Fix Connection timeout

core/cmd/dragontcp-client/chunk.go

@@ -263,20 +263,31 @@ func (l *requestLane) single(mode byte, sid wire.SessionID, seq uint64, payload
 	var lastErr error
 	for attempt := 0; attempt < 2; attempt++ {
 		if err := l.ensureLocked(); err != nil {
+			// A dial timeout is a dead path for this logical tunnel. Do not
+			// spend another timeout replaying the same request.
+			if isTransportTimeout(err) {
+				return 0, nil, err
+			}
 			lastErr = err
 			continue
 		}
 		reused := l.pc.requests > 0
 		_ = l.pc.conn.SetDeadline(time.Now().Add(timeout))
 		if err := wire.WriteRequestProfileEncoding(l.pc.conn, mode, sid, seq, payload, l.headerMask, l.coverProfile.Clear); err != nil {
-			lastErr = err
 			l.transportFailureLocked(reused)
+			if isTransportTimeout(err) {
+				return 0, nil, err
+			}
+			lastErr = err
 			continue
 		}
 		status, body, err := wire.ReadResponseProfile(l.pc.conn, l.headerMask)
 		if err != nil {
-			lastErr = err
 			l.transportFailureLocked(reused)
+			if isTransportTimeout(err) {
+				return 0, nil, err
+			}
+			lastErr = err
 			continue
 		}
 		l.pc.requests++
@@ -307,14 +318,20 @@ func (l *requestLane) download(sid wire.SessionID, startOffset, ackOffset uint64
 	var lastErr error
 	for attempt := 0; attempt < 2; attempt++ {
 		if err := l.ensureLocked(); err != nil {
+			if isTransportTimeout(err) {
+				return nil, 0, err
+			}
 			lastErr = err
 			continue
 		}
 		reused := l.pc.requests > 0
 		_ = l.pc.conn.SetDeadline(time.Now().Add(timeout))
 		if err := wire.WriteRequestProfileEncoding(l.pc.conn, wire.ModeDownload, sid, startOffset, payload, l.headerMask, l.coverProfile.Clear); err != nil {
-			lastErr = err
 			l.transportFailureLocked(reused)
+			if isTransportTimeout(err) {
+				return nil, 0, err
+			}
+			lastErr = err
 			continue
 		}
 
@@ -324,8 +341,11 @@ func (l *requestLane) download(sid wire.SessionID, startOffset, ackOffset uint64
 		for i := 0; i < count; i++ {
 			status, body, err := wire.ReadResponseProfile(l.pc.conn, l.headerMask)
 			if err != nil {
-				lastErr = err
 				l.transportFailureLocked(reused)
+				if isTransportTimeout(err) {
+					return nil, 0, err
+				}
+				lastErr = err
 				goto retry
 			}
 			lastStatus = status
@@ -585,6 +605,7 @@ type chunkConn struct {
 	maxPipeline    int
 
 	closeOnce sync.Once
+	dead      atomic.Bool
 }
 
 // appendChunkParts keeps the single-response fast path zero-copy. For a batch,
@@ -755,6 +776,10 @@ func (c *chunkConn) fillReadBuffer() error {
 			minFailures = 0
 		}
 		if err != nil {
+			if isTransportTimeout(err) {
+				c.dead.Store(true)
+				return err
+			}
 			if c.pipeline > c.minPipeline {
 				old := c.pipeline
 				c.pipeline /= 2
@@ -829,6 +854,10 @@ func (c *chunkConn) Write(p []byte) (int, error) {
 		n := minInt(size, len(p))
 		status, body, err := c.uploadLane.single(wire.ModeUpload, c.sid, c.upOffset, p[:n])
 		if err != nil {
+			if isTransportTimeout(err) {
+				c.dead.Store(true)
+				return total, err
+			}
 			old, next := c.upSizer.FailureReason(size, err)
 			if old == next && next == c.opts.minSize {
 				minFailures++
@@ -858,9 +887,13 @@ func (c *chunkConn) Write(p []byte) (int, error) {
 
 func (c *chunkConn) Close() error {
 	c.closeOnce.Do(func() {
-		// Reuse the upload lane rather than dialling a connection just to say
-		// goodbye; that was a second wasted dial per flow.
-		_, _, _ = c.uploadLane.single(wire.ModeClose, c.sid, 0, nil)
+		// A timed-out tunnel is already dead. Do not redial merely to send a
+		// graceful CLOSE for a session that the caller is abandoning.
+		if !c.dead.Load() {
+			// Reuse the upload lane rather than dialling a connection just to say
+			// goodbye; that was a second wasted dial per flow.
+			_, _, _ = c.uploadLane.single(wire.ModeClose, c.sid, 0, nil)
+		}
 		c.uploadLane.Close()
 		c.downloadLane.Close()
 	})