Accessing `status.url` from ISVC Returns 503

[bartoszmajsak]

Summary

Steps to Reproduce

1. Deploy the latest incubation of odh-operator sources using manifests from here.
2. This deployment uses the image referenced here.
3. Change image to v0.12.0 to mitigate reconcile issue mentioned in Outdated images in opendatahub-operator breaks ISVC reconcile #233
4. Run the smoke test against KServe.
5. Observe 503

Workaround

Use .status.components.predictor.url instead. This seems to be working fine

ISVC

apiVersion: v1
items:
- apiVersion: serving.kserve.io/v1beta1
  kind: InferenceService
  metadata:
    annotations:
      kubectl.kubernetes.io/last-applied-configuration: |
        {"apiVersion":"serving.kserve.io/v1beta1","kind":"InferenceService","metadata":{"annotations":{"serving.knative.openshift.io/enablePassthrough":"true","sidecar.istio.io/inject":"true"},"name":"sklearn-v2-iris","namespace":"kserve-test"},"spec":{"predictor":{"model":{"modelFormat":{"name":"sklearn"},"protocolVersion":"v2","runtime":"kserve-sklearnserver","storageUri":"gs://kfserving-examples/models/sklearn/1.0/model"}}}}
      security.opendatahub.io/enable-auth: "true"
      serving.knative.openshift.io/enablePassthrough: "true"
      sidecar.istio.io/inject: "true"
    creationTimestamp: "2024-07-09T15:50:50Z"
    finalizers:
    - inferenceservice.finalizers
    generation: 1
    name: sklearn-v2-iris
    namespace: kserve-test
    resourceVersion: "155594"
    uid: 9e242bdf-ee5c-4522-9bd9-542dd0c84cb7
  spec:
    predictor:
      model:
        modelFormat:
          name: sklearn
        name: ""
        protocolVersion: v2
        resources: {}
        runtime: kserve-sklearnserver
        storageUri: gs://kfserving-examples/models/sklearn/1.0/model
  status:
    address:
      url: http://sklearn-v2-iris.kserve-test.svc.cluster.local
    components:
      predictor:
        address:
          url: http://sklearn-v2-iris-predictor.kserve-test.svc.cluster.local
        latestCreatedRevision: sklearn-v2-iris-predictor-00002
        latestReadyRevision: sklearn-v2-iris-predictor-00002
        latestRolledoutRevision: sklearn-v2-iris-predictor-00002
        previousRolledoutRevision: sklearn-v2-iris-predictor-00001
        traffic:
        - latestRevision: true
          percent: 100
          revisionName: sklearn-v2-iris-predictor-00002
        url: https://sklearn-v2-iris-predictor-kserve-test.apps.rosa.q2rzk-kbxfh-3tj.3r5r.p3.openshiftapps.com
    conditions:
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      status: "True"
      type: IngressReady
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      severity: Info
      status: "True"
      type: LatestDeploymentReady
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      severity: Info
      status: "True"
      type: PredictorConfigurationReady
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      status: "True"
      type: PredictorReady
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      severity: Info
      status: "True"
      type: PredictorRouteReady
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      status: "True"
      type: Ready
    - lastTransitionTime: "2024-07-09T16:00:38Z"
      severity: Info
      status: "True"
      type: RoutesReady
    modelStatus:
      copies:
        failedCopies: 0
        totalCopies: 1
      states:
        activeModelState: Loaded
        targetModelState: Loaded
      transitionStatus: UpToDate
    observedGeneration: 2
    url: https://sklearn-v2-iris-kserve-test.apps.rosa.q2rzk-kbxfh-3tj.3r5r.p3.openshiftapps.com
kind: List
metadata:
  resourceVersion: ""

[israel-hdez]

Run the smoke test against KServe.

Do you still remember which smoke test you ran?