Uploaded image for project: 'Open Data Hub'
  1. Open Data Hub
  2. ODH-506

JupyterHub Memory Calculation Function fails when node memory doesn't have a unit

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Major Major
    • None
    • None
    • None
    • Open Data Hub Sprint 12, Open Data Hub Sprint 13

        allocatable:
          cpu: "94"
          ephemeral-storage: "1519561896383"
          hugepages-1Gi: "0"
          hugepages-2Mi: "0"
          memory: "381935474387"
          pods: "250" 

      JupyterHub fails when attempting to spawn a new notebook if one of the nodes does not have a unit attached to the memory. In the above code snippet, you can see "memory" is just an int.
      This function fails: https://github.com/red-hat-data-services/jupyterhub-singleuser-profiles/blob/master/jupyterhub_singleuser_profiles/openshift.py#L150
      12:34
      Specifically, line 157 - memory isn’t set
      The JSP code needs to be updated to have a case for when there isn't a unit
      cc maros.roman

              acorvin@redhat.com Alex Corvin
              aasthana@redhat.com Anish Asthana
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Created:
                Updated:
                Resolved: