What causes "wait detect message canceled" in fatedier/frp?

No defect: treat as a clean cancel; ensure the caller distinguishes ctx.Err() from real punch failures

wait detect message canceled

Error message

wait detect message canceled

What it means

MakeHole aborted because the parent context was canceled while waiting for detect messages across the multiple listen sockets. This is a caller-initiated cancellation (shutdown, visit canceled, upper-layer deadline), not a network failure itself.

Source

Thrown at pkg/nathole/nathole.go:286

			if err != nil {
				lConn.Close()
				return
			}
			select {
			case resultCh <- result{lConn: lConn, raddr: addr}:
			default:
				lConn.Close()
			}
		}(conn)
	}

	select {
	case result := <-resultCh:
		return result.lConn, result.raddr, nil
	case <-time.After(timeout):
		return nil, nil, fmt.Errorf("wait detect message timeout")
	case <-ctx.Done():
		return nil, nil, fmt.Errorf("wait detect message canceled")
	}
}

func waitDetectMessage(
	ctx context.Context, conn *net.UDPConn, sid string, key []byte,
	timeout time.Duration, role string,
) (*net.UDPAddr, error) {
	xl := xlog.FromContextSafe(ctx)
	for {
		buf := pool.GetBuf(1024)
		_ = conn.SetReadDeadline(time.Now().Add(timeout))
		n, raddr, err := conn.ReadFromUDP(buf)
		_ = conn.SetReadDeadline(time.Time{})
		if err != nil {
			pool.PutBuf(buf)
			return nil, err
		}
		xl.Debugf("get udp message local %s, from %s", conn.LocalAddr(), raddr)

View on GitHub (pinned to 6c8a8d0a97)

Solutions

No defect: treat as a clean cancel; ensure the caller distinguishes ctx.Err() from real punch failures
If premature, lengthen or remove the parent deadline so it exceeds DetectBehavior.ReadTimeoutMs
Avoid stopping frpc or closing the visitor while P2P setup is in flight if the connection is wanted

Defensive patterns

Strategy: try-catch

Try / catch

if ctx.Err() != nil {
    // cancellation, not a punch failure — do not count toward retry budgets
    return ctx.Err()
}

Prevention

Check ctx.Err() first when MakeHole returns an error to separate shutdown from failure
Keep parent deadlines longer than the nathole detect timeout
Do not stop frpc mid-visit if P2P setup is expected to finish

When it happens

Trigger: The ctx passed into MakeHole is canceled while the select over resultCh/time.After is blocked — e.g. frpc shuts down, the visitor proxy is closed, or a parent timeout fires first.

Common situations: User cancels the visit or stops frpc during hole punching; a wrapping context deadline (shorter than the detect timeout) expires; graceful shutdown racing an in-flight P2P connection setup.

Related errors

AI-assisted analysis of fatedier/frp@6c8a8d0a97 (2026-08-15). Data as JSON: /api/errors/7874520918ff7023. Report an issue: GitHub.