optimize MetricsBindAddress to MetricsBindPort

Signed-off-by: phoenixwu0229 <[email protected]>

charts/hami/templates/scheduler/deployment.yaml

@@ -84,7 +84,7 @@ spec:
             - --cert_file=/tls/tls.crt
             - --key_file=/tls/tls.key
             - --scheduler-name={{ .Values.schedulerName }}
-            - --metrics-bind-address={{ .Values.scheduler.metricsBindAddress }}
+            - --metrics-bind-port={{ .Values.scheduler.metricsBindPort }}
             - --node-scheduler-policy={{ .Values.scheduler.defaultSchedulerPolicy.nodeSchedulerPolicy }}
             - --gpu-scheduler-policy={{ .Values.scheduler.defaultSchedulerPolicy.gpuSchedulerPolicy }}
             - --device-config-file=/device-config.yaml

charts/hami/templates/scheduler/service.yaml

@@ -21,7 +21,7 @@ spec:
       protocol: TCP
     - name: monitor
       port: {{ .Values.scheduler.service.monitorPort }}
-      targetPort: 9395
+      targetPort: {{ .Values.scheduler.metricsBindPort }}
       nodePort: {{ .Values.scheduler.service.monitorPort }}
       protocol: TCP
   selector:

charts/hami/values.yaml

@@ -52,7 +52,7 @@ scheduler:
   defaultSchedulerPolicy:
     nodeSchedulerPolicy: binpack
     gpuSchedulerPolicy: spread
-  metricsBindAddress: ":9395"
+  metricsBindPort: 9395
   livenessProbe: false
   leaderElect: true
   kubeScheduler:

cmd/scheduler/main.go

@@ -61,7 +61,7 @@ func init() {
 	rootCmd.Flags().Int32Var(&config.DefaultResourceNum, "default-gpu", 1, "default gpu to allocate")
 	rootCmd.Flags().StringVar(&config.NodeSchedulerPolicy, "node-scheduler-policy", util.NodeSchedulerPolicyBinpack.String(), "node scheduler policy")
 	rootCmd.Flags().StringVar(&config.GPUSchedulerPolicy, "gpu-scheduler-policy", util.GPUSchedulerPolicySpread.String(), "GPU scheduler policy")
-	rootCmd.Flags().StringVar(&config.MetricsBindAddress, "metrics-bind-address", ":9395", "The TCP address that the scheduler should bind to for serving prometheus metrics(e.g. 127.0.0.1:9395, :9395)")
+	rootCmd.Flags().StringVar(&config.MetricsBindPort, "metrics-bind-port", ":9395", "The port that the scheduler should bind to for serving prometheus metrics(e.g. 9395)")
 	rootCmd.Flags().StringToStringVar(&config.NodeLabelSelector, "node-label-selector", nil, "key=value pairs separated by commas")
 	rootCmd.PersistentFlags().AddGoFlagSet(device.GlobalFlagSet())
 	rootCmd.AddCommand(version.VersionCmd)
@@ -76,7 +76,7 @@ func start() {
 
 	// start monitor metrics
 	go sher.RegisterFromNodeAnnotations()
-	go initMetrics(config.MetricsBindAddress)
+	go initMetrics(config.MetricsBindPort)
 
 	// start http server
 	router := httprouter.New()

cmd/scheduler/metrics.go

@@ -19,6 +19,7 @@ package main
 import (
 	"fmt"
 	"log"
+	"net"
 	"net/http"
 	"strings"
 
@@ -246,7 +247,7 @@ func NewClusterManager(zone string, reg prometheus.Registerer) *ClusterManager {
 	return c
 }
 
-func initMetrics(bindAddress string) {
+func initMetrics(bindPort string) {
 	// Since we are dealing with custom Collector implementations, it might
 	// be a good idea to try it out with a pedantic registry.
 	klog.Info("Initializing metrics for scheduler")
@@ -257,5 +258,7 @@ func initMetrics(bindAddress string) {
 	NewClusterManager("vGPU", reg)
 
 	http.Handle("/metrics", promhttp.HandlerFor(reg, promhttp.HandlerOpts{}))
-	log.Fatal(http.ListenAndServe(bindAddress, nil))
+	addr := net.JoinHostPort("0.0.0.0", bindPort)
+	klog.Infof("Starting metrics server on %s", addr)
+	log.Fatal(http.ListenAndServe(addr, nil))
 }

pkg/scheduler/config/config.go

@@ -19,9 +19,9 @@ package config
 import "github.com/Project-HAMi/HAMi/pkg/util"
 
 var (
-	HTTPBind           string
-	SchedulerName      string
-	MetricsBindAddress string
+	HTTPBind        string
+	SchedulerName   string
+	MetricsBindPort string
 
 	DefaultMem         int32
 	DefaultCores       int32