enhance the log init process

cmd/bbr/health.go

@@ -25,7 +25,7 @@ import (
 	healthPb "google.golang.org/grpc/health/grpc_health_v1"
 	"google.golang.org/grpc/status"
 
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 )
 
 type healthServer struct {

cmd/bbr/main.go

@@ -22,8 +22,6 @@ import (
 	"os"
 
 	"github.com/go-logr/logr"
-	uberzap "go.uber.org/zap"
-	"go.uber.org/zap/zapcore"
 	"google.golang.org/grpc"
 	healthPb "google.golang.org/grpc/health/grpc_health_v1"
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -33,9 +31,9 @@ import (
 	metricsserver "sigs.k8s.io/controller-runtime/pkg/metrics/server"
 
 	"sigs.k8s.io/gateway-api-inference-extension/internal/runnable"
+	"sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/bbr/metrics"
 	runserver "sigs.k8s.io/gateway-api-inference-extension/pkg/bbr/server"
-	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 var (
@@ -51,7 +49,7 @@ var (
 		"metrics-port", 9090, "The metrics port")
 	streaming = flag.Bool(
 		"streaming", false, "Enables streaming support for Envoy full-duplex streaming mode")
-	logVerbosity = flag.Int("v", logging.DEFAULT, "number for the log level verbosity")
+	logVerbosity = flag.Int("log-verbosity", logging.DEFAULT, "number for the log level verbosity, higher verbosity means more (and less important) logs will be generated")
 
 	setupLog = ctrl.Log.WithName("setup")
 )
@@ -66,7 +64,9 @@ func run() error {
 	opts := zap.Options{Development: true}
 	opts.BindFlags(flag.CommandLine)
 	flag.Parse()
-	initLogging(&opts)
+
+	// --- set up logger
+	ctrl.SetLogger(logging.InitLogging(&opts))
 
 	// Print all flag values
 	flags := make(map[string]any)
@@ -138,20 +138,3 @@ func registerHealthServer(mgr manager.Manager, logger logr.Logger, port int) err
 	}
 	return nil
 }
-
-func initLogging(opts *zap.Options) {
-	useV := true
-	flag.Visit(func(f *flag.Flag) {
-		if f.Name == "zap-log-level" {
-			useV = false
-		}
-	})
-	if useV {
-		// See https://pkg.go.dev/sigs.k8s.io/controller-runtime/pkg/log/zap#Options.Level
-		lvl := -1 * (*logVerbosity)
-		opts.Level = uberzap.NewAtomicLevelAt(zapcore.Level(int8(lvl)))
-	}
-
-	logger := zap.New(zap.UseFlagOptions(opts), zap.RawZapOpts(uberzap.AddCaller()))
-	ctrl.SetLogger(logger)
-}

cmd/epp/runner/health.go

@@ -27,8 +27,8 @@ import (
 	healthPb "google.golang.org/grpc/health/grpc_health_v1"
 	"google.golang.org/grpc/status"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datastore"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 type healthServer struct {

cmd/epp/runner/runner.go

@@ -30,8 +30,6 @@ import (
 
 	"github.com/go-logr/logr"
 	"github.com/prometheus/client_golang/prometheus"
-	uberzap "go.uber.org/zap"
-	"go.uber.org/zap/zapcore"
 	"google.golang.org/grpc"
 	healthPb "google.golang.org/grpc/health/grpc_health_v1"
 	"k8s.io/apimachinery/pkg/runtime/schema"
@@ -44,6 +42,7 @@ import (
 	metricsserver "sigs.k8s.io/controller-runtime/pkg/metrics/server"
 
 	"sigs.k8s.io/gateway-api-inference-extension/internal/runnable"
+	"sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/common"
 	backendmetrics "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/backend/metrics"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/config/loader"
@@ -63,7 +62,6 @@ import (
 	testfilter "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/framework/plugins/test/filter"
 	runserver "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/server"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/env"
-	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/version"
 )
 
@@ -81,7 +79,6 @@ var (
 	poolName       = flag.String("pool-name", runserver.DefaultPoolName, "Name of the InferencePool this Endpoint Picker is associated with.")
 	poolGroup      = flag.String("pool-group", runserver.DefaultPoolGroup, "group of the InferencePool this Endpoint Picker is associated with.")
 	poolNamespace  = flag.String("pool-namespace", "", "Namespace of the InferencePool this Endpoint Picker is associated with.")
-	logVerbosity   = flag.Int("v", logging.DEFAULT, "number for the log level verbosity")
 	secureServing  = flag.Bool("secure-serving", runserver.DefaultSecureServing, "Enables secure serving. Defaults to true.")
 	healthChecking = flag.Bool("health-checking", runserver.DefaultHealthChecking, "Enables health checking")
 	certPath       = flag.String("cert-path", runserver.DefaultCertPath, "The path to the certificate for secure serving. The certificate and private key files "+
@@ -134,13 +131,11 @@ func (r *Runner) WithSchedulerConfig(schedulerConfig *scheduling.SchedulerConfig
 }
 
 func (r *Runner) Run(ctx context.Context) error {
-	opts := zap.Options{
-		Development: true,
-	}
+	opts := zap.Options{Development: true}
 	opts.BindFlags(flag.CommandLine)
 	flag.Parse()
-	initLogging(&opts)
 
+	ctrl.SetLogger(logging.InitLogging(&opts))
 	setupLog.Info("GIE build", "commit-sha", version.CommitSHA, "build-ref", version.BuildRef)
 
 	// Validate flags
@@ -430,24 +425,6 @@ func setupDatalayer() (datalayer.EndpointFactory, error) {
 	return factory, nil
 }
 
-func initLogging(opts *zap.Options) {
-	// Unless -zap-log-level is explicitly set, use -v
-	useV := true
-	flag.Visit(func(f *flag.Flag) {
-		if f.Name == "zap-log-level" {
-			useV = false
-		}
-	})
-	if useV {
-		// See https://pkg.go.dev/sigs.k8s.io/controller-runtime/pkg/log/zap#Options.Level
-		lvl := -1 * (*logVerbosity)
-		opts.Level = uberzap.NewAtomicLevelAt(zapcore.Level(int8(lvl)))
-	}
-
-	logger := zap.New(zap.UseFlagOptions(opts), zap.RawZapOpts(uberzap.AddCaller()))
-	ctrl.SetLogger(logger)
-}
-
 // registerExtProcServer adds the ExtProcServerRunner as a Runnable to the manager.
 func registerExtProcServer(mgr manager.Manager, runner *runserver.ExtProcServerRunner, logger logr.Logger) error {
 	if err := mgr.Add(runner.AsRunnable(logger)); err != nil {

config/charts/body-based-routing/templates/bbr.yaml

@@ -19,8 +19,8 @@ spec:
         imagePullPolicy: {{ .Values.bbr.image.pullPolicy | default "Always" }}
         args:
         - "--streaming"
-        - "--v"
-        - "3"
+        - "--zap-log-level"
+        - "info"
         ports:
         - containerPort: {{ .Values.bbr.port }}
         # health check

config/charts/inferencepool/templates/epp-deployment.yaml

@@ -43,6 +43,8 @@ spec:
         {{- end }}
         - --zap-encoder
         - "json"
+        - --zap-log-level
+        - "info"
         - --config-file
         - "/config/{{ .Values.inferenceExtension.pluginsConfigFile }}"
         {{- if eq (.Values.inferencePool.modelServerType | default "vllm") "triton-tensorrt-llm" }}

pkg/epp/util/logging/logger.go → internal/telemetry/logging/logger.go

@@ -18,13 +18,21 @@ package logging
 
 import (
 	"context"
+	"os"
 
 	"github.com/go-logr/logr"
 	uberzap "go.uber.org/zap"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 	"sigs.k8s.io/controller-runtime/pkg/log/zap"
 )
 
+const (
+	DEFAULT = 2
+	VERBOSE = 3
+	DEBUG   = 4
+	TRACE   = 5
+)
+
 // NewTestLogger creates a new Zap logger using the dev mode.
 func NewTestLogger() logr.Logger {
 	return zap.New(zap.UseDevMode(true), zap.RawZapOpts(uberzap.AddCaller()))
@@ -34,3 +42,15 @@ func NewTestLogger() logr.Logger {
 func NewTestLoggerIntoContext(ctx context.Context) context.Context {
 	return log.IntoContext(ctx, zap.New(zap.UseDevMode(true), zap.RawZapOpts(uberzap.AddCaller())))
 }
+
+// Fatal calls logger.Error followed by os.Exit(1).
+//
+// This is a utility function and should not be used in production code!
+func Fatal(logger logr.Logger, err error, msg string, keysAndValues ...any) {
+	logger.Error(err, msg, keysAndValues...)
+	os.Exit(1)
+}
+
+func InitLogging(opts *zap.Options) logr.Logger {
+	return zap.New(zap.UseFlagOptions(opts), zap.RawZapOpts(uberzap.AddCaller()))
+}

pkg/bbr/handlers/request.go

@@ -24,8 +24,8 @@ import (
 	eppb "github.com/envoyproxy/go-control-plane/envoy/service/ext_proc/v3"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/bbr/metrics"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 const modelHeader = "X-Gateway-Model-Name"

pkg/bbr/handlers/request_test.go

@@ -29,8 +29,8 @@ import (
 	metricsutils "k8s.io/component-base/metrics/testutil"
 	crmetrics "sigs.k8s.io/controller-runtime/pkg/metrics"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/bbr/metrics"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 func TestHandleRequestBody(t *testing.T) {

pkg/bbr/handlers/server.go

@@ -27,7 +27,7 @@ import (
 	"google.golang.org/grpc/status"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	requtil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/request"
 )
 

pkg/bbr/handlers/server_test.go

@@ -26,7 +26,7 @@ import (
 	"google.golang.org/protobuf/testing/protocmp"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 )
 
 func TestProcessRequestBody(t *testing.T) {

pkg/epp/backend/metrics/fake.go

@@ -26,9 +26,9 @@ import (
 	"k8s.io/apimachinery/pkg/types"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/backend"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 // FakePodMetrics is an implementation of PodMetrics that doesn't run the async refresh loop.

pkg/epp/backend/metrics/logger.go

@@ -24,9 +24,9 @@ import (
 	"github.com/go-logr/logr"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/metrics"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 const (

pkg/epp/backend/metrics/metrics_test.go

@@ -31,8 +31,8 @@ import (
 	"google.golang.org/protobuf/proto"
 	"k8s.io/apimachinery/pkg/types"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/backend"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 // --- Test Helpers ---

pkg/epp/backend/metrics/pod_metrics.go

@@ -27,9 +27,9 @@ import (
 	corev1 "k8s.io/api/core/v1"
 	"k8s.io/apimachinery/pkg/types"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/backend"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 const (

pkg/epp/config/loader/configloader_test.go

@@ -27,14 +27,14 @@ import (
 	"k8s.io/utils/ptr"
 
 	configapi "sigs.k8s.io/gateway-api-inference-extension/apix/config/v1alpha1"
+	"sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/config"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/plugins"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/framework"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/framework/plugins/multi/prefix"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/framework/plugins/picker"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/framework/plugins/profile"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/scheduling/types"
-	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/test/utils"
 )
 

pkg/epp/controller/inferenceobjective_reconciler.go

@@ -28,9 +28,9 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/predicate"
 
 	"sigs.k8s.io/gateway-api-inference-extension/apix/v1alpha2"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/common"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datastore"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 type InferenceObjectiveReconciler struct {

pkg/epp/controller/inferencepool_reconciler.go

@@ -27,9 +27,9 @@ import (
 
 	v1 "sigs.k8s.io/gateway-api-inference-extension/api/v1"
 	"sigs.k8s.io/gateway-api-inference-extension/apix/v1alpha2"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/common"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datastore"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 // InferencePoolReconciler utilizes the controller runtime to reconcile Instance Gateway resources

pkg/epp/controller/pod_reconciler.go

@@ -30,8 +30,8 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/log"
 	"sigs.k8s.io/controller-runtime/pkg/predicate"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datastore"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 	podutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/pod"
 )
 

pkg/epp/datalayer/collector.go

@@ -19,12 +19,12 @@ package datalayer
 import (
 	"context"
 	"errors"
+
 	"sync"
 	"time"
 
 	"sigs.k8s.io/controller-runtime/pkg/log"
-
-	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
+	"sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 )
 
 // TODO:

pkg/epp/datalayer/metrics/logger.go

@@ -23,9 +23,9 @@ import (
 	"github.com/go-logr/logr"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/metrics"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 const debugPrintInterval = 5 * time.Second

pkg/epp/datastore/datastore.go

@@ -31,10 +31,10 @@ import (
 
 	v1 "sigs.k8s.io/gateway-api-inference-extension/api/v1"
 	"sigs.k8s.io/gateway-api-inference-extension/apix/v1alpha2"
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	backendmetrics "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/backend/metrics"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer"
 	dlmetrics "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/datalayer/metrics"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 	podutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/pod"
 )
 

pkg/epp/flowcontrol/controller/internal/processor.go

@@ -29,10 +29,10 @@ import (
 	"k8s.io/utils/clock"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 
+	logutil "sigs.k8s.io/gateway-api-inference-extension/internal/telemetry/logging"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/flowcontrol/contracts"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/flowcontrol/framework"
 	"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/flowcontrol/types"
-	logutil "sigs.k8s.io/gateway-api-inference-extension/pkg/epp/util/logging"
 )
 
 // maxCleanupWorkers caps the number of concurrent workers for background cleanup tasks. This prevents a single shard