Merge branch 'main' into refactor/relay-foreign-cache

# Conflicts:
#	shared/relay/client/manager.go

shared/relay/client/dialer/ws/close_generic.go

@@ -0,0 +1,9 @@
+//go:build !js
+
+package ws
+
+// closeConn closes the underlying WebSocket immediately, skipping the close
+// handshake.
+func (c *Conn) closeConn() error {
+	return c.Conn.CloseNow()
+}

shared/relay/client/dialer/ws/close_js.go

@@ -0,0 +1,25 @@
+//go:build js
+
+package ws
+
+import (
+	"github.com/coder/websocket"
+	log "github.com/sirupsen/logrus"
+)
+
+// closeConn closes the browser WebSocket without blocking the caller.
+//
+// The browser close API only accepts codes 1000 and 3000-4999, so CloseNow's
+// 1001 (going away) throws an InvalidAccessError. Close with a valid code
+// waits for the browser close event before returning, which can park the
+// calling goroutine (the relay teardown path holds its mutexes while closing)
+// until the close handshake finishes. Run the close in the background and
+// report success; a teardown close error is not actionable.
+func (c *Conn) closeConn() error {
+	go func() {
+		if err := c.Conn.Close(websocket.StatusNormalClosure, ""); err != nil {
+			log.Debugf("failed to close relay websocket: %v", err)
+		}
+	}()
+	return nil
+}

shared/relay/client/dialer/ws/conn.go

@@ -77,5 +77,5 @@ func (c *Conn) SetDeadline(t time.Time) error {
 }
 
 func (c *Conn) Close() error {
-	return c.Conn.CloseNow()
+	return c.closeConn()
 }

shared/relay/client/early_msg_buffer.go

@@ -10,7 +10,7 @@ import (
 
 const (
 	earlyMsgTTL      = 5 * time.Second
-	earlyMsgCapacity = 1000
+	earlyMsgCapacity = 10000
 )
 
 // earlyMsgBuffer buffers transport messages that arrive before the corresponding

shared/relay/client/manager_cleanup_test.go

@@ -0,0 +1,50 @@
+package client
+
+import (
+	"context"
+	"testing"
+	"time"
+)
+
+// TestCleanUpUnusedRelays_DoesNotBlockOnRealHangingDial drives a real, hanging foreign
+// relay dial and asserts the foreign store cleanup does not stall behind it.
+func TestCleanUpUnusedRelays_DoesNotBlockOnRealHangingDial(t *testing.T) {
+	serverAddr, accepted := stallingRelayListener(t)
+
+	mCtx, mCancel := context.WithCancel(context.Background())
+	t.Cleanup(mCancel)
+
+	m := NewManager(mCtx, nil, "alice", 1280)
+
+	dialDone := make(chan struct{})
+	go func() {
+		defer close(dialDone)
+		_, _ = m.foreign.OpenConn(mCtx, "peerKey", RelayServer{Addr: serverAddr})
+	}()
+
+	select {
+	case <-accepted:
+	case <-time.After(5 * time.Second):
+		t.Fatal("relay dial did not reach the listener")
+	}
+
+	cleanupDone := make(chan struct{})
+	go func() {
+		defer close(cleanupDone)
+		m.foreign.cleanupUnused()
+	}()
+
+	select {
+	case <-cleanupDone:
+	case <-time.After(2 * time.Second):
+		t.Fatal("cleanupUnused blocked on an in-progress relay dial")
+	}
+
+	// Release the hanging dial so the goroutine can exit cleanly.
+	mCancel()
+	select {
+	case <-dialDone:
+	case <-time.After(5 * time.Second):
+		t.Fatal("foreign OpenConn did not return after context cancellation")
+	}
+}

shared/relay/client/manager_relaystates_test.go

@@ -0,0 +1,96 @@
+package client
+
+import (
+	"context"
+	"net"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+)
+
+// stallingRelayListener accepts TCP connections and holds them open without ever
+// responding, so a relay handshake dialed against it blocks until its context is
+// cancelled. accepted is signalled once per incoming connection so a caller can
+// wait until a dial has actually reached the listener. It returns the
+// "rel://host:port" URL to dial.
+func stallingRelayListener(t *testing.T) (string, <-chan struct{}) {
+	t.Helper()
+
+	ln, err := net.Listen("tcp", "127.0.0.1:0")
+	require.NoError(t, err)
+
+	accepted := make(chan struct{}, 1)
+	var mu sync.Mutex
+	var conns []net.Conn
+	go func() {
+		for {
+			c, err := ln.Accept()
+			if err != nil {
+				return
+			}
+			mu.Lock()
+			conns = append(conns, c)
+			mu.Unlock()
+			select {
+			case accepted <- struct{}{}:
+			default:
+			}
+		}
+	}()
+	t.Cleanup(func() {
+		_ = ln.Close()
+		mu.Lock()
+		for _, c := range conns {
+			_ = c.Close()
+		}
+		mu.Unlock()
+	})
+
+	return "rel://" + ln.Addr().String(), accepted
+}
+
+// TestRelayStates_DoesNotBlockOnRealHangingDial is a regression test for
+// RelayStates() called by a "status -d command" hanging behind an in-progress
+// foreign relay dial.
+func TestRelayStates_DoesNotBlockOnRealHangingDial(t *testing.T) {
+	serverAddr, accepted := stallingRelayListener(t)
+
+	mCtx, mCancel := context.WithCancel(context.Background())
+	t.Cleanup(mCancel)
+
+	m := NewManager(mCtx, nil, "alice", 1280)
+
+	dialDone := make(chan struct{})
+	go func() {
+		defer close(dialDone)
+		_, _ = m.foreign.OpenConn(mCtx, "peerKey", RelayServer{Addr: serverAddr})
+	}()
+
+	select {
+	case <-accepted:
+	case <-time.After(5 * time.Second):
+		t.Fatal("relay dial did not reach the listener")
+	}
+
+	done := make(chan []RelayConnState, 1)
+	go func() {
+		done <- m.RelayStates()
+	}()
+
+	select {
+	case states := <-done:
+		require.Empty(t, states, "a relay still being dialed carries no state and must be omitted")
+	case <-time.After(2 * time.Second):
+		t.Fatal("RelayStates blocked on a foreign relay whose Connect() is in progress")
+	}
+
+	// Release the hanging dial so the goroutine can exit cleanly.
+	mCancel()
+	select {
+	case <-dialDone:
+	case <-time.After(5 * time.Second):
+		t.Fatal("foreign OpenConn did not return after context cancellation")
+	}
+}