Add tests

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

client/internal/conn_mgr.go

@@ -34,8 +34,6 @@ const (
 // - Handling connection establishment based on peer signaling
 //
 // The implementation is not thread-safe; it is protected by engine.syncMsgMux.
-// The only exception is ActivatePeer, which is safe for concurrent use so the
-// DNS warm-up path can call it without contending on the engine mutex.
 type ConnMgr struct {
 	peerStore        *peerstore.Store
 	statusRecorder   *peer.Status
@@ -44,10 +42,6 @@ type ConnMgr struct {
 	rosenpassEnabled bool
 
 	lazyConnMgr *manager.Manager
-	// lazyConnMgrMu guards the lazyConnMgr pointer for readers outside the
-	// engine loop (ActivatePeer). Writers hold it in addition to
-	// engine.syncMsgMux; all other reads stay under engine.syncMsgMux only.
-	lazyConnMgrMu sync.RWMutex
 
 	wg            sync.WaitGroup
 	lazyCtx       context.Context
@@ -244,20 +238,12 @@ func (e *ConnMgr) RemovePeerConn(peerKey string) {
 	conn.Log.Infof("removed peer from lazy conn manager")
 }
 
-// ActivatePeer wakes an idle lazy connection. Unlike the rest of ConnMgr it is
-// safe for concurrent use: the lazy manager pointer is read under lazyConnMgrMu
-// and the manager itself is internally synchronized, so callers outside the
-// engine loop (DNS warm-up) do not need engine.syncMsgMux.
 func (e *ConnMgr) ActivatePeer(ctx context.Context, conn *peer.Conn) {
-	e.lazyConnMgrMu.RLock()
-	lazyConnMgr := e.lazyConnMgr
-	started := lazyConnMgr != nil && e.lazyCtxCancel != nil
-	e.lazyConnMgrMu.RUnlock()
-	if !started {
+	if !e.isStartedWithLazyMgr() {
 		return
 	}
 
-	if found := lazyConnMgr.ActivatePeer(conn.GetKey()); found {
+	if found := e.lazyConnMgr.ActivatePeer(conn.GetKey()); found {
 		if err := conn.Open(ctx); err != nil {
 			conn.Log.Errorf("failed to open connection: %v", err)
 		}
@@ -282,21 +268,16 @@ func (e *ConnMgr) Close() {
 
 	e.lazyCtxCancel()
 	e.wg.Wait()
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = nil
-	e.lazyConnMgrMu.Unlock()
 }
 
 func (e *ConnMgr) initLazyManager(engineCtx context.Context) {
 	cfg := manager.Config{
 		InactivityThreshold: inactivityThresholdEnv(),
 	}
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = manager.NewManager(cfg, engineCtx, e.peerStore, e.iface)
+
 	e.lazyCtx, e.lazyCtxCancel = context.WithCancel(engineCtx)
-	e.lazyConnMgrMu.Unlock()
 
 	e.wg.Add(1)
 	go func() {
@@ -335,10 +316,7 @@ func (e *ConnMgr) closeManager(ctx context.Context) {
 
 	e.lazyCtxCancel()
 	e.wg.Wait()
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = nil
-	e.lazyConnMgrMu.Unlock()
 
 	for _, peerID := range e.peerStore.PeersPubKey() {
 		e.peerStore.PeerConnOpen(ctx, peerID)

client/internal/conn_mgr_test.go

@@ -1,21 +1,10 @@
 package internal
 
 import (
-	"context"
-	"net"
-	"net/netip"
 	"os"
-	"sync"
 	"testing"
-	"time"
 
-	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
-
-	"github.com/netbirdio/netbird/client/iface/wgaddr"
 	"github.com/netbirdio/netbird/client/internal/lazyconn"
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-	"github.com/netbirdio/netbird/monotime"
 )
 
 func TestResolveLazyForce(t *testing.T) {
@@ -49,58 +38,3 @@ func TestResolveLazyForce(t *testing.T) {
 		})
 	}
 }
-
-type mockLazyWGIface struct{}
-
-func (mockLazyWGIface) RemovePeer(string) error { return nil }
-func (mockLazyWGIface) UpdatePeer(string, []netip.Prefix, time.Duration, *net.UDPAddr, *wgtypes.Key) error {
-	return nil
-}
-func (mockLazyWGIface) IsUserspaceBind() bool                    { return false }
-func (mockLazyWGIface) Address() wgaddr.Address                  { return wgaddr.Address{} }
-func (mockLazyWGIface) LastActivities() map[string]monotime.Time { return nil }
-func (mockLazyWGIface) MTU() uint16                              { return 1280 }
-
-// TestConnMgr_ActivatePeerConcurrentWithLifecycle exercises ActivatePeer from
-// non-engine goroutines (the DNS warm-up path) racing the manager lifecycle,
-// which stays on the engine loop. Run with -race: it fails if ActivatePeer
-// still requires engine.syncMsgMux for safety.
-func TestConnMgr_ActivatePeerConcurrentWithLifecycle(t *testing.T) {
-	t.Setenv(lazyconn.EnvLazyConn, "on")
-
-	status := peer.NewRecorder("https://mgm")
-	store := peerstore.NewConnStore()
-	connMgr := NewConnMgr(&EngineConfig{}, status, store, mockLazyWGIface{})
-
-	conn := newTestPeerConn(t, "peerA")
-
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-	connMgr.Start(ctx)
-
-	done := make(chan struct{})
-	var wg sync.WaitGroup
-	for range 4 {
-		wg.Add(1)
-		go func() {
-			defer wg.Done()
-			for {
-				select {
-				case <-done:
-					return
-				default:
-					connMgr.ActivatePeer(ctx, conn)
-				}
-			}
-		}()
-	}
-
-	// Let the activators spin against the started manager, then tear it down
-	// underneath them and let them spin against the stopped manager.
-	time.Sleep(100 * time.Millisecond)
-	connMgr.Close()
-	time.Sleep(50 * time.Millisecond)
-
-	close(done)
-	wg.Wait()
-}

client/internal/dns/local/local.go

@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"net"
 	"net/netip"
-	"os"
 	"slices"
 	"strings"
 	"sync"
@@ -37,43 +36,7 @@ type resolver interface {
 // record is left alone (it points at something outside our mesh, e.g.
 // a non-peer upstream).
 type PeerConnectivity interface {
-	IsConnectedByIP(ip netip.Addr) (known, connected bool)
-}
-
-// PeerActivator wakes lazy-connection peers on demand. The local resolver calls
-// it with the tunnel IPs an answer points at, so a peer that is idle (lazily
-// disconnected) starts connecting at DNS-resolution time rather than racing the
-// client's first request packet. nil disables warm-up.
-type PeerActivator interface {
-	// ActivatePeersByIP triggers wake-up for the peer(s) owning addrs and blocks
-	// until one is connected or ctx (a short per-query budget) expires. It is a
-	// fast no-op for unknown or already-connected addresses.
-	ActivatePeersByIP(ctx context.Context, addrs []netip.Addr)
-}
-
-const (
-	defaultLazyWarmupTimeout = 2 * time.Second
-	envLazyWarmupTimeout     = "NB_DNS_LAZY_WARMUP_TIMEOUT"
-)
-
-// lazyWarmupTimeoutFromEnv returns the per-query budget for waking a
-// lazy-connection peer a DNS answer points at. Tunable via
-// NB_DNS_LAZY_WARMUP_TIMEOUT (a Go duration). Parsed once at construction time.
-func lazyWarmupTimeoutFromEnv() time.Duration {
-	v := os.Getenv(envLazyWarmupTimeout)
-	if v == "" {
-		return defaultLazyWarmupTimeout
-	}
-	d, err := time.ParseDuration(v)
-	if err != nil {
-		log.Warnf("invalid %s value %q, using default %s: %v", envLazyWarmupTimeout, v, defaultLazyWarmupTimeout, err)
-		return defaultLazyWarmupTimeout
-	}
-	if d <= 0 {
-		log.Warnf("non-positive %s value %q, using default %s", envLazyWarmupTimeout, v, defaultLazyWarmupTimeout)
-		return defaultLazyWarmupTimeout
-	}
-	return d
+	IsConnectedByIP(ip string) (known, connected bool)
 }
 
 type Resolver struct {
@@ -88,12 +51,6 @@ type Resolver struct {
 	// filter and preserves the legacy "return whatever is registered"
 	// behaviour for callers that never wire a status source.
 	peerConn PeerConnectivity
-	// peerActivator, when non-nil, is called at resolution time to warm the
-	// lazy connection to the peer(s) an answer points at. nil disables warm-up.
-	peerActivator PeerActivator
-	// warmupTimeout is the per-query budget for the lazy-connection warm-up
-	// wait, resolved from the environment once at construction time.
-	warmupTimeout time.Duration
 
 	ctx    context.Context
 	cancel context.CancelFunc
@@ -102,12 +59,11 @@ type Resolver struct {
 func NewResolver() *Resolver {
 	ctx, cancel := context.WithCancel(context.Background())
 	return &Resolver{
-		records:       make(map[dns.Question][]dns.RR),
-		domains:       make(map[domain.Domain]struct{}),
-		zones:         make(map[domain.Domain]bool),
-		warmupTimeout: lazyWarmupTimeoutFromEnv(),
-		ctx:           ctx,
-		cancel:        cancel,
+		records: make(map[dns.Question][]dns.RR),
+		domains: make(map[domain.Domain]struct{}),
+		zones:   make(map[domain.Domain]bool),
+		ctx:     ctx,
+		cancel:  cancel,
 	}
 }
 
@@ -120,14 +76,6 @@ func (d *Resolver) SetPeerConnectivity(p PeerConnectivity) {
 	d.peerConn = p
 }
 
