fix: Enhance event logging and improve shutdown handling in processors

Author: Aias00

internal/transport/netty_client.go

@@ -154,7 +154,24 @@ func (c *NettyClient) SetEventHandler(handler EventHandler) {
 }
 
 func (c *NettyClient) triggerEvent(eventType EventType, err error) {
-	log.Printf("Triggering event: %d, error: %v", eventType, err)
+	eventName := ""
+	switch eventType {
+	case EventConnected:
+		eventName = "Connected"
+	case EventDisconnected:
+		eventName = "Disconnected"
+	case EventConnectFailed:
+		eventName = "ConnectFailed"
+	default:
+		eventName = fmt.Sprintf("Unknown(%d)", eventType)
+	}
+
+	if err != nil {
+		log.Printf("Triggering event: %s, error: %v", eventName, err)
+	} else {
+		log.Printf("Triggering event: %s", eventName)
+	}
+
 	if c.eventHandler != nil {
 		c.eventHandler(Event{
 			Type:    eventType,

internal/transport/processors.go

@@ -19,6 +19,8 @@ package transport
 
 import (
 	"fmt"
+	"log"
+	"time"
 
 	pb "hertzbeat.apache.org/hertzbeat-collector-go/api"
 )
@@ -80,17 +82,25 @@ func NewGoOfflineProcessor(client *NettyClient) *GoOfflineProcessor {
 }
 
 func (p *GoOfflineProcessor) Process(msg *pb.Message) (*pb.Message, error) {
-	// Handle go offline message - shutdown the client
-	if p.client != nil {
-		// Stop heartbeat and close connection
-		p.client.Shutdown()
-	}
-	return &pb.Message{
+	// Handle go offline message - first return response, then shutdown
+	// Create response first
+	response := &pb.Message{
 		Type:      pb.MessageType_GO_OFFLINE,
 		Direction: pb.Direction_RESPONSE,
 		Identity:  msg.Identity,
 		Msg:       []byte("offline ack"),
-	}, nil
+	}
+
+	// Schedule shutdown after a brief delay to allow response to be sent
+	if p.client != nil {
+		go func() {
+			time.Sleep(100 * time.Millisecond) // Brief delay to ensure response is sent
+			log.Printf("Shutting down client as requested by manager")
+			p.client.Shutdown()
+		}()
+	}
+
+	return response, nil
 }
 
 // GoCloseProcessor handles go close messages