diff --git a/config/manifests/inferencepool-resources.yaml b/config/manifests/inferencepool-resources.yaml index 993b7bf62..3d978292c 100644 --- a/config/manifests/inferencepool-resources.yaml +++ b/config/manifests/inferencepool-resources.yaml @@ -4,7 +4,6 @@ apiVersion: inference.networking.x-k8s.io/v1alpha2 kind: InferencePool metadata: - labels: name: vllm-llama3-8b-instruct spec: targetPortNumber: 8000 @@ -54,6 +53,8 @@ spec: args: - -poolName - "vllm-llama3-8b-instruct" + - "-poolNamespace" + - "default" - -v - "4" - --zap-encoder diff --git a/pkg/epp/controller/inferencepool_reconciler.go b/pkg/epp/controller/inferencepool_reconciler.go index c92d4eccb..0738181f0 100644 --- a/pkg/epp/controller/inferencepool_reconciler.go +++ b/pkg/epp/controller/inferencepool_reconciler.go @@ -21,7 +21,6 @@ import ( "reflect" "k8s.io/apimachinery/pkg/api/errors" - "k8s.io/apimachinery/pkg/types" "k8s.io/client-go/tools/record" ctrl "sigs.k8s.io/controller-runtime" "sigs.k8s.io/controller-runtime/pkg/client" @@ -36,9 +35,8 @@ import ( // will have the proper controller that will create/manage objects on behalf of the server pool. type InferencePoolReconciler struct { client.Client - Record record.EventRecorder - PoolNamespacedName types.NamespacedName - Datastore datastore.Datastore + Record record.EventRecorder + Datastore datastore.Datastore } func (c *InferencePoolReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) { diff --git a/pkg/epp/controller/inferencepool_reconciler_test.go b/pkg/epp/controller/inferencepool_reconciler_test.go index 7e5d4801f..b7e283346 100644 --- a/pkg/epp/controller/inferencepool_reconciler_test.go +++ b/pkg/epp/controller/inferencepool_reconciler_test.go @@ -96,7 +96,7 @@ func TestInferencePoolReconciler(t *testing.T) { pmf := backendmetrics.NewPodMetricsFactory(&backendmetrics.FakePodMetricsClient{}, time.Second) datastore := datastore.NewDatastore(ctx, pmf) - inferencePoolReconciler := &InferencePoolReconciler{PoolNamespacedName: namespacedName, Client: fakeClient, Datastore: datastore} + inferencePoolReconciler := &InferencePoolReconciler{Client: fakeClient, Datastore: datastore} // Step 1: Inception, only ready pods matching pool1 are added to the store. if _, err := inferencePoolReconciler.Reconcile(ctx, req); err != nil { diff --git a/pkg/epp/server/controller_manager.go b/pkg/epp/server/controller_manager.go index aaad89768..ce5cfc893 100644 --- a/pkg/epp/server/controller_manager.go +++ b/pkg/epp/server/controller_manager.go @@ -39,8 +39,8 @@ func init() { utilruntime.Must(v1alpha2.Install(scheme)) } -// DefaultManagerOptions returns the default options used to create the manager. -func DefaultManagerOptions(namespace, name string) ctrl.Options { +// defaultManagerOptions returns the default options used to create the manager. +func defaultManagerOptions(namespace string, name string) ctrl.Options { return ctrl.Options{ Scheme: scheme, Cache: cache.Options{ @@ -71,7 +71,7 @@ func DefaultManagerOptions(namespace, name string) ctrl.Options { // NewDefaultManager creates a new controller manager with default configuration. func NewDefaultManager(namespace, name string, restConfig *rest.Config) (ctrl.Manager, error) { - manager, err := ctrl.NewManager(restConfig, DefaultManagerOptions(namespace, name)) + manager, err := ctrl.NewManager(restConfig, defaultManagerOptions(namespace, name)) if err != nil { return nil, fmt.Errorf("failed to create controller manager: %v", err) } diff --git a/pkg/epp/server/runserver.go b/pkg/epp/server/runserver.go index aa048e6e6..65a6e7879 100644 --- a/pkg/epp/server/runserver.go +++ b/pkg/epp/server/runserver.go @@ -87,11 +87,7 @@ func (r *ExtProcServerRunner) SetupWithManager(ctx context.Context, mgr ctrl.Man if err := (&controller.InferencePoolReconciler{ Datastore: r.Datastore, Client: mgr.GetClient(), - PoolNamespacedName: types.NamespacedName{ - Name: r.PoolName, - Namespace: r.PoolNamespace, - }, - Record: mgr.GetEventRecorderFor("InferencePool"), + Record: mgr.GetEventRecorderFor("InferencePool"), }).SetupWithManager(mgr); err != nil { return fmt.Errorf("failed setting up InferencePoolReconciler: %w", err) } diff --git a/site-src/implementations/gateways.md b/site-src/implementations/gateways.md index d4e919be0..a5717d6ae 100644 --- a/site-src/implementations/gateways.md +++ b/site-src/implementations/gateways.md @@ -5,10 +5,12 @@ This project has several implementations that are planned or in progress: * [Envoy AI Gateway][1] * [Kgateway][2] * [Google Kubernetes Engine][3] +* [Istio][4] [1]:#envoy-gateway [2]:#kgateway [3]:#google-kubernetes-engine +[4]:#istio ## Envoy AI Gateway