-// SetPeerActivator wires the DNS-time lazy-connection warm-up. Pass nil to
-// disable. Safe to call multiple times; the latest value wins.
-func (d *Resolver) SetPeerActivator(a PeerActivator) {
-	d.mu.Lock()
-	defer d.mu.Unlock()
-	d.peerActivator = a
-}
-
 func (d *Resolver) MatchSubdomains() bool {
 	return true
 }
@@ -174,9 +122,6 @@ func (d *Resolver) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 	replyMessage.RecursionAvailable = true
 
 	result := d.lookupRecords(logger, question)
-	// Warm before filtering: activation flips a lazily-idle target to connected,
-	// which then lets it survive the disconnected-peer filter below.
-	d.warmLazyPeers(question, result.records)
 	result.records = d.filterDisconnectedPeerAnswers(logger, question, result.records)
 	replyMessage.Authoritative = !result.hasExternalData
 	replyMessage.Answer = result.records
@@ -550,8 +495,8 @@ func (d *Resolver) filterDisconnectedPeerAnswers(logger *log.Entry, question dns
 	kept := make([]dns.RR, 0, len(records))
 	var dropped int
 	for _, rr := range records {
-		ip, ok := extractRecordAddr(rr)
-		if !ok {
+		ip := extractRecordIP(rr)
+		if ip == "" {
 			kept = append(kept, rr)
 			continue
 		}
@@ -573,54 +518,22 @@ func (d *Resolver) filterDisconnectedPeerAnswers(logger *log.Entry, question dns
 	return kept
 }
 
-// warmLazyPeers triggers lazy-connection wake-up for the peers a resolved
-// answer points at and waits briefly for one to connect, so the caller's first
-// request doesn't race the connection establishment. Warm-up is scoped to
-// match-only (non-authoritative) zones — the synthesized private-service zones
-// and user-created zones whose records point at specific peers. The account's
-// peer zone is authoritative, so plain peer-name lookups never trigger warm-up;
-// otherwise resolving any peer's name would wake its idle connection, defeating
-// laziness mesh-wide. No-op when no activator is wired (lazy connections
-// disabled) or the answer carries no peer IPs.
-func (d *Resolver) warmLazyPeers(question dns.Question, records []dns.RR) {
-	d.mu.RLock()
-	activator := d.peerActivator
-	var nonAuth, found bool
-	if activator != nil {
-		nonAuth, found = d.findZone(question.Name)
-	}
-	d.mu.RUnlock()
-	if activator == nil || !found || !nonAuth {
-		return
-	}
-
-	var addrs []netip.Addr
-	for _, rr := range records {
-		if addr, ok := extractRecordAddr(rr); ok {
-			addrs = append(addrs, addr)
-		}
-	}
-	if len(addrs) == 0 {
-		return
-	}
-
-	ctx, cancel := context.WithTimeout(d.ctx, d.warmupTimeout)
-	defer cancel()
-	activator.ActivatePeersByIP(ctx, addrs)
-}
-
-// extractRecordAddr returns the IP address carried by an A or AAAA record.
-// ok is false for any other record type or a record with no address.
-func extractRecordAddr(rr dns.RR) (netip.Addr, bool) {
+// extractRecordIP returns the dotted-decimal / colon-hex IP carried by
+// an A or AAAA record, or "" for any other record type.
+func extractRecordIP(rr dns.RR) string {
 	switch r := rr.(type) {
 	case *dns.A:
-		addr, ok := netip.AddrFromSlice(r.A)
-		return addr.Unmap(), ok
+		if r.A == nil {
+			return ""
+		}
+		return r.A.String()
 	case *dns.AAAA:
-		addr, ok := netip.AddrFromSlice(r.AAAA)
-		return addr.Unmap(), ok
+		if r.AAAA == nil {
+			return ""
+		}
+		return r.AAAA.String()
 	}
-	return netip.Addr{}, false
+	return ""
 }
 
 // Update replaces all zones and their records

client/internal/dns/local/local_test.go

@@ -37,8 +37,8 @@ type mockPeerConnectivity struct {
 	byIP map[string]struct{ known, connected bool }
 }
 
-func (m mockPeerConnectivity) IsConnectedByIP(ip netip.Addr) (known, connected bool) {
-	v, ok := m.byIP[ip.String()]
+func (m mockPeerConnectivity) IsConnectedByIP(ip string) (known, connected bool) {
+	v, ok := m.byIP[ip]
 	if !ok {
 		return false, false
 	}

client/internal/dns/local/warmup_test.go

@@ -1,172 +0,0 @@
-package local
-
-import (
-	"context"
-	"net"
-	"net/netip"
-	"sync"
-	"testing"
-	"time"
-
-	"github.com/miekg/dns"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-
-	"github.com/netbirdio/netbird/client/internal/dns/test"
-	nbdns "github.com/netbirdio/netbird/dns"
-)
-
-// recordingActivator records the addresses it was asked to warm and returns
-// immediately, so ServeDNS is not blocked by the test.
-type recordingActivator struct {
-	mu     sync.Mutex
-	called bool
-	addrs  []netip.Addr
-}
-
-func (r *recordingActivator) ActivatePeersByIP(_ context.Context, addrs []netip.Addr) {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	r.called = true
-	r.addrs = append(r.addrs, addrs...)
-}
-
-func serveA(t *testing.T, resolver *Resolver, name string) *dns.Msg {
-	t.Helper()
-	var resp *dns.Msg
-	w := &test.MockResponseWriter{WriteMsgFunc: func(m *dns.Msg) error { resp = m; return nil }}
-	resolver.ServeDNS(w, new(dns.Msg).SetQuestion(name, dns.TypeA))
-	return resp
-}
-
-// serviceZone registers rec in a match-only (non-authoritative) zone, the shape
-// the synthesized private-service zones arrive in.
-func serviceZone(t *testing.T, resolver *Resolver, zone string, records ...nbdns.SimpleRecord) {
-	t.Helper()
-	resolver.Update([]nbdns.CustomZone{{
-		Domain:           zone,
-		Records:          records,
-		NonAuthoritative: true,
-	}})
-}
-
-func TestLocalResolver_WarmsLazyPeerOnResolve(t *testing.T) {
-	rec := nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"}
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud", rec)
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	resp := serveA(t, resolver, rec.Name)
-	require.NotNil(t, resp, "resolver must answer")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A record")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.True(t, act.called, "activator must be invoked for a service-zone A answer")
-	assert.Contains(t, act.addrs, netip.MustParseAddr("100.64.0.7"), "activator must receive the answer's peer IP")
-}
-
-func TestLocalResolver_NoActivatorNoWarmup(t *testing.T) {
-	// With no activator wired the resolver behaves exactly as before.
-	rec := nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"}
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud", rec)
-
-	resp := serveA(t, resolver, rec.Name)
-	require.NotNil(t, resp, "resolver must still answer without an activator")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A record")
-}
-
-func TestLocalResolver_NoWarmupForMissingRecord(t *testing.T) {
-	// A query that resolves to nothing must not invoke the activator (no IPs).
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud",
-		nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"})
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	serveA(t, resolver, "absent.proxy.netbird.cloud.")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.False(t, act.called, "activator must not be invoked when there is no answer")
-}
-
-func TestLocalResolver_NoWarmupInAuthoritativeZone(t *testing.T) {
-	// The account's peer zone is authoritative; resolving a peer's name there
-	// must not wake its lazy connection — warm-up is scoped to match-only
-	// (non-authoritative) zones such as the synthesized private-service zones.
-	rec := nbdns.SimpleRecord{Name: "peer.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.9"}
-	resolver := NewResolver()
-	resolver.Update([]nbdns.CustomZone{{
-		Domain:  "netbird.cloud",
-		Records: []nbdns.SimpleRecord{rec},
-	}})
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	resp := serveA(t, resolver, rec.Name)
-	require.NotNil(t, resp, "resolver must answer")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A record")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.False(t, act.called, "activator must not be invoked for authoritative-zone answers")
-}
-
-func TestLazyWarmupTimeoutFromEnv(t *testing.T) {
-	tests := []struct {
-		name   string
-		value  string
-		envSet bool
-		want   time.Duration
-	}{
-		{name: "unset uses default", want: defaultLazyWarmupTimeout},
-		{name: "valid overrides", value: "5s", envSet: true, want: 5 * time.Second},
-		{name: "invalid falls back", value: "not-a-duration", envSet: true, want: defaultLazyWarmupTimeout},
-		{name: "negative falls back", value: "-1s", envSet: true, want: defaultLazyWarmupTimeout},
-		{name: "zero falls back", value: "0s", envSet: true, want: defaultLazyWarmupTimeout},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			if tt.envSet {
-				t.Setenv(envLazyWarmupTimeout, tt.value)
-			}
-			assert.Equal(t, tt.want, lazyWarmupTimeoutFromEnv())
-			assert.Equal(t, tt.want, NewResolver().warmupTimeout, "constructor must resolve the timeout once")
-		})
-	}
-}
-
-func TestExtractRecordAddr(t *testing.T) {
-	t.Run("A record yields unmapped v4", func(t *testing.T) {
-		// net.ParseIP returns the 16-byte v4-in-v6 form, the same shape
-		// miekg/dns stores after parsing an A record; the extracted address
-		// must compare equal to a plain v4 netip.Addr.
-		addr, ok := extractRecordAddr(&dns.A{A: net.ParseIP("100.64.0.7")})
-		require.True(t, ok)
-		assert.True(t, addr.Is4())
-		assert.Equal(t, netip.MustParseAddr("100.64.0.7"), addr)
-	})
-
-	t.Run("AAAA record yields v6", func(t *testing.T) {
-		addr, ok := extractRecordAddr(&dns.AAAA{AAAA: net.ParseIP("fd00::1")})
-		require.True(t, ok)
-		assert.Equal(t, netip.MustParseAddr("fd00::1"), addr)
-	})
-
-	t.Run("A record without address", func(t *testing.T) {
-		_, ok := extractRecordAddr(&dns.A{})
-		assert.False(t, ok)
-	})
-
-	t.Run("non-address record", func(t *testing.T) {
-		_, ok := extractRecordAddr(&dns.CNAME{Target: "target.netbird.cloud."})
-		assert.False(t, ok)
-	})
-}

client/internal/dns/mock_server.go

@@ -8,7 +8,6 @@ import (
 	"github.com/miekg/dns"
 
 	dnsconfig "github.com/netbirdio/netbird/client/internal/dns/config"
-	"github.com/netbirdio/netbird/client/internal/dns/local"
 	nbdns "github.com/netbirdio/netbird/dns"
 	"github.com/netbirdio/netbird/route"
 	"github.com/netbirdio/netbird/shared/management/domain"
@@ -93,11 +92,6 @@ func (m *MockServer) SetFirewall(Firewall) {
 	// Mock implementation - no-op
 }
 
-// SetPeerActivator mock implementation of SetPeerActivator from Server interface
-func (m *MockServer) SetPeerActivator(local.PeerActivator) {
-	// Mock implementation - no-op
-}
-
 // BeginBatch mock implementation of BeginBatch from Server interface
 func (m *MockServer) BeginBatch() {
 	// Mock implementation - no-op

client/internal/dns/server.go

@@ -82,7 +82,6 @@ type Server interface {
 	PopulateManagementDomain(mgmtURL *url.URL) error
 	SetRouteSources(selected, active func() route.HAMap)
 	SetFirewall(Firewall)
-	SetPeerActivator(local.PeerActivator)
 }
 
 type nsGroupsByDomain struct {
@@ -492,13 +491,6 @@ func (s *DefaultServer) SetFirewall(fw Firewall) {
 	}
 }
 
-// SetPeerActivator wires the DNS-time lazy-connection warm-up on the local
-// resolver. Injected after the connection manager exists (it does not at
-// DNS-server construction time). Pass nil to disable.
-func (s *DefaultServer) SetPeerActivator(a local.PeerActivator) {
-	s.localResolver.SetPeerActivator(a)
-}
-
 // Stop stops the server
 func (s *DefaultServer) Stop() {
 	s.ctxCancel()
@@ -1443,11 +1435,11 @@ type localPeerConnectivity struct {
 
 // IsConnectedByIP looks the IP up in the peerstore and surfaces both
 // the known and connected bits. Used by Resolver.filterDisconnectedPeerAnswers.
-func (l localPeerConnectivity) IsConnectedByIP(ip netip.Addr) (known, connected bool) {
+func (l localPeerConnectivity) IsConnectedByIP(ip string) (known, connected bool) {
 	if l.status == nil {
 		return false, false
 	}
-	state, ok := l.status.PeerStateByIP(ip.String())
+	state, ok := l.status.PeerStateByIP(ip)
 	if !ok {
 		return false, false
 	}

client/internal/dns_peer_activator.go

@@ -1,76 +0,0 @@
-package internal
-
-import (
-	"context"
-	"net/netip"
-	"time"
-
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-)
-
-const dnsActivationPollInterval = 50 * time.Millisecond
-
-// dnsPeerActivator wakes lazy-connection peers from the DNS resolution path. It
-// implements dns/local.PeerActivator. DNS queries run on their own goroutines,
-// so it only touches state that is safe for concurrent use — ConnMgr.ActivatePeer,
-// peerstore.Store and peer.Status — and never takes the engine's syncMsgMux,
-// keeping DNS resolution from contending with network-map processing.
-type dnsPeerActivator struct {
-	connMgr   *ConnMgr
-	peerStore *peerstore.Store
-	status    *peer.Status
-	// ctx is the engine's long-lived context. The connection dial is tied to it
-	// (not the per-query DNS wait budget) so a handshake that outlasts the wait
-	// still completes in the background rather than being cancelled at the deadline.
-	ctx context.Context
-}
-
-// ActivatePeersByIP triggers wake-up for the peer(s) owning addrs and waits
-// until one is connected or ctx (the per-query DNS wait budget) expires.
-// Activation itself is tied to the engine's long-lived context so the dial
-// survives a wait that times out. Unknown or already-connected addresses are
-// skipped, so the steady-state (warm) path adds no latency.
-func (a *dnsPeerActivator) ActivatePeersByIP(ctx context.Context, addrs []netip.Addr) {
-	if a == nil || a.connMgr == nil {
-		return
-	}
-
-	var pending []string
-	for _, addr := range addrs {
-		ip := addr.String()
-		st, ok := a.status.PeerStateByIP(ip)
-		if !ok || st.ConnStatus == peer.StatusConnected {
-			continue
-		}
-		conn, ok := a.peerStore.PeerConn(st.PubKey)
-		if !ok {
-			continue
-		}
-		a.connMgr.ActivatePeer(a.ctx, conn)
-		pending = append(pending, ip)
-	}
-
-	if len(pending) == 0 {
-		return
-	}
-	a.waitConnected(ctx, pending)
-}
-
-// waitConnected blocks until any of ips reports a connected peer or ctx expires.
-func (a *dnsPeerActivator) waitConnected(ctx context.Context, ips []string) {
-	ticker := time.NewTicker(dnsActivationPollInterval)
-	defer ticker.Stop()
-	for {
-		for _, ip := range ips {
-			if st, ok := a.status.PeerStateByIP(ip); ok && st.ConnStatus == peer.StatusConnected {
-				return
-			}
-		}
-		select {
-		case <-ctx.Done():
-			return
-		case <-ticker.C:
-		}
-	}
-}

client/internal/dns_peer_activator_test.go

@@ -1,129 +0,0 @@
-package internal
-
-import (
-	"context"
-	"net/netip"
-	"testing"
-	"time"
-
-	"github.com/stretchr/testify/require"
-
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-)
-
-func newTestPeerConn(t *testing.T, key string) *peer.Conn {
-	t.Helper()
-	conn, err := peer.NewConn(peer.ConnConfig{
-		Key:      key,
-		LocalKey: "local",
-		WgConfig: peer.WgConfig{
-			AllowedIps: []netip.Prefix{netip.MustParsePrefix("100.64.0.1/32")},
-		},
-	}, peer.ServiceDependencies{})
-	require.NoError(t, err)
-	return conn
-}
-
-func newTestDNSPeerActivator(t *testing.T) (*dnsPeerActivator, *peer.Status, *peerstore.Store) {
-	t.Helper()
-	status := peer.NewRecorder("https://mgm")
-	store := peerstore.NewConnStore()
-	// ConnMgr without Start: the lazy manager is nil, so ActivatePeer is a
-	// no-op — these tests exercise the activator's skip/wait logic.
-	connMgr := NewConnMgr(&EngineConfig{}, status, store, nil)
-	return &dnsPeerActivator{
-		connMgr:   connMgr,
-		peerStore: store,
-		status:    status,
-		ctx:       context.Background(),
-	}, status, store
-}
-
-func TestDNSPeerActivator_NilSafe(t *testing.T) {
-	var a *dnsPeerActivator
-	a.ActivatePeersByIP(context.Background(), []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-}
-
-// TestDNSPeerActivator_SkipsUnknownAndConnectedPeers verifies the steady-state
-// (warm) path adds no latency: already-connected and unknown addresses never
-// enter the wait loop.
-func TestDNSPeerActivator_SkipsUnknownAndConnectedPeers(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", "fd00::1"))
-	require.NoError(t, status.UpdatePeerState(peer.State{PubKey: "peerA", ConnStatus: peer.StatusConnected}))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{
-		netip.MustParseAddr("100.64.0.1"),  // known, connected -> skipped
-		netip.MustParseAddr("fd00::1"),     // known via IPv6, connected -> skipped
-		netip.MustParseAddr("100.64.0.99"), // unknown -> skipped
-	})
-	require.Less(t, time.Since(start), time.Second, "no pending peer must mean no wait")
-}
-
-// TestDNSPeerActivator_WaitsForPendingPeerToConnect verifies the wait loop
-// returns as soon as a pending peer reports connected, well before the
-// per-query budget expires.
-func TestDNSPeerActivator_WaitsForPendingPeerToConnect(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	go func() {
-		time.Sleep(150 * time.Millisecond)
-		_ = status.UpdatePeerState(peer.State{PubKey: "peerA", ConnStatus: peer.StatusConnected})
-	}()
-
-	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	elapsed := time.Since(start)
-
-	require.GreaterOrEqual(t, elapsed, 100*time.Millisecond, "must wait for the pending peer")
-	require.Less(t, elapsed, 5*time.Second, "must return on connect, not at the deadline")
-}
-
-// TestDNSPeerActivator_ReturnsAtBudgetWhenPeerStaysIdle verifies a peer that
-// never connects releases the DNS response at the per-query budget instead of
-// blocking it indefinitely.
-func TestDNSPeerActivator_ReturnsAtBudgetWhenPeerStaysIdle(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 300*time.Millisecond)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	elapsed := time.Since(start)
-
-	require.GreaterOrEqual(t, elapsed, 250*time.Millisecond, "must wait out the budget for a pending peer")
-	require.Less(t, elapsed, 5*time.Second, "must not block past the budget")
-}
-
-// TestDNSPeerActivator_NoWaitWithoutPeerConn verifies a known-but-idle peer
-// with no connection object in the store is not waited on: there is nothing to
-// activate, so waiting could only ever time out.
-func TestDNSPeerActivator_NoWaitWithoutPeerConn(t *testing.T) {
-	a, status, _ := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	require.Less(t, time.Since(start), time.Second, "peer without a conn must not be waited on")
-}

client/internal/engine.go

@@ -654,16 +654,6 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 	e.connMgr = NewConnMgr(e.config, e.statusRecorder, e.peerStore, wgIface)
 	e.connMgr.Start(e.ctx)
 
-	// Wire DNS-time lazy-connection warm-up now that the connection manager
-	// exists (it does not at DNS-server construction time). A DNS answer that
-	// points at an idle peer then wakes it before the client's first request.
-	e.dnsServer.SetPeerActivator(&dnsPeerActivator{
-		connMgr:   e.connMgr,
-		peerStore: e.peerStore,
-		status:    e.statusRecorder,
-		ctx:       e.ctx,
-	})
-
 	e.srWatcher = guard.NewSRWatcher(e.signal, e.relayManager, e.mobileDep.IFaceDiscover, iceCfg)
 	e.srWatcher.Start(peer.IsForceRelayed())
 

client/internal/peer/worker_relay.go

@@ -54,19 +54,15 @@ func (w *WorkerRelay) OnNewOffer(remoteOfferAnswer *OfferAnswer) {
 	w.relaySupportedOnRemotePeer.Store(true)
 
 	// the relayManager will return with error in case if the connection has lost with relay server
-	currentRelayAddress, _, err := w.relayManager.RelayInstanceAddress()
+	_, _, err := w.relayManager.RelayInstanceAddress()
 	if err != nil {
 		w.log.Errorf("failed to handle new offer: %s", err)
 		return
 	}
 
-	srv := w.preferredRelayServer(currentRelayAddress, remoteOfferAnswer.RelaySrvAddress)
-	var serverIP netip.Addr
-	if srv == remoteOfferAnswer.RelaySrvAddress {
-		serverIP = remoteOfferAnswer.RelaySrvIP
-	}
-
-	relayedConn, err := w.relayManager.OpenConn(w.peerCtx, srv, w.config.Key, serverIP)
+	preferForeign := !w.isController
+	remoteRelayServer := relayClient.RelayServer{Addr: remoteOfferAnswer.RelaySrvAddress, IP: remoteOfferAnswer.RelaySrvIP}
+	relayedConn, err := w.relayManager.OpenConn(w.peerCtx, remoteRelayServer, w.config.Key, preferForeign)
 	if err != nil {
 		if errors.Is(err, relayClient.ErrConnAlreadyExists) {
 			w.log.Debugf("handled offer by reusing existing relay connection")
@@ -80,14 +76,13 @@ func (w *WorkerRelay) OnNewOffer(remoteOfferAnswer *OfferAnswer) {
 	w.relayedConn = relayedConn
 	w.relayLock.Unlock()
 
-	err = w.relayManager.AddCloseListener(srv, w.onRelayClientDisconnected)
-	if err != nil {
-		log.Errorf("failed to add close listener: %s", err)
+	if err := w.relayManager.AddCloseListener(relayedConn.RemoteAddr().String(), w.onRelayClientDisconnected); err != nil {
+		w.log.Errorf("failed to add close listener: %s", err)
 		_ = relayedConn.Close()
 		return
 	}
 
-	w.log.Debugf("peer conn opened via Relay: %s", srv)
+	w.log.Debugf("peer conn opened via Relay: %s", relayedConn.RemoteAddr())
 	go w.conn.onRelayConnectionIsReady(RelayConnInfo{
 		relayedConn:     relayedConn,
 		rosenpassPubKey: remoteOfferAnswer.RosenpassPubKey,
@@ -126,13 +121,6 @@ func (w *WorkerRelay) isRelaySupported(answer *OfferAnswer) bool {
 	return answer.RelaySrvAddress != ""
 }
 
-func (w *WorkerRelay) preferredRelayServer(myRelayAddress, remoteRelayAddress string) string {
-	if w.isController {
-		return myRelayAddress
-	}
-	return remoteRelayAddress
-}
-
 func (w *WorkerRelay) onRelayClientDisconnected() {
 	go w.conn.onRelayDisconnected()
 }

client/internal/routemanager/dynamic/route.go

@@ -185,7 +185,7 @@ func (r *Route) startResolver(ctx context.Context) {
 }
 
 func (r *Route) update(ctx context.Context) error {
-	resolved, err := r.resolveDomains(ctx)
+	resolved, err := r.resolveDomains()
 	if err != nil {
 		if len(resolved) == 0 {
 			return fmt.Errorf("resolve domains: %w", err)
@@ -199,9 +199,9 @@ func (r *Route) update(ctx context.Context) error {
 	return nil
 }
 
-func (r *Route) resolveDomains(ctx context.Context) (domainMap, error) {
+func (r *Route) resolveDomains() (domainMap, error) {
 	results := make(chan resolveResult)
-	go r.resolve(ctx, results)
+	go r.resolve(results)
 
 	resolved := domainMap{}
 	var merr *multierror.Error
@@ -217,7 +217,7 @@ func (r *Route) resolveDomains(ctx context.Context) (domainMap, error) {
 	return resolved, nberrors.FormatErrorOrNil(merr)
 }
 
-func (r *Route) resolve(ctx context.Context, results chan resolveResult) {
+func (r *Route) resolve(results chan resolveResult) {
 	var wg sync.WaitGroup
 
 	for _, d := range r.route.Domains {
@@ -225,10 +225,10 @@ func (r *Route) resolve(ctx context.Context, results chan resolveResult) {
 		go func(domain domain.Domain) {
 			defer wg.Done()
 
-			ips, err := r.getIPsFromResolver(ctx, domain)
+			ips, err := r.getIPsFromResolver(domain)
 			if err != nil {
 				log.Tracef("Failed to resolve domain %s with private resolver: %v", domain.SafeString(), err)
-				ips, err = lookupHostIPs(ctx, domain)
+				ips, err = net.LookupIP(domain.PunycodeString())
 				if err != nil {
 					results <- resolveResult{domain: domain, err: fmt.Errorf("resolve d %s: %w", domain.SafeString(), err)}
 					return
@@ -364,20 +364,6 @@ func determinePrefixChanges(oldPrefixes, newPrefixes []netip.Prefix) (toAdd, toR
 	return
 }
 
-// lookupHostIPs resolves d via the system resolver, honoring ctx cancellation.
-func lookupHostIPs(ctx context.Context, d domain.Domain) ([]net.IP, error) {
-	addrs, err := net.DefaultResolver.LookupIPAddr(ctx, d.PunycodeString())
-	if err != nil {
-		return nil, err
-	}
-
-	ips := make([]net.IP, 0, len(addrs))
-	for _, addr := range addrs {
-		ips = append(ips, addr.IP)
-	}
-	return ips, nil
-}
-
 func combinePrefixes(oldPrefixes, removedPrefixes, addedPrefixes []netip.Prefix) []netip.Prefix {
 	prefixSet := make(map[netip.Prefix]struct{})
 	for _, prefix := range oldPrefixes {

client/internal/routemanager/dynamic/route_generic.go

@@ -3,12 +3,11 @@
 package dynamic
 
 import (
-	"context"
 	"net"
 
 	"github.com/netbirdio/netbird/shared/management/domain"
 )
 
-func (r *Route) getIPsFromResolver(ctx context.Context, domain domain.Domain) ([]net.IP, error) {
-	return lookupHostIPs(ctx, domain)
+func (r *Route) getIPsFromResolver(domain domain.Domain) ([]net.IP, error) {
+	return net.LookupIP(domain.PunycodeString())
 }

client/internal/routemanager/dynamic/route_ios.go

@@ -3,7 +3,6 @@
 package dynamic
 
 import (
-	"context"
 	"fmt"
 	"net"
 	"time"
@@ -17,7 +16,7 @@ import (
 
 const dialTimeout = 10 * time.Second
 
-func (r *Route) getIPsFromResolver(ctx context.Context, domain domain.Domain) ([]net.IP, error) {
+func (r *Route) getIPsFromResolver(domain domain.Domain) ([]net.IP, error) {
 	privateClient, err := nbdns.GetClientPrivate(r.wgInterface, r.resolverAddr.Addr(), dialTimeout)
 	if err != nil {
 		return nil, fmt.Errorf("error while creating private client: %s", err)
@@ -33,7 +32,7 @@ func (r *Route) getIPsFromResolver(ctx context.Context, domain domain.Domain) ([
 		msg := new(dns.Msg)
 		msg.SetQuestion(fqdn, qtype)
 
-		response, _, err := nbdns.ExchangeWithFallback(ctx, privateClient, msg, r.resolverAddr.String())
+		response, _, err := nbdns.ExchangeWithFallback(nil, privateClient, msg, r.resolverAddr.String())
 		if err != nil {
 			if queryErr == nil {
 				queryErr = fmt.Errorf("DNS query for %s (type %d) after %s: %w", domain.SafeString(), qtype, time.Since(startTime), err)

client/server/server.go

@@ -1081,10 +1081,7 @@ func (s *Server) Down(ctx context.Context, _ *proto.DownRequest) (*proto.DownRes
 
 	if err := s.cleanupConnection(); err != nil {
 		s.mutex.Unlock()
-		if errors.Is(err, ErrServiceNotUp) {
-			log.Debugf("Down called while service not up: %v", err)
-			return nil, err
-		}
+		// todo review to update the status in case any type of error
 		log.Errorf("failed to shut down properly: %v", err)
 		return nil, err
 	}
@@ -1157,7 +1154,7 @@ func (s *Server) cleanupConnection() error {
 	// making the run loop the sole owner of engine shutdown.
 	if engine != nil {
 		if err := engine.Stop(); err != nil {
-			log.Errorf("failed to stop engine during cleanup: %v", err)
+			return err
 		}
 	}
 

client/ui/frontend/src/components/empty-state/DaemonOutdatedOverlay.tsx

@@ -1,13 +1,10 @@
-import { useEffect, useState } from "react";
 import { useTranslation } from "react-i18next";
 import { AlertTriangleIcon, DownloadIcon } from "lucide-react";
 import { Browser } from "@wailsio/runtime";
-import { Version } from "@bindings/services";
 import { Button } from "@/components/buttons/Button";
 import { useStatus } from "@/contexts/StatusContext.tsx";
 
 const RELEASES_URL = "https://github.com/netbirdio/netbird/releases/latest";
-const RC_RELEASES_URL = "https://pkgs.netbird.io/releases/rc";
 
 function openUrl(url: string) {
     Browser.OpenURL(url).catch(() => globalThis.open(url, "_blank"));
@@ -15,26 +12,7 @@ function openUrl(url: string) {
 
 export const DaemonOutdatedOverlay = () => {
     const { t } = useTranslation();
-    const { status, isDaemonOutdated } = useStatus();
-
-    const [guiVersion, setGuiVersion] = useState<string>("-");
-    const clientVersion = status?.daemonVersion ?? "—";
-
-    const isRc = /-rc/i.test(guiVersion) || /-rc/i.test(clientVersion);
-    const downloadUrl = isRc ? RC_RELEASES_URL : RELEASES_URL;
-
-    useEffect(() => {
-        if (!isDaemonOutdated) return;
-        let cancelled = false;
-        Version.GUI()
-            .then((v) => {
-                if (!cancelled) setGuiVersion(v);
-            })
-            .catch((err) => console.error("[DaemonOutdatedOverlay] GUI version error", err));
-        return () => {
-            cancelled = true;
-        };
-    }, [isDaemonOutdated]);
+    const { isDaemonOutdated } = useStatus();
 
     if (!isDaemonOutdated) return null;
 
@@ -60,37 +38,10 @@ export const DaemonOutdatedOverlay = () => {
                     <p className={"text-sm text-nb-gray-300"}>{t("daemon.outdated.description")}</p>
                 </div>
 
-                <div className={"flex flex-col items-center gap-0.5 text-center"}>
-                    <p className={"text-sm font-semibold text-nb-gray-100"}>
-                        {clientVersion === "development" ? (
-                            <span>
-                                {t("settings.about.clientName")}{" "}
-                                <span className={"font-mono text-yellow-400"}>
-                                    {t("settings.about.development")}
-                                </span>
-                            </span>
-                        ) : (
-                            t("settings.about.client", { version: clientVersion })
-                        )}
-                    </p>
-                    <p className={"text-sm font-medium text-nb-gray-250"}>
-                        {guiVersion === "development" ? (
-                            <span>
-                                {t("settings.about.guiName")}{" "}
-                                <span className={"font-mono text-yellow-400"}>
-                                    {t("settings.about.development")}
-                                </span>
-                            </span>
-                        ) : (
-                            t("settings.about.gui", { version: guiVersion })
-                        )}
-                    </p>
-                </div>
-
                 <div className={"wails-no-draggable"}>
-                    <Button variant={"primary"} size={"xs"} onClick={() => openUrl(downloadUrl)}>
+                    <Button variant={"primary"} size={"xs"} onClick={() => openUrl(RELEASES_URL)}>
                         <DownloadIcon size={14} />
-                        {t("daemon.outdated.download")}
+                        {t("update.card.getInstaller")}
                     </Button>
                 </div>
             </div>

client/ui/frontend/src/contexts/ProfileContext.tsx

@@ -28,7 +28,6 @@ type ProfileContextValue = {
     loaded: boolean;
     refresh: () => Promise<void>;
     switchProfile: (id: string) => Promise<void>;
-    switchProfileNoConnect: (id: string) => Promise<void>;
     addProfile: (name: string) => Promise<string>;
     removeProfile: (id: string) => Promise<void>;
     renameProfile: (id: string, newName: string) => Promise<void>;
@@ -113,16 +112,6 @@ export const ProfileProvider = ({ children }: { children: ReactNode }) => {
         [username, refresh],
     );
 
-    // Manage-profiles variant: switches without connecting, so the user can
-    // still adjust the management URL before bringing the connection up.
-    const switchProfileNoConnect = useCallback(
-        async (id: string) => {
-            await ProfileSwitcher.SwitchActiveNoConnect({ profileName: id, username });
-            await refresh();
-        },
-        [username, refresh],
-    );
-
     // addProfile creates a profile by display name and returns the
     // daemon-generated ID, so the caller can immediately address it by ID.
     const addProfile = useCallback(
@@ -169,7 +158,6 @@ export const ProfileProvider = ({ children }: { children: ReactNode }) => {
             loaded,
             refresh,
             switchProfile,
-            switchProfileNoConnect,
             addProfile,
             removeProfile,
             renameProfile,
@@ -183,7 +171,6 @@ export const ProfileProvider = ({ children }: { children: ReactNode }) => {
             loaded,
             refresh,
             switchProfile,
-            switchProfileNoConnect,
             addProfile,
             removeProfile,
             renameProfile,

client/ui/frontend/src/modules/profiles/ProfilesTab.tsx

@@ -45,7 +45,7 @@ export function ProfilesTab() {
         activeProfileId,
         loaded,
         username,
-        switchProfileNoConnect,
+        switchProfile,
         addProfile,
         removeProfile,
         renameProfile,
@@ -100,7 +100,7 @@ export function ProfilesTab() {
             confirmLabel: t("profile.switch.confirm"),
         });
         if (!ok) return;
-        await guarded(i18next.t("profile.error.switchTitle"), () => switchProfileNoConnect(id));
+        await guarded(i18next.t("profile.error.switchTitle"), () => switchProfile(id));
     };
 
     const handleDeregister = async (id: string, name: string) => {
@@ -129,13 +129,14 @@ export function ProfilesTab() {
         await guarded(i18next.t("profile.error.createTitle"), async () => {
             const id = await addProfile(name);
             // SetConfig is keyed by the new profile's ID, so it writes the
-            // not-yet-active profile before the switch makes it current.
+            // not-yet-active profile. Write before switching so any reconnect
+            // targets the right deployment.
             if (!isNetbirdCloud(managementUrl)) {
                 await SettingsSvc.SetConfig(
                     new SetConfigParams({ profileName: id, username, managementUrl }),
                 );
             }
-            await switchProfileNoConnect(id);
+            await switchProfile(id);
         });
     };
 

client/ui/frontend/src/modules/session/SessionExpirationDialog.tsx

@@ -73,13 +73,6 @@ export default function SessionExpirationDialog() {
 
         let offCancel: (() => void) | undefined;
 
-        // Return the dialog to its interactive state and dismiss the browser popup
-        const resetDialog = () => {
-            offCancel?.();
-            WindowManager.CloseBrowserLogin().catch(console.error);
-            setBusy(false);
-        };
-
         try {
             const start = await Session.RequestExtend({ hint: "" });
             const uri = start.verificationUriComplete || start.verificationUri;
@@ -112,22 +105,25 @@ export default function SessionExpirationDialog() {
             if (outcome.kind === "cancel") {
                 waitPromise.cancel?.();
                 waitPromise.catch(() => {});
-                resetDialog();
                 return;
             }
 
             // Another surface owns this flow; keep the dialog open to retry.
             if (outcome.result.preempted) {
-                resetDialog();
                 return;
             }
-            WindowManager.CloseRenewFlow().catch(console.error);
+
+            // Close before the popup so the restore can't flash this window back.
+            WindowManager.CloseSessionExpiration().catch(console.error);
         } catch (e) {
-            resetDialog();
             await errorDialog({
                 Title: t("sessionExpiration.extendFailedTitle"),
                 Message: formatErrorMessage(e),
             });
+        } finally {
+            offCancel?.();
+            WindowManager.CloseBrowserLogin().catch(console.error);
+            setBusy(false);
         }
     }, [busy, t]);
 
@@ -143,11 +139,12 @@ export default function SessionExpirationDialog() {
             });
             WindowManager.CloseSessionExpiration().catch(console.error);
         } catch (e) {
-            setBusy(false);
             await errorDialog({
                 Title: t("sessionExpiration.logoutFailedTitle"),
                 Message: formatErrorMessage(e),
             });
+        } finally {
+            setBusy(false);
         }
     }, [busy, t]);
 

client/ui/frontend/src/modules/welcome/WelcomeStepTray.tsx

@@ -22,9 +22,6 @@ type WelcomeStepTrayProps = {
 export function WelcomeStepTray({ onContinue }: Readonly<WelcomeStepTrayProps>) {
     const { t } = useTranslation();
     const trayScreenshot = trayScreenshotForOS();
-    // macOS has no tray — the icon sits in the menu bar, so the copy says so.
-    const titleKey = isMacOS() ? "welcome.titleMac" : "welcome.title";
-    const descriptionKey = isMacOS() ? "welcome.descriptionMac" : "welcome.description";
 
     return (
         <>
@@ -39,9 +36,9 @@ export function WelcomeStepTray({ onContinue }: Readonly<WelcomeStepTrayProps>)
 
             <div className={"flex w-full flex-col gap-1"}>
                 <DialogHeading id={"nb-welcome-title"} align={"left"}>
-                    {t(titleKey)}
+                    {t("welcome.title")}
                 </DialogHeading>
-                <DialogDescription align={"left"}>{t(descriptionKey)}</DialogDescription>
+                <DialogDescription align={"left"}>{t("welcome.description")}</DialogDescription>
             </div>
 
             <DialogActions>

client/ui/i18n/locales/de/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Suchen Sie NetBird in der Taskleiste"
     },
-    "welcome.titleMac": {
-        "message": "Suchen Sie NetBird in der Menüleiste"
-    },
     "welcome.description": {
         "message": "NetBird läuft in Ihrer Taskleiste. Klicken Sie auf das Symbol, um sich zu verbinden, Profile zu wechseln oder die Einstellungen zu öffnen."
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird läuft in Ihrer Menüleiste. Klicken Sie auf das Symbol, um sich zu verbinden, Profile zu wechseln oder die Einstellungen zu öffnen."
-    },
     "welcome.continue": {
         "message": "Weiter"
     },
@@ -1299,13 +1293,10 @@
         "message": "Dokumentation"
     },
     "daemon.outdated.title": {
-        "message": "NetBird Client ist veraltet"
+        "message": "NetBird-Dienst ist veraltet"
     },
     "daemon.outdated.description": {
-        "message": "Die neue GUI ist nicht mit Ihrem älteren Client kompatibel. Aktualisieren Sie Ihren Client, um die neue Anwendung zu verwenden."
-    },
-    "daemon.outdated.download": {
-        "message": "Neueste Version herunterladen"
+        "message": "Aktualisieren Sie den NetBird-Dienst, um diese App zu verwenden."
     },
     "error.jwt_clock_skew": {
         "message": "Anmeldung fehlgeschlagen: Die Uhr dieses Geräts ist nicht mit dem Server synchron. Bitte synchronisieren Sie die Systemuhr und versuchen Sie es erneut."

client/ui/i18n/locales/en/common.json

@@ -1377,19 +1377,11 @@
     },
     "welcome.title": {
         "message": "Look for NetBird in your tray",
-        "description": "Heading on the first onboarding step, pointing the user to the tray icon. Shown on Windows and Linux; macOS uses welcome.titleMac."
-    },
-    "welcome.titleMac": {
-        "message": "Look for NetBird in your menu bar",
-        "description": "Heading on the first onboarding step on macOS, pointing the user to the menu bar icon. Use your language's Apple term for the macOS menu bar."
+        "description": "Heading on the first onboarding step, pointing the user to the tray icon. 'tray' = system tray / menu bar."
     },
     "welcome.description": {
         "message": "NetBird lives in your tray. Click the icon to connect, switch profiles, or open settings.",
-        "description": "Body of the first onboarding step explaining the tray icon. Shown on Windows and Linux; macOS uses welcome.descriptionMac."
-    },
-    "welcome.descriptionMac": {
-        "message": "NetBird lives in your menu bar. Click the icon to connect, switch profiles, or open settings.",
-        "description": "Body of the first onboarding step on macOS explaining the menu bar icon. Use your language's Apple term for the macOS menu bar."
+        "description": "Body of the first onboarding step explaining the tray icon."
     },
     "welcome.continue": {
         "message": "Continue",
@@ -1732,16 +1724,12 @@
         "description": "Documentation link on the daemon-unavailable overlay."
     },
     "daemon.outdated.title": {
-        "message": "NetBird Client Is Outdated",
-        "description": "Title of the overlay shown when the NetBird client (daemon) is too old to drive this UI."
+        "message": "NetBird Service Is Outdated",
+        "description": "Title of the overlay shown when the NetBird background service is too old to drive this UI."
     },
     "daemon.outdated.description": {
-        "message": "The new GUI isn't compatible with the older NetBird client. Update your client to use the new application.",
-        "description": "Body of the daemon-outdated overlay explaining that the GUI is newer than the client and the client must be updated."
-    },
-    "daemon.outdated.download": {
-        "message": "Download Latest",
-        "description": "Button on the daemon-outdated overlay that opens the download page for the latest release."
+        "message": "Update the NetBird service to use this app.",
+        "description": "Body of the daemon-outdated overlay telling the user to upgrade the service."
     },
     "error.jwt_clock_skew": {
         "message": "Sign-in failed: this device's clock is out of sync with the server. Please sync your system clock and try again.",

client/ui/i18n/locales/es/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Busque NetBird en su bandeja del sistema"
     },
-    "welcome.titleMac": {
-        "message": "Busque NetBird en su barra de menús"
-    },
     "welcome.description": {
         "message": "NetBird reside en su bandeja del sistema. Haga clic en el icono para conectarse, cambiar de perfil o abrir la configuración."
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird reside en su barra de menús. Haga clic en el icono para conectarse, cambiar de perfil o abrir la configuración."
-    },
     "welcome.continue": {
         "message": "Continuar"
     },
@@ -1299,13 +1293,10 @@
         "message": "Documentación"
     },
     "daemon.outdated.title": {
-        "message": "NetBird Client está desactualizado"
+        "message": "El servicio de NetBird está desactualizado"
     },
     "daemon.outdated.description": {
-        "message": "La nueva GUI no es compatible con su cliente anterior. Actualice su cliente para usar la nueva aplicación."
-    },
-    "daemon.outdated.download": {
-        "message": "Descargar la última versión"
+        "message": "Actualice el servicio de NetBird para usar esta aplicación."
     },
     "error.jwt_clock_skew": {
         "message": "Error al iniciar sesión: el reloj de este dispositivo no está sincronizado con el servidor. Sincronice el reloj del sistema e inténtelo de nuevo."

client/ui/i18n/locales/fr/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Cherchez NetBird dans votre barre d’état système"
     },
-    "welcome.titleMac": {
-        "message": "Cherchez NetBird dans votre barre des menus"
-    },
     "welcome.description": {
         "message": "NetBird se trouve dans votre barre d’état système. Cliquez sur l’icône pour vous connecter, changer de profil ou ouvrir les paramètres."
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird se trouve dans votre barre des menus. Cliquez sur l’icône pour vous connecter, changer de profil ou ouvrir les paramètres."
-    },
     "welcome.continue": {
         "message": "Continuer"
     },
@@ -1299,13 +1293,10 @@
         "message": "Documentation"
     },
     "daemon.outdated.title": {
-        "message": "Le Client NetBird est obsolète"
+        "message": "Le service NetBird est obsolète"
     },
     "daemon.outdated.description": {
-        "message": "La nouvelle GUI n'est pas compatible avec votre ancien client. Mettez à jour votre client pour utiliser la nouvelle application."
-    },
-    "daemon.outdated.download": {
-        "message": "Télécharger la dernière version"
+        "message": "Mettez à jour le service NetBird pour utiliser cette application."
     },
     "error.jwt_clock_skew": {
         "message": "Échec de la connexion : l’horloge de cet appareil n’est pas synchronisée avec le serveur. Veuillez synchroniser l’horloge de votre système et réessayer."

client/ui/i18n/locales/hu/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Keresse a NetBirdöt a tálcán"
     },
-    "welcome.titleMac": {
-        "message": "Keresse a NetBirdöt a menüsorban"
-    },
     "welcome.description": {
         "message": "A NetBird a tálcán fut. Kattintson az ikonra a csatlakozáshoz, profilváltáshoz vagy a beállítások megnyitásához."
     },
-    "welcome.descriptionMac": {
-        "message": "A NetBird a menüsorban fut. Kattintson az ikonra a csatlakozáshoz, profilváltáshoz vagy a beállítások megnyitásához."
-    },
     "welcome.continue": {
         "message": "Folytatás"
     },
@@ -1299,13 +1293,10 @@
         "message": "Dokumentáció"
     },
     "daemon.outdated.title": {
-        "message": "A NetBird Kliens elavult"
+        "message": "A NetBird szolgáltatás elavult"
     },
     "daemon.outdated.description": {
-        "message": "Az új GUI nem kompatibilis a régebbi klienseddel. Frissítsd a klienst az új alkalmazás használatához."
-    },
-    "daemon.outdated.download": {
-        "message": "Legújabb letöltése"
+        "message": "Frissítsd a NetBird szolgáltatást az alkalmazás használatához."
     },
     "error.jwt_clock_skew": {
         "message": "A bejelentkezés sikertelen: az eszköz órája eltér a szerverétől. Kérjük, szinkronizálja a rendszer óráját, majd próbálja újra."

client/ui/i18n/locales/it/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Cerchi NetBird nella tray"
     },
-    "welcome.titleMac": {
-        "message": "Cerchi NetBird nella barra dei menu"
-    },
     "welcome.description": {
         "message": "NetBird risiede nella tray. Clicchi sull'icona per connettersi, cambiare profilo o aprire le impostazioni."
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird risiede nella barra dei menu. Clicchi sull'icona per connettersi, cambiare profilo o aprire le impostazioni."
-    },
     "welcome.continue": {
         "message": "Continua"
     },
@@ -1299,13 +1293,10 @@
         "message": "Documentazione"
     },
     "daemon.outdated.title": {
-        "message": "NetBird Client è obsoleto"
+        "message": "Il servizio NetBird è obsoleto"
     },
     "daemon.outdated.description": {
-        "message": "La nuova GUI non è compatibile con il tuo client precedente. Aggiorna il client per usare la nuova applicazione."
-    },
-    "daemon.outdated.download": {
-        "message": "Scarica l'ultima versione"
+        "message": "Aggiorna il servizio NetBird per usare questa app."
     },
     "error.jwt_clock_skew": {
         "message": "Accesso non riuscito: l'orologio di questo dispositivo non è sincronizzato con il server. Sincronizzi l'orologio di sistema e riprovi."

client/ui/i18n/locales/ja/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "トレイの NetBird を確認してください"
     },
-    "welcome.titleMac": {
-        "message": "メニューバーの NetBird を確認してください"
-    },
     "welcome.description": {
         "message": "NetBird はトレイに常駐します。アイコンをクリックして、接続、プロファイルの切り替え、設定を開くことができます。"
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird はメニューバーに常駐します。アイコンをクリックして、接続、プロファイルの切り替え、設定を開くことができます。"
-    },
     "welcome.continue": {
         "message": "続ける"
     },

client/ui/i18n/locales/pt/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Procure o NetBird na sua bandeja"
     },
-    "welcome.titleMac": {
-        "message": "Procure o NetBird na sua barra de menus"
-    },
     "welcome.description": {
         "message": "O NetBird fica na sua bandeja. Clique no ícone para conectar, alternar perfis ou abrir as configurações."
     },
-    "welcome.descriptionMac": {
-        "message": "O NetBird fica na sua barra de menus. Clique no ícone para conectar, alternar perfis ou abrir as configurações."
-    },
     "welcome.continue": {
         "message": "Continuar"
     },
@@ -1299,13 +1293,10 @@
         "message": "Documentação"
     },
     "daemon.outdated.title": {
-        "message": "O NetBird Client está desatualizado"
+        "message": "O serviço NetBird está desatualizado"
     },
     "daemon.outdated.description": {
-        "message": "A nova GUI não é compatível com o seu cliente mais antigo. Atualize o seu cliente para usar o novo aplicativo."
-    },
-    "daemon.outdated.download": {
-        "message": "Baixar a versão mais recente"
+        "message": "Atualize o serviço NetBird para usar este aplicativo."
     },
     "error.jwt_clock_skew": {
         "message": "Falha no login: o relógio deste dispositivo está fora de sincronia com o servidor. Sincronize o relógio do sistema e tente novamente."

client/ui/i18n/locales/ru/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "Найдите NetBird в системном трее"
     },
-    "welcome.titleMac": {
-        "message": "Найдите NetBird в строке меню"
-    },
     "welcome.description": {
         "message": "NetBird находится в системном трее. Нажмите на значок, чтобы подключиться, переключить профиль или открыть настройки."
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird находится в строке меню. Нажмите на значок, чтобы подключиться, переключить профиль или открыть настройки."
-    },
     "welcome.continue": {
         "message": "Продолжить"
     },
@@ -1299,13 +1293,10 @@
         "message": "Документация"
     },
     "daemon.outdated.title": {
-        "message": "Клиент NetBird устарел"
+        "message": "Служба NetBird устарела"
     },
     "daemon.outdated.description": {
-        "message": "Новый GUI несовместим с вашим более старым клиентом. Обновите клиент, чтобы использовать новое приложение."
-    },
-    "daemon.outdated.download": {
-        "message": "Скачать последнюю версию"
+        "message": "Обновите службу NetBird, чтобы использовать это приложение."
     },
     "error.jwt_clock_skew": {
         "message": "Не удалось войти: часы этого устройства рассинхронизированы с сервером. Синхронизируйте системные часы и повторите попытку."

client/ui/i18n/locales/zh-CN/common.json

@@ -1034,15 +1034,9 @@
     "welcome.title": {
         "message": "在托盘中查找 NetBird"
     },
-    "welcome.titleMac": {
-        "message": "在菜单栏中查找 NetBird"
-    },
     "welcome.description": {
         "message": "NetBird 驻留在您的托盘中。点击图标即可连接、切换配置文件或打开设置。"
     },
-    "welcome.descriptionMac": {
-        "message": "NetBird 驻留在您的菜单栏中。点击图标即可连接、切换配置文件或打开设置。"
-    },
     "welcome.continue": {
         "message": "继续"
     },
@@ -1299,13 +1293,10 @@
         "message": "文档"
     },
     "daemon.outdated.title": {
-        "message": "NetBird 客户端版本过旧"
+        "message": "NetBird 服务版本过旧"
     },
     "daemon.outdated.description": {
-        "message": "新版 GUI 与您较旧的客户端不兼容。请更新客户端以使用新应用。"
-    },
-    "daemon.outdated.download": {
-        "message": "下载最新版本"
+        "message": "请更新 NetBird 服务以使用此应用。"
     },
     "error.jwt_clock_skew": {
         "message": "登录失败：此设备的时钟与服务器不同步。请同步您的系统时钟后重试。"

client/ui/services/profileswitcher.go

@@ -12,15 +12,13 @@ import (
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 )
 
-// ProfileSwitcher holds the switch policy shared by the tray and React
-// frontend so both flip profiles identically. SwitchActive (plain selection:
-// header dropdown, tray submenu) always connects after the switch;
-// SwitchActiveNoConnect (manage-profiles screen) never does, so the user can
-// still adjust the management URL before connecting. prevStatus from
-// DaemonFeed.Get at entry only decides the teardown:
+// ProfileSwitcher holds the reconnect policy shared by the tray and React
+// frontend so both flip profiles identically. The policy keys off prevStatus
+// from DaemonFeed.Get at SwitchActive entry:
 //
-//	Connected/Connecting/NeedsLogin/LoginFailed/SessionExpired → Down first.
-//	Idle → no Down.
+//	Connected/Connecting → Switch + Down + Up; optimistic Connecting paint.
+//	NeedsLogin/LoginFailed/SessionExpired → Switch + Down; clear stale error for re-login.
+//	Idle → Switch only.
 type ProfileSwitcher struct {
 	profiles   *Profiles
 	connection *Connection
@@ -31,40 +29,29 @@ func NewProfileSwitcher(profiles *Profiles, connection *Connection, feed *Daemon
 	return &ProfileSwitcher{profiles: profiles, connection: connection, feed: feed}
 }
 
-// SwitchActive switches to the named profile and always connects afterwards.
+// SwitchActive switches to the named profile applying the reconnect policy.
 func (s *ProfileSwitcher) SwitchActive(ctx context.Context, p ProfileRef) error {
-	return s.switchActive(ctx, p, true)
-}
-
-// SwitchActiveNoConnect switches to the named profile without connecting,
-// tearing down any existing connection first.
-func (s *ProfileSwitcher) SwitchActiveNoConnect(ctx context.Context, p ProfileRef) error {
-	return s.switchActive(ctx, p, false)
-}
-
-func (s *ProfileSwitcher) switchActive(ctx context.Context, p ProfileRef, connect bool) error {
 	prevStatus := ""
-	if s.feed != nil {
-		if st, err := s.feed.Get(ctx); err == nil {
-			prevStatus = st.Status
-		} else {
-			log.Warnf("profileswitcher: get status: %v", err)
-		}
+	if st, err := s.feed.Get(ctx); err == nil {
+		prevStatus = st.Status
+	} else {
+		log.Warnf("profileswitcher: get status: %v", err)
 	}
 
-	needsDown := strings.EqualFold(prevStatus, StatusConnected) ||
-		strings.EqualFold(prevStatus, StatusConnecting) ||
+	wasActive := strings.EqualFold(prevStatus, StatusConnected) ||
+		strings.EqualFold(prevStatus, StatusConnecting)
+	needsDown := wasActive ||
 		strings.EqualFold(prevStatus, StatusNeedsLogin) ||
 		strings.EqualFold(prevStatus, StatusLoginFailed) ||
 		strings.EqualFold(prevStatus, StatusSessionExpired)
 
-	log.Infof("profileswitcher: switch profile=%q prevStatus=%q connect=%v needsDown=%v",
-		p.ProfileName, prevStatus, connect, needsDown)
+	log.Infof("profileswitcher: switch profile=%q prevStatus=%q wasActive=%v needsDown=%v",
+		p.ProfileName, prevStatus, wasActive, needsDown)
 
-	// Optimistic Connecting paint plus stale-push suppression during Down (see
-	// DaemonFeed suppression table); also arms the login-watch that pops
-	// browser-login when the new profile turns out to need SSO.
-	if connect && s.feed != nil {
+	// Optimistic Connecting paint only when wasActive: those prevStatuses emit
+	// stale Connected + transient Idle pushes during Down that must be
+	// suppressed until Up resumes the stream (see DaemonFeed suppression table).
+	if wasActive {
 		s.feed.BeginProfileSwitch()
 	}
 
@@ -89,9 +76,9 @@ func (s *ProfileSwitcher) switchActive(ctx context.Context, p ProfileRef, connec
 		}
 	}
 
-	if connect {
+	if wasActive {
 		if err := s.connection.Up(ctx, UpParams(p)); err != nil {
-			return fmt.Errorf("connect %q: %w", p.ProfileName, err)
+			return fmt.Errorf("reconnect %q: %w", p.ProfileName, err)
 		}
 	}
 

client/ui/services/windowmanager.go

@@ -185,38 +185,37 @@ func (s *WindowManager) OpenBrowserLogin(uri string) {
 			startURL = "/#/dialog/browser-login?uri=" + url.QueryEscape(uri)
 		}
 		s.hideOtherWindowsLocked("browser-login")
+		// Prefer the main window's screen (multi-monitor); falls back to OS-default centering.
+		var screen *application.Screen
+		if s.mainWindow != nil {
+			if sc, err := s.mainWindow.GetScreen(); err == nil {
+				screen = sc
+			}
+		}
 		opts := DialogWindowOptions("browser-login", s.title("window.title.signIn"), startURL, s.linuxIcon)
 		// Not always-on-top: it would obscure the browser tab the user logs in through.
 		opts.AlwaysOnTop = false
 		opts.InitialPosition = application.WindowCentered
-		// Open on the active (where users cursor is) display, like the session-expiration dialog.
-		opts.Screen = s.getScreenBasedOnCursorPosition()
+		opts.Screen = screen
 		s.browserLogin = s.app.Window.NewWithOptions(opts)
 		bl := s.browserLogin
+		// Red-X close means cancel: emit the event so startLogin() tears down the SSO wait.
 		bl.OnWindowEvent(events.Common.WindowClosing, func(_ *application.WindowEvent) {
+			s.app.Event.Emit(EventBrowserLoginCancel)
 			s.mu.Lock()
-			// Only a live user red-X still has this registered; programmatic closers
-			// nil s.browserLogin first and clean up themselves. Guarding here stops a
-			// stale close event from wiping a replacement popup's state.
-			userClosed := s.browserLogin == bl
-			if userClosed {
-				s.browserLogin = nil
-				s.restoreHiddenWindowsLocked()
-			}
+			s.browserLogin = nil
+			s.restoreHiddenWindowsLocked()
 			s.mu.Unlock()
-			if userClosed {
-				s.app.Event.Emit(EventBrowserLoginCancel)
-			}
 		})
-		s.centerOnCursorScreen(s.browserLogin)
+		s.centerWhenReady(s.browserLogin)
 		return
 	}
 	if uri != "" {
 		s.browserLogin.SetURL("/#/dialog/browser-login?uri=" + url.QueryEscape(uri))
 	}
-	s.centerOnCursorScreen(s.browserLogin)
 	s.browserLogin.Show()
 	s.browserLogin.Focus()
+	s.centerWhenReady(s.browserLogin)
 }
 
 // BrowserLoginWindow returns the live SSO popup, or nil. While non-nil it is the
@@ -239,15 +238,6 @@ func (s *WindowManager) CloseBrowserLogin() {
 	s.mu.Lock()
 	w := s.browserLogin
 	s.browserLogin = nil
-	// The WindowClosing hook no-ops on a programmatic close, so restore here —
-	// but only if a popup was actually open. The frontend calls this even when no
-	// popup was ever shown (e.g. resetDialog() after an early RequestExtend failure,
-	// or connection.ts's catch path), and hiddenForLogin is shared with
-	// OpenInstallProgress, so an unconditional restore could re-show windows a
-	// still-running install-progress is hiding.
-	if w != nil {
-		s.restoreHiddenWindowsLocked()
-	}
 	s.mu.Unlock()
 	if w != nil {
 		w.Close()
@@ -289,35 +279,6 @@ func (s *WindowManager) CloseSessionExpiration() {
 	}
 }
 
-// CloseRenewFlow tears down the SSO session-renewal UI in a single call: it
-// closes the browser-login popup and the session-expiration window together.
-func (s *WindowManager) CloseRenewFlow() {
-	s.mu.Lock()
-	bl := s.browserLogin
-	se := s.sessionExpiration
-	s.browserLogin = nil
-	s.sessionExpiration = nil
-	if se != nil {
-		kept := s.hiddenForLogin[:0]
-		for _, w := range s.hiddenForLogin {
-			if w != se {
-				kept = append(kept, w)
-			}
-		}
-		s.hiddenForLogin = kept
-	}
-	s.restoreHiddenWindowsLocked()
-	s.mu.Unlock()
-
-	// Close after unlock so the re-entrant handlers can take s.mu.
-	if bl != nil {
-		bl.Close()
-	}
-	if se != nil {
-		se.Close()
-	}
-}
-
 // OpenInstallProgress shows the install-progress window and hides the rest for the duration
 // (restored on close). It owns its own result polling since the daemon restarts mid-install.
 func (s *WindowManager) OpenInstallProgress(version string) {

client/ui/tray.go

@@ -30,8 +30,6 @@ const (
 
 	statusError = "Error"
 
