RHOAIENG-34472: [WIP] GIE v1 Migration

.github/workflows/e2e-test-llmisvc.yml

@@ -99,7 +99,8 @@ jobs:
         run: |
           # Run only CPU tests for now using pytest markers (cluster_)
           # Available GPU vendors: amd, nvidia, intel
-          ./test/scripts/gh-actions/run-e2e-tests.sh "llminferenceservice and cluster_cpu" 2 "istio-gatewayapi-ext"
+          # Use parallelism=1 to eliminate HTTP 409 cleanup race conditions
+          ./test/scripts/gh-actions/run-e2e-tests.sh "llminferenceservice and cluster_cpu" 1 "istio-gatewayapi-ext"
 
       - name: Check system status
         if: always()

Dockerfile

@@ -6,6 +6,8 @@ WORKDIR /go/src/github.com/kserve/kserve
 COPY go.mod  go.mod
 COPY go.sum  go.sum
 
+# Allow Go to automatically download the required toolchain version
+ENV GOTOOLCHAIN=auto
 RUN go mod download
 
 COPY cmd/    cmd/
@@ -17,6 +19,11 @@ RUN CGO_ENABLED=0 GOOS=linux GOFLAGS=-mod=mod go build -a -o manager ./cmd/manag
 
 # Generate third-party licenses
 COPY LICENSE LICENSE
+# go-licenses is temporarily disabled due to incompatibility with Go 1.24+
+# go-licenses v1.6.0 has a compilation error with Go 1.24 due to github.com/otiai10/copy dependency
+# go-licenses v2 is available as pre-release but not yet stable
+# See: https://github.com/google/go-licenses/issues/312
+# Uncomment these lines once go-licenses v2 is stable or v1.x adds Go 1.24+ support
 # RUN go install github.com/google/go-licenses@latest
 # Forbidden Licenses: https://github.com/google/licenseclassifier/blob/e6a9bb99b5a6f71d5a34336b8245e305f5430f99/license_type.go#L341
 # RUN go-licenses check ./cmd/... ./pkg/... --disallowed_types="forbidden,unknown"

Makefile

@@ -99,24 +99,40 @@ manifests: controller-gen yq
 	echo '{{- end }}' >> charts/kserve-resources/templates/localmodelnode/role.yaml
 
 	@$(CONTROLLER_GEN) object:headerFile="hack/boilerplate.go.txt" paths=./pkg/apis/serving/v1alpha1
+	@$(CONTROLLER_GEN) object:headerFile="hack/boilerplate.go.txt" paths=./pkg/apis/serving/v1alpha2
 	@$(CONTROLLER_GEN) object:headerFile="hack/boilerplate.go.txt" paths=./pkg/apis/serving/v1beta1
 
 	# Remove validation for the LLMInferenceServiceConfig API so that we can use Go templates to inject values at runtime.
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.rules.items.properties.matches.items.properties.path.x-kubernetes-validations)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.rules.items.properties.matches.items.properties.path.x-kubernetes-validations)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.rules.items.properties.filters.items.properties.urlRewrite.properties.path.x-kubernetes-validations)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.rules.items.properties.filters.items.properties.urlRewrite.properties.path.x-kubernetes-validations)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.parentRefs.items.properties.namespace.pattern)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.route.properties.http.properties.spec.properties.parentRefs.items.properties.namespace.pattern)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	# Remove validation for the LLMInferenceServiceConfig API so that we can override only specific values.
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.pool.properties.spec.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.pool.properties.spec.required)' -i config/crd/full/serving.kserve.io_llminferenceserviceconfigs.yaml
 	# Remove validation for the LLMInferenceService API so that we can override only specific values.
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.prefill.properties.worker.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
+	@$(YQ) 'del(.spec.versions[1].schema.openAPIV3Schema.properties.spec.properties.router.properties.scheduler.properties.template.required)' -i config/crd/full/serving.kserve.io_llminferenceservices.yaml
 
 	#remove the required property on framework as name field needs to be optional
 	@$(YQ) 'del(.spec.versions[0].schema.openAPIV3Schema.properties.spec.properties.*.properties.*.required)' -i config/crd/full/serving.kserve.io_inferenceservices.yaml
@@ -145,7 +161,7 @@ manifests: controller-gen yq
 	cp config/crd/minimal/* charts/kserve-crd-minimal/templates/
 	rm charts/kserve-crd-minimal/templates/kustomization.yaml
 
-	kubectl kustomize https://github.com/kubernetes-sigs/gateway-api-inference-extension/config/crd?ref=$(GIE_VERSION) > config/crd/external/gateway-inference-extension/gateway-inference-extension.yaml
+	kubectl kustomize "https://github.com/kubernetes-sigs/gateway-api-inference-extension/config/crd?ref=$(GIE_VERSION)" > config/crd/external/gateway-inference-extension/gateway-inference-extension.yaml
 
 	# InferenceModels is needed for now but after we start to use inferenceObjectives, we can remove this.
 	curl https://raw.githubusercontent.com/kubernetes-sigs/gateway-api-inference-extension/refs/tags/v0.5.1/config/crd/bases/inference.networking.x-k8s.io_inferencemodels.yaml  >> config/crd/external/gateway-inference-extension/gateway-inference-extension.yaml
@@ -198,6 +214,9 @@ test: fmt vet manifests envtest test-qpext
 test-qpext:
 	cd qpext && go test -v ./... -cover
 
+test-llmisvc: fmt vet manifests envtest
+	KUBEBUILDER_ASSETS="$$($(ENVTEST) use $(ENVTEST_K8S_VERSION) -p path)" go test --timeout 20m ./pkg/controller/llmisvc/... -coverprofile coverage.out -coverpkg ./pkg/... ./cmd...
+
 # Build manager binary
 manager: generate fmt vet go-lint
 	go build -o bin/manager ./cmd/manager
@@ -305,7 +324,7 @@ docker-build:
 
 # Push the docker image
 docker-push:
-	docker push ${IMG}
+	${ENGINE} push ${IMG}
 
 docker-build-agent:
 	${ENGINE} buildx build ${ARCH} -f agent.Dockerfile . -t ${KO_DOCKER_REPO}/${AGENT_IMG}

Makefile.tools.mk

@@ -17,7 +17,7 @@ POETRY = $(PYTHON_BIN)/poetry
 
 ## Tool versions.
 GOLANGCI_LINT_VERSION ?= v1.64.8
-CONTROLLER_TOOLS_VERSION ?= v0.16.2
+CONTROLLER_TOOLS_VERSION ?= v0.17.2
 ENVTEST_VERSION ?= latest
 YQ_VERSION ?= v4.28.1
 HELM_DOCS_VERSION ?= v1.12.0

agent.Dockerfile

@@ -6,6 +6,8 @@ WORKDIR /go/src/github.com/kserve/kserve
 COPY go.mod  go.mod
 COPY go.sum  go.sum
 
+# Allow Go to automatically download the required toolchain version
+ENV GOTOOLCHAIN=auto
 RUN go mod download
 
 COPY cmd/    cmd/
@@ -17,10 +19,15 @@ RUN CGO_ENABLED=0 GOOS=linux GOFLAGS=-mod=mod go build -a -o agent ./cmd/agent
 
 # Generate third-party licenses
 COPY LICENSE LICENSE
+# go-licenses is temporarily disabled due to incompatibility with Go 1.24+
+# go-licenses v1.6.0 has a compilation error with Go 1.24 due to github.com/otiai10/copy dependency
+# go-licenses v2 is available as pre-release but not yet stable
+# See: https://github.com/google/go-licenses/issues/312
+# Uncomment these lines once go-licenses v2 is stable or v1.x adds Go 1.24+ support
 # RUN go install github.com/google/go-licenses@latest
 # Forbidden Licenses: https://github.com/google/licenseclassifier/blob/e6a9bb99b5a6f71d5a34336b8245e305f5430f99/license_type.go#L341
 # RUN go-licenses check ./cmd/... ./pkg/... --disallowed_types="forbidden,unknown"
-#RUN go-licenses save --save_path third_party/library ./cmd/agent
+# RUN go-licenses save --save_path third_party/library ./cmd/agent
 
 # Copy the inference-agent into a thin image
 FROM registry.access.redhat.com/ubi9/ubi-minimal:latest

charts/kserve-crd-minimal/templates/serving.kserve.io_clusterservingruntimes.yaml

@@ -2,7 +2,7 @@ apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   annotations:
-    controller-gen.kubebuilder.io/version: v0.16.2
+    controller-gen.kubebuilder.io/version: v0.17.2
   name: clusterservingruntimes.serving.kserve.io
 spec:
   group: serving.kserve.io

charts/kserve-crd-minimal/templates/serving.kserve.io_clusterstoragecontainers.yaml

@@ -2,7 +2,7 @@ apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   annotations:
-    controller-gen.kubebuilder.io/version: v0.16.2
+    controller-gen.kubebuilder.io/version: v0.17.2
   name: clusterstoragecontainers.serving.kserve.io
 spec:
   group: serving.kserve.io

charts/kserve-crd-minimal/templates/serving.kserve.io_inferencegraphs.yaml

@@ -2,7 +2,7 @@ apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   annotations:
-    controller-gen.kubebuilder.io/version: v0.16.2
+    controller-gen.kubebuilder.io/version: v0.17.2
   name: inferencegraphs.serving.kserve.io
 spec:
   group: serving.kserve.io

charts/kserve-crd-minimal/templates/serving.kserve.io_inferenceservices.yaml

@@ -2,7 +2,7 @@ apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   annotations:
-    controller-gen.kubebuilder.io/version: v0.16.2
+    controller-gen.kubebuilder.io/version: v0.17.2
   name: inferenceservices.serving.kserve.io
 spec:
   group: serving.kserve.io