: true
        OpenShiftPodSecurityAdmission: true
        RetroactiveDefaultStorageClass: false
        RotateKubeletServerCertificate: true
      serverTLSBootstrap: true
      tlsMinVersion: VersionTLS12
      tlsCipherSuites:
        - TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256
        - TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256
        - TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384
        - TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384
        - TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256
        - TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305_SHA256
      sh-5.1# 
      

      Expected results:

      1 the shutdownGracePeriod and shutdownGracePeriodCriticalPods are included in /etc/kubernetes/kubelet.conf, and set to 0 by default.

      Additional info:

       

              bnemeth@redhat.com Balazs Nemeth
              rhn-support-minmli Min Li
              None
              None
              Sunil Choudhary Sunil Choudhary
              None
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

                Created:
                Updated:
                Resolved: