Uploaded image for project: 'Red Hat OpenShift AI Engineering'
  1. Red Hat OpenShift AI Engineering
  2. RHOAIENG-6343

[2.9.z] DW components (CodeFlare, Ray, Kueue) aren't set as Managed in Cloud Service

XMLWordPrintable

    • False
    • Hide

      None

      Show
      None
    • False
    • Release Notes
    • No
    • Hide
      Cloud services only:
      Some components are set to `Removed` after installing OpenShift AI*
      After you install OpenShift AI, the `managementState` field for the `codeflare`, `kueue`, and `ray` components is incorrectly set to `Removed` instead of `Managed` in the `DataScienceCluster` custom resource.

      Workaround:
      1. In the OpenShift web console, log in as a cluster administrator.
      2. Click *Operators* → *Installed Operators* and then click the Red Hat OpenShift AI Operator.
      3. Click the *Data Science Cluster* tab.
      4. Select the *default-dsc* object.
      5. Click the *YAML* tab and in the `spec.components` section, set the value of the `managementState` field for `codeflare`, `kueue`, and `ray` to `Managed`.
      6. Click Save.
      Show
      Cloud services only: Some components are set to `Removed` after installing OpenShift AI* After you install OpenShift AI, the `managementState` field for the `codeflare`, `kueue`, and `ray` components is incorrectly set to `Removed` instead of `Managed` in the `DataScienceCluster` custom resource. Workaround: 1. In the OpenShift web console, log in as a cluster administrator. 2. Click *Operators* → *Installed Operators* and then click the Red Hat OpenShift AI Operator. 3. Click the *Data Science Cluster* tab. 4. Select the *default-dsc* object. 5. Click the *YAML* tab and in the `spec.components` section, set the value of the `managementState` field for `codeflare`, `kueue`, and `ray` to `Managed`. 6. Click Save.
    • Known Issue
    • Done
    • No
    • Testable

      With the RHOAI 2.9.0 RC1 build, the default DSC instance that is created for the Cloud Service installation type has the DW components management state set to Removed instead of the expected Managed state:

        olminstall git:(helper-channel) oc get dsc default-dsc -ojson | jq ".spec"                      
{
  "components": {
    "codeflare": {
      "managementState": "Removed"
    },
    "dashboard": {
      "managementState": "Managed"
    },
    "datasciencepipelines": {
      "managementState": "Managed"
    },
    "kserve": {
      "managementState": "Managed",
      "serving": {
        "ingressGateway": {
          "certificate": {
            "type": "SelfSigned"
          }
        },
        "managementState": "Managed",
        "name": "knative-serving"
      }
    },
    "kueue": {
      "managementState": "Removed"
    },
    "modelmeshserving": {
      "managementState": "Managed"
    },
    "ray": {
      "managementState": "Removed"
    },
    "trustyai": {
      "managementState": "Removed"
    },
    "workbenches": {
      "managementState": "Managed"
    }
  }
}

      Expectation is that components CodeFlare, Ray and Kueue are set as Managed in above.

      Note that the DSC template content is as expected.

      This issue seem to lie in this code.

        1. image-2024-05-16-17-44-36-478.png
          image-2024-05-16-17-44-36-478.png
          163 kB
        2. image-2024-04-29-15-32-46-716.png
          image-2024-04-29-15-32-46-716.png
          117 kB
        3. image-2024-04-29-09-44-22-047.png
          image-2024-04-29-09-44-22-047.png
          96 kB

            rh-ee-wenzhou Wen Zhou
            jstourac@redhat.com Jan Stourac
            Adrián Sanz Gómiz Adrián Sanz Gómiz
            RHOAI Platform
            Votes:
            0 Vote for this issue
            Watchers:
            12 Start watching this issue

              Created:
              Updated:
              Resolved: