Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-3443

[4.12] Descheduler pod is OOM killed when using descheduler-operator profiles on big clusters

    XMLWordPrintable

Details

    • Bug
    • Resolution: Done
    • Normal
    • None
    • 4.10
    • kube-scheduler
    • None
    • Moderate
    • False
    • Hide

      None

      Show
      None
    • Hide
      The {product-title} {product-version} release changes the behavior of the descheduler during initialization, so that the descheduler no longer consumes memory to the point where it could lead to out-of-memory (OOM) issues.

      In previous releases of {product-title}, if you deployed the Kube Descheduler Operator on an {product-title} cluster, and you changed the default resource limit values for a deployed pod, the pod's values would revert to the default values. This caused the descheduler to incorrectly record events when it evicted the pod from a node.

      (link:https://issues.redhat.com/browse/OCPBUGS-3443)[*OCPBUGS-3443*]
      Show
      The {product-title} {product-version} release changes the behavior of the descheduler during initialization, so that the descheduler no longer consumes memory to the point where it could lead to out-of-memory (OOM) issues. In previous releases of {product-title}, if you deployed the Kube Descheduler Operator on an {product-title} cluster, and you changed the default resource limit values for a deployed pod, the pod's values would revert to the default values. This caused the descheduler to incorrectly record events when it evicted the pod from a node. (link: https://issues.redhat.com/browse/OCPBUGS-3443) [* OCPBUGS-3443 *]
    • Bug Fix
    • Done

    Description

      Description of problem:

      When deploying the Kube Descheduler Operator on a OCP Cluster
is not possible to modify the resources(request/limits) of the deployed pods.

This is problematic in big environments as it eventually hits an OOM scenario where the pod is restarting.

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

      4.10

      How reproducible:

      Always

      Steps to Reproduce:

      1. Deploy operator
2. Try to modify the resource values of the pod/deployment
3. Operator defaults the values back.

      Actual results:

      Not possible to change resources in the pod

      Expected results:

      Pod running with different resource values.

      Additional info:

       

      Attachments

        Issue Links

          clones

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-1995 Descheduler pod is OOM killed when using descheduler-operator profiles on big clusters

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed
          depends on

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-3444 [4.13] Descheduler pod is OOM killed when using descheduler-operator profiles on big clusters

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed
          is cloned by

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-3444 [4.13] Descheduler pod is OOM killed when using descheduler-operator profiles on big clusters

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed
          is depended on by

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-1995 Descheduler pod is OOM killed when using descheduler-operator profiles on big clusters

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed

          Activity

            People

              jchaloup@redhat.com Jan Chaloupka
              rhn-support-mblach Miguel Blach
              Rama Kasturi Narra Rama Kasturi Narra
              Darragh Fitzmaurice Darragh Fitzmaurice
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: