Uploaded image for project: 'Red Hat OpenShift Data Science'
  1. Red Hat OpenShift Data Science
  2. RHODS-6590

[Model Serving] Model status will never update if the runtime pod is unschedulable

    XMLWordPrintable

Details

    • False
    • None
    • False
    • Testable
    • No
    • No
    • No
    • Pending
    • None

    Description

      Description of problem:

      When deploying a model, if the run time pod cannot be schedule because there are not enough resources, its status will never update from the loading animation.
      This is because the pod remains in "Pending" state until it can be scheduled.

      Prerequisites (if any, like setup, operators/versions):

      Steps to Reproduce

      1. Deploy a model via model serving using more resources than currently available
      2. Wait for its status in the list of deployed models to become either pass or fail

      Actual results:

      The status never updates, as it is likely using the Pod status (Pending) to determine what to show

      Expected results:

      After some time the user should be informed that the pod is currently unschedulable and that the model was not successfully deployed

      Reproducibility (Always/Intermittent/Only Once):

      always

      Build Details:

      rhods 1.21

      Workaround:

      Additional info:

      Attachments

        Activity

          People

            Unassigned Unassigned
            rhn-support-lgiorgi Luca Giorgi
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: