chore(chart-deps): update kserve to version v0.20.0

[svcAPLBot]

This PR updates the dependency kserve-resources to version v0.20.0.

[Copilot]

Pull request overview

This PR bumps the vendored/managed KServe Helm artifacts to align with kserve-resources v0.20.0, updating version references across the chart metadata, dependency pinning, and associated documentation/config defaults.

Changes:

• Update KServe chart/dependency version references from v0.19.0 → v0.20.0.
• Update README badges/examples and apps catalog metadata to reflect 0.20.0.
• Extend the KServe controller ConfigMap defaults with new storage initializer OCI settings (enableOciModelSupport, ociModelMode).

Reviewed changes

Copilot reviewed 6 out of 7 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
charts/kserve/values.yaml	Bumps default kserve.version to v0.20.0.
charts/kserve/README.md	Updates version badges and install example to v0.20.0.
charts/kserve/files/common/configmap.yaml	Adds v0.20.0 storage initializer OCI-related settings.
charts/kserve/Chart.yaml	Updates chart version and appVersion to v0.20.0.
charts/dependencies.yaml	Pins kserve-crd and kserve-resources to v0.20.0.
apps.yaml	Updates KServe app catalog entry version to 0.20.0.

Suppressed comments (1)

charts/kserve/files/common/configmap.yaml:161

• data.storageInitializer is defined twice in this ConfigMap (the second definition starting at line 161 overrides the earlier one). Duplicate mapping keys are invalid YAML and can break tooling; it also makes it unclear which initializer config is intended to be used.

           "ociModelMode": "modelcar",
           "cpuModelcar": "10m",
           "memoryModelcar": "15Mi"
       }
     storageInitializer: |-

[Copilot]

Pull request overview

Copilot reviewed 6 out of 7 changed files in this pull request and generated no new comments.

Suppressed comments (1)

charts/dependencies.yaml:67

• PR description says it updates only the kserve-resources dependency, but this change also bumps kserve-crd to v0.20.0. Please update the PR description (or split the change) so reviewers/users understand both dependencies are being upgraded together.

  - name: kserve-crd
    version: v0.20.0
    repository: oci://ghcr.io/kserve/charts/kserve-crd

[svcAPLBot]

Comparison of Helm chart templating output:

# kserve-resources/templates/common/clusterstoragecontainer.yaml

@@ spec.container.image @@
! ± value change
- kserve/storage-initializer:v0.19.0
+ kserve/storage-initializer:v0.20.0

# kserve-resources/templates/common/configmap.yaml

@@ data._example @@
! ± value change in multiline text (two inserts, no deletions)
  ################################
  #                              #
  #    EXAMPLE CONFIGURATION     #
  #                              #
  
  [144 lines unchanged)]
  
         "cpuLimit": "1",
         "caBundleConfigMapName": "",
         "caBundleVolumeMountPath": "/etc/ssl/custom-certs",
         "enableModelcar": false,
+        "enableOciModelSupport": false,
+        "ociModelMode": "modelcar",
         "cpuModelcar": "10m",
         "memoryModelcar": "15Mi"
     }
   storageInitializer: |-
  
  [21 lines unchanged)]
  
  
         # enableModelcar enabled allows you to directly access an OCI container image by
         # using a source URL with an "oci://" schema.
         "enableModelcar": false,
+ 
+        # enableOciModelSupport enables any OCI-backed model storage path (modelcar, native ImageVolume, or fetch).
+        # This is the newer master switch; enableModelcar is kept as a backcompat alias for the "modelcar" mode.
+        "enableOciModelSupport": false,
+ 
+        # ociModelMode selects the materialization strategy when a storageUri uses oci:// without an explicit
+        # suffix. Valid values: "modelcar" (default sidecar), "native" (K8s ImageVolume), "fetch" (init-container).
+        "ociModelMode": "modelcar",
  
         # cpuModelcar is the cpu request and limit that is used for the passive modelcar container. It can be
         # set very low, but should be allowed by any Kubernetes LimitRange that might apply.
         "cpuModelcar": "10m",
  
  [434 lines unchanged)]
  
       "reconcilationFrequencyInSecs": 60,
       # This is to disable localmodel pv and pvc management for namespaces without isvcs
       "disableVolumeManagement": false
     }

@@ data.agent @@
! ± value change in multiline text (one insert, one deletion)
  {
    "cpuLimit": "1",
    "cpuRequest": "100m",
-   "image": "kserve/agent:v0.19.0",
+   "image": "kserve/agent:v0.20.0",
    "memoryLimit": "1Gi",
    "memoryRequest": "100Mi"
  }

@@ data.batcher @@
! ± value change in multiline text (one insert, one deletion)
  {
    "cpuLimit": "1",
    "cpuRequest": "1",
-   "image": "kserve/agent:v0.19.0",
+   "image": "kserve/agent:v0.20.0",
    "maxBatchSize": "32",
    "maxLatency": "5000",
    "memoryLimit": "1Gi",
    "memoryRequest": "1Gi"
  }

@@ data.explainers @@
! ± value change in multiline text (one insert, one deletion)
  {
    "art": {
-     "defaultImageVersion": "v0.19.0",
+     "defaultImageVersion": "v0.20.0",
      "image": "kserve/art-explainer"
    }
  }

@@ data.localModel @@
! ± value change in multiline text (one insert, one deletion)
  {
-   "defaultJobImage": "kserve/storage-initializer:v0.19.0",
+   "defaultJobImage": "kserve/storage-initializer:v0.20.0",
    "disableVolumeManagement": false,
    "enabled": false,
    "fsGroup": 1000,
    "jobNamespace": "kserve-localmodel-jobs",
    "jobTTLSecondsAfterFinished": 3600,
    "reconcilationFrequencyInSecs": 60
  }

@@ data.logger @@
! ± value change in multiline text (one insert, one deletion)
  {
    "cpuLimit": "1",
    "cpuRequest": "100m",
    "defaultUrl": "http://default-broker",
-   "image": "kserve/agent:v0.19.0",
+   "image": "kserve/agent:v0.20.0",
    "memoryLimit": "1Gi",
    "memoryRequest": "100Mi"
  }

@@ data.router @@
! ± value change in multiline text (one insert, one deletion)
  {
    "cpuLimit": "1",
    "cpuRequest": "100m",
-   "image": "kserve/router:v0.19.0",
+   "image": "kserve/router:v0.20.0",
    "imagePullPolicy": "IfNotPresent",
    "imagePullSecrets": [],
    "memoryLimit": "1Gi",
    "memoryRequest": "100Mi"
  }

@@ data.storageInitializer @@
! ± value change in multiline text (one insert, one deletion)
  {
    "caBundleConfigMapName": "",
    "caBundleVolumeMountPath": "/etc/ssl/custom-certs",
    "cpuLimit": "1",
    "cpuModelcar": "10m",
    "cpuRequest": "100m",
    "enableModelcar": true,
-   "image": "kserve/storage-initializer:v0.19.0",
+   "image": "kserve/storage-initializer:v0.20.0",
    "memoryLimit": "1Gi",
    "memoryModelcar": "15Mi",
    "memoryRequest": "100Mi",
    "uidModelcar": 1010
  }

# kserve-resources/templates/kserve/resources.yaml

@@ spec.template.spec.containers.manager.image @@
# apps/v1/Deployment/kserve/kserve-controller-manager
! ± value change
- kserve/kserve-controller:v0.19.0
+ kserve/kserve-controller:v0.20.0

# otomi-api/templates/core-config.yaml

@@ data.core.yaml @@
! ± value change in multiline text (one insert, one deletion)
  adminApps:
  - deps:
    - prometheus
    name: alertmanager
  
  [327 lines unchanged)]
  
      title: Knative Operator
    kserve:
      about: Standardized Distributed Generative and Predictive AI Inference Platform
        for Scalable, Multi-Framework Deployment on Kubernetes.
-     appVersion: 0.19.0
+     appVersion: 0.20.0
      chartName: kserve
      isAlpha: true
      license: Apache 2.0
      maintainers: Kserve
  
  [258 lines unchanged)]
  
    path: /explore?orgId=1&left=%7B"datasource":"loki","queries":%5B%7B"refId":"A","expr":"","queryType":"range","datasource":%7B"type":"loki","uid":"loki"%7D%7D%5D,"range":%7B"from":"now-1h","to":"now"%7D%7D
    useHost: grafana
  - name: tekton
    ownHost: true

# otomi-api/templates/deployment.yaml

# rabbitmq-cluster-operator/templates/messaging-topology-operator/validating-webhook-configuration.yaml

# values-repo.yaml