What causes "channel for udp listener closed" in fatedier/frp?

Treat it as a shutdown signal: break out of the accept loop and clean up per-connection state, do not retry

channel for udp listener closed

Error message

channel for udp listener closed

What it means

Returned by UDPListener.Accept when the internal accept channel has been closed. Accept models itself after net.Listener, but connections arrive via a channel; once the listener is closed and the channel is closed, a receive yields ok==false and this error is produced. It is the normal 'listener is shutting down' signal, equivalent to net.ErrClosed for a TCP listener.

Source

Thrown at pkg/util/net/udp.go:241

	}()
	l.writeCh <- packet
	return
}

func (l *UDPListener) WriteMsg(buf []byte, remoteAddr *net.UDPAddr) (err error) {
	// only set remote addr here
	packet := &UDPPacket{
		Buf:        buf,
		RemoteAddr: remoteAddr,
	}
	err = l.writeUDPPacket(packet)
	return
}

func (l *UDPListener) Accept() (net.Conn, error) {
	conn, ok := <-l.acceptCh
	if !ok {
		return conn, fmt.Errorf("channel for udp listener closed")
	}
	return conn, nil
}

func (l *UDPListener) Close() error {
	if !l.closeFlag {
		l.closeFlag = true
		if l.readConn != nil {
			l.readConn.Close()
		}
	}
	return nil
}

func (l *UDPListener) Addr() net.Addr {
	return l.addr
}

View on GitHub (pinned to 6c8a8d0a97)

Solutions

Treat it as a shutdown signal: break out of the accept loop and clean up per-connection state, do not retry
Ensure Close() is the last operation: cancel the parent context first so accept loops exit before teardown
If it appears outside of shutdown, look for double-Close or premature Close caused by a failed initialization path

Example fix

// before
for {
    conn, err := l.Accept()
    if err != nil { log.Fatalf("accept: %v", err) }
}

// after
for {
    conn, err := l.Accept()
    if err != nil {
        if strings.Contains(err.Error(), "channel for udp listener closed") { break }
        log.Fatalf("accept: %v", err)
    }
}

Defensive patterns

Strategy: try-catch

Try / catch

for {
    conn, err := l.Accept()
    if err != nil {
        if strings.Contains(err.Error(), "channel for udp listener closed") {
            break // normal shutdown
        }
        return err
    }
    go handle(conn)
}

Prevention

Model Accept error as shutdown sentinel and always break, never log.Fatal
Coordinate Close with accept loops via context so loops exit first
Ignore this line in logs when it coincides with process shutdown

When it happens

Trigger: Calling Accept() after Close() on the same UDPListener, or after the reader goroutine has shut down and closed acceptCh; typical in accept loops like `for { conn, err := l.Accept(); if err != nil { break } ... }`.

Common situations: frps or a vhost UDP worker shutting down while its accept loop is still running; test harness closing the listener then calling Accept once more. Seeing this in logs during process shutdown is expected and benign.

Related errors

AI-assisted analysis of fatedier/frp@6c8a8d0a97 (2026-08-15). Data as JSON: /api/errors/1285e6b14a2f6098. Report an issue: GitHub.