Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-54617

Spot Machine should be Failed if during provisioning Machine becomes Deallocated on Azure

XMLWordPrintable

    • Quality / Stability / Reliability
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • CLOUD Sprint 269, CLOUD Sprint 270
    • 2
    • Done
    • Bug Fix
    • Hide
      * Previously, {azure-short} spot machines that were evicted before their node became ready could get stuck in the `provisioned` state. With this release, {azure-short} spot instances now use a delete eviction policy. This policy ensures that the machines correctly transition to the `failed` state upon preemption. (link:https://issues.redhat.com/browse/OCPBUGS-54617[OCPBUGS-54617])
      Show
      * Previously, {azure-short} spot machines that were evicted before their node became ready could get stuck in the `provisioned` state. With this release, {azure-short} spot instances now use a delete eviction policy. This policy ensures that the machines correctly transition to the `failed` state upon preemption. (link: https://issues.redhat.com/browse/OCPBUGS-54617 [ OCPBUGS-54617 ])
    • None
    • None
    • None
    • None

      Description of problem:

       - When provisioning  Spot VM via Machineset ,if the cloud provider stops the VM between Machine gets provisioned and it joins to the cluster as Node , the machine remains powered-off (Deallocated in Azure clusters)  causing machine-controller get stuck in a loop waiting machine to join the node:
      [controller.go:318] machine-name-xxxx: has no node yet, requeuing

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

          

      How reproducible:

          Scale up a Spot

      Steps to Reproduce:

          1.
          2.
          3.
          

      Actual results:

      After machine joins as Node everithing works as expected and workloads are scheduled in surviving nodes.    

      Expected results:

      If the SpotVM gets Deallocated during provisionig phase, openshift should reconcile the status of the machine-controller.

      Additional info:

          

              rmanak@redhat.com Radek Manak
              rhn-support-lperezbe Luis Perez Besa
              None
              None
              Zhaohua Sun Zhaohua Sun
              None
              Votes:
              1 Vote for this issue
              Watchers:
              10 Start watching this issue

                Created:
                Updated:
                Resolved: