feat(multiplayer): expose allocator quota metrics

server/cmd/allocator/main.go

@@ -5,6 +5,7 @@ import (
 	"database/sql"
 	"flag"
 	"log"
+	"net/http"
 	"os"
 	"os/signal"
 	"syscall"
@@ -27,6 +28,7 @@ func main() {
 	workloadSecret := flag.String("workload-secret", os.Getenv("COSMIC_CLASH_WORKLOAD_SECRET"), "HMAC secret for control-plane-issued workload tokens (see workload/signed_token.go); must match cmd/control-plane's own --workload-secret. Unset skips minting a cosmic-clash.io/workload-token annotation entirely")
 	allocationQuota := flag.Int("allocation-quota", 0, "optional per-replica allocation attempts per region per quota window; zero disables this local guard")
 	allocationQuotaWindow := flag.Duration("allocation-quota-window", time.Minute, "window for --allocation-quota")
+	metricsAddr := flag.String("metrics-addr", envOrDefault("COSMIC_CLASH_ALLOCATOR_METRICS_ADDR", ":9091"), "allocator Prometheus metrics address; empty disables metrics")
 	flag.Parse()
 	if *dsn == "" || *agonesURL == "" {
 		fatalf("--dsn/COSMIC_CLASH_POSTGRES_DSN and --agones-url/COSMIC_CLASH_AGONES_URL are required")
@@ -62,6 +64,7 @@ func main() {
 		}
 		log.Printf("allocator: enabled per-replica regional allocation quota=%d window=%s", *allocationQuota, *allocationQuotaWindow)
 	}
+	metrics := allocator.NewMetrics()
 	client := agones.Client{BaseURL: *agonesURL, Namespace: *namespace, WorkloadSecret: []byte(*workloadSecret)}
 	worker := allocator.Worker{
 		Claims: store.AllocatingMatchClaims{DB: db, Transport: *transport},
@@ -70,12 +73,30 @@ func main() {
 			Durable:  store.AllocationRegistry{DB: db},
 			Quota:    store.AllocationQuota{DB: db},
 			Budget:   budget,
+			Metrics:  metrics,
 			Now:      now,
 		},
 		Now: now,
 	}
 	ctx, stop := signal.NotifyContext(context.Background(), os.Interrupt, syscall.SIGTERM)
 	defer stop()
+	var metricsServer *http.Server
+	if *metricsAddr != "" {
+		metricsServer = &http.Server{Addr: *metricsAddr, Handler: allocator.MetricsHandler(metrics)}
+		go func() {
+			if err := metricsServer.ListenAndServe(); err != nil && err != http.ErrServerClosed {
+				log.Printf("allocator: metrics server: %v", err)
+			}
+		}()
+		log.Printf("allocator: metrics listening on %s", *metricsAddr)
+	}
+	defer func() {
+		if metricsServer != nil {
+			shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+			defer cancel()
+			_ = metricsServer.Shutdown(shutdownCtx)
+		}
+	}()
 	ticker := time.NewTicker(*interval)
 	defer ticker.Stop()
 	for {