-	quitDownTimeout = 5 * time.Second
-
 	urlGitHubRepo     = "https://github.com/netbirdio/netbird"
 	urlGitHubReleases = "https://github.com/netbirdio/netbird/releases/latest"
 	urlDocs           = "https://docs.netbird.io"
@@ -448,28 +446,11 @@ func (t *Tray) buildMenu() *application.Menu {
 	menu.AddSeparator()
 	menu.Add(t.loc.T("tray.menu.quit")).
 		SetAccelerator("CmdOrCtrl+Q").
-		OnClick(func(*application.Context) { t.handleQuit() })
+		OnClick(func(*application.Context) { t.app.Quit() })
 
 	return menu
 }
 
-func (t *Tray) handleQuit() {
-	t.profileMu.Lock()
-	if t.switchCancel != nil {
-		t.switchCancel()
-		t.switchCancel = nil
-	}
-	t.profileMu.Unlock()
-	t.svc.DaemonFeed.CancelProfileSwitch()
-
-	ctx, cancel := context.WithTimeout(context.Background(), quitDownTimeout)
-	defer cancel()
-	if err := t.svc.Connection.Down(ctx); err != nil {
-		log.Errorf("disconnect on quit: %v", err)
-	}
-	t.app.Quit()
-}
-
 // handleConnect receives the clicked item from the buildMenu closure —
 // t.upItem is menuMu-guarded and must not be read here.
 func (t *Tray) handleConnect(upItem *application.MenuItem) {

e2e/agentnetwork/chat_test.go

@@ -91,22 +91,14 @@ func availableProviders() []providerCase {
 		if region == "" {
 			region = "us-east-1"
 		}
-		// A valid Bedrock inference-profile id (region prefix + date + version),
-		// overridable per account. `global.` profiles can be invoked from any
-		// region; set AWS_BEDROCK_MODEL to match the enabled profile for the token.
-		model := os.Getenv("AWS_BEDROCK_MODEL")
-		if model == "" {
-			model = "global.anthropic.claude-haiku-4-5-20251001-v1:0"
-		}
-		ps = append(ps, providerCase{name: "bedrock", catalogID: "bedrock_api", upstream: "https://bedrock-runtime." + region + ".amazonaws.com", apiKey: k, model: model, kind: harness.WireBedrock})
+		ps = append(ps, providerCase{name: "bedrock", catalogID: "bedrock_api", upstream: "https://bedrock-runtime." + region + ".amazonaws.com", apiKey: k, model: "us.anthropic.claude-haiku-4-5", kind: harness.WireBedrock})
 	}
 	return ps
 }
 
 // providerRequest builds a create request for a matrix provider: enabled, with
-// a uniquely-priced model registered under the id an operator would paste —
-// the normalized catalog id for Bedrock, the raw form elsewhere (including the
-// "@version" Vertex id, which the router must normalize to route).
+// a uniquely-priced model for body-routed providers and none for the
+// path-routed Vertex (whose model lives in the request path).
 func providerRequest(pc providerCase) api.AgentNetworkProviderRequest {
 	req := api.AgentNetworkProviderRequest{
 		Name:        pc.name,
@@ -115,19 +107,10 @@ func providerRequest(pc providerCase) api.AgentNetworkProviderRequest {
 		ApiKey:      &pc.apiKey,
 		Enabled:     ptr(true),
 	}
-	// The router matches the normalized catalog id. Bedrock's request model
-	// travels as a region-prefixed inference-profile id in the URL path
-	// (us.anthropic...), which the router strips before matching, so register
-	// the normalized form here or routing fails as model_not_routable. Vertex is
-	// registered with the raw "@version" id the Google console documents — the
-	// router normalizes the registered id, so the operator can paste either
-	// form; keeping the raw form here guards that normalization end to end.
-	modelID := pc.model
-	if pc.kind == harness.WireBedrock {
-		modelID = catalogModel(pc)
-	}
-	req.Models = &[]api.AgentNetworkProviderModel{
-		{Id: modelID, InputPer1k: 0.001, OutputPer1k: 0.002},
+	if pc.kind != harness.WireVertex {
+		req.Models = &[]api.AgentNetworkProviderModel{
+			{Id: pc.model, InputPer1k: 0.001, OutputPer1k: 0.002},
+		}
 	}
 	return req
 }
@@ -218,13 +201,11 @@ func TestProvidersMatrix(t *testing.T) {
 	t.Cleanup(func() { _ = cl.Terminate(context.Background()) })
 
 	require.NoError(t, cl.WaitConnected(ctx, 90*time.Second), "client must connect to management")
-	// Resolve first: the DNS lookup triggers the lazy-connection warm-up, waking
-	// the proxy peer so WaitProxyPeer then observes it connected.
-	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
-	require.NoError(t, err, "resolve agent-network endpoint to proxy IP")
 	if err := cl.WaitProxyPeer(ctx, 180*time.Second); err != nil {
 		t.Fatalf("client did not see the proxy peer: %v\n=== proxy logs ===\n%s", err, px.Logs(context.Background()))
 	}
+	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
+	require.NoError(t, err, "resolve agent-network endpoint to proxy IP")
 
 	for _, pc := range matrix {
 		pc := pc

e2e/agentnetwork/guardrail_test.go

@@ -4,7 +4,6 @@ package agentnetwork
 
 import (
 	"context"
-	"regexp"
 	"strings"
 	"testing"
 	"time"
@@ -16,29 +15,13 @@ import (
 	"github.com/netbirdio/netbird/shared/management/http/api"
 )
 
-// bedrockRegionPrefixes and bedrockVersionSuffix mirror the proxy's Bedrock
-// model normalization (region/inference-profile prefix + version suffix) so the
-// provider is registered under the same catalog key the router matches against.
-var (
-	bedrockRegionPrefixes = []string{"us.", "eu.", "apac.", "global."}
-	bedrockVersionSuffix  = regexp.MustCompile(`-(\d{8}-)?v\d+(:\d+)?$`)
-)
-
 // catalogModel returns the normalized catalog id the proxy stamps for a
-// path-routed provider's configured model — the form the router and guardrail
-// allowlist compare against (Bedrock region prefix + version stripped, Vertex
-// @version stripped).
+// path-routed provider's configured model — the form the guardrail allowlist is
+// compared against (region prefix / @version stripped).
 func catalogModel(pc providerCase) string {
 	switch pc.kind {
 	case harness.WireBedrock:
-		m := pc.model
-		for _, p := range bedrockRegionPrefixes {
-			if strings.HasPrefix(m, p) {
-				m = m[len(p):]
-				break
-			}
-		}
-		return bedrockVersionSuffix.ReplaceAllString(m, "")
+		return strings.TrimPrefix(pc.model, "us.")
 	case harness.WireVertex:
 		return strings.SplitN(pc.model, "@", 2)[0]
 	default:
@@ -125,15 +108,7 @@ func TestModelAllowlistEnforced(t *testing.T) {
 		require.NoError(t, perr, "create provider %s", pc.name)
 		id := prov.Id
 		ids = append(ids, id)
-		// Vertex allowlists the raw "@version" id an operator copies from the
-		// Google console; the guardrail must normalize the entry to match the
-		// version-stripped request model. Other providers allowlist the
-		// normalized catalog id.
-		if pc.kind == harness.WireVertex {
-			allowed = append(allowed, pc.model)
-		} else {
-			allowed = append(allowed, catalogModel(pc))
-		}
+		allowed = append(allowed, catalogModel(pc))
 		t.Cleanup(func() { _ = srv.DeleteProvider(context.Background(), id) })
 	}
 
@@ -172,13 +147,11 @@ func TestModelAllowlistEnforced(t *testing.T) {
 	t.Cleanup(func() { _ = cl.Terminate(context.Background()) })
 
 	require.NoError(t, cl.WaitConnected(ctx, 90*time.Second), "client must connect to management")
-	// Resolve first: the DNS lookup triggers the lazy-connection warm-up, waking
-	// the proxy peer so WaitProxyPeer then observes it connected.
-	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
-	require.NoError(t, err, "resolve agent-network endpoint to proxy IP")
 	if err := cl.WaitProxyPeer(ctx, 180*time.Second); err != nil {
 		t.Fatalf("client did not see the proxy peer: %v\n=== proxy logs ===\n%s", err, px.Logs(context.Background()))
 	}
+	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
+	require.NoError(t, err, "resolve agent-network endpoint to proxy IP")
 
 	for _, pc := range providers {
 		pc := pc
@@ -190,25 +163,6 @@ func TestModelAllowlistEnforced(t *testing.T) {
 			// the upstream), regardless of whether it is a real catalog model.
 			assert.Equal(t, 403, sendModel(ctx, t, cl, settings.Endpoint, proxyIP, pc, disallowedModel(pc)),
 				"model outside the allowlist must be denied for %s", pc.name)
-
-			if pc.kind != harness.WireVertex {
-				return
-			}
-			// The unversioned model id — the shape an Anthropic SDK client sends
-			// when configured without an "@version" suffix (customer-reported) —
-			// must pass the same allowlist entry and resolve upstream.
-			assert.Equal(t, 200, sendModel(ctx, t, cl, settings.Endpoint, proxyIP, pc, catalogModel(pc)),
-				"unversioned model id must be permitted for %s", pc.name)
-			// Token counting: the URL carries the literal "count-tokens"
-			// pseudo-model and the real model travels in the body (Claude Code
-			// sends one such call per request). Allowed body model → served.
-			code, _, err := cl.VertexCountTokens(ctx, settings.Endpoint, proxyIP, pc.project, pc.region, pc.model)
-			require.NoError(t, err, "count-tokens request must reach the proxy for %s", pc.name)
-			assert.Equal(t, 200, code, "count-tokens with an allowlisted body model must be permitted for %s", pc.name)
-			// Disallowed body model → denied before the upstream.
-			code, _, err = cl.VertexCountTokens(ctx, settings.Endpoint, proxyIP, pc.project, pc.region, disallowedModel(pc))
-			require.NoError(t, err, "count-tokens request must reach the proxy for %s", pc.name)
-			assert.Equal(t, 403, code, "count-tokens with a body model outside the allowlist must be denied for %s", pc.name)
 		})
 	}
 }

e2e/agentnetwork/skiptls_test.go

@@ -104,13 +104,11 @@ func TestProviderSkipTLSVerification(t *testing.T) {
 	t.Cleanup(func() { _ = cl.Terminate(context.Background()) })
 
 	require.NoError(t, cl.WaitConnected(ctx, 90*time.Second), "client must connect to management")
-	// Resolve first: the DNS lookup triggers the lazy-connection warm-up, waking
-	// the proxy peer so WaitProxyPeer then observes it connected.
-	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
-	require.NoError(t, err, "resolve endpoint to proxy IP")
 	if err := cl.WaitProxyPeer(ctx, 180*time.Second); err != nil {
 		t.Fatalf("client did not see the proxy peer: %v\n=== proxy logs ===\n%s", err, px.Logs(context.Background()))
 	}
+	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
+	require.NoError(t, err, "resolve endpoint to proxy IP")
 
 	// Positive: skip=true reaches the self-signed upstream. Retry to absorb
 	// tunnel/DNS jitter on the first call; success also proves the path works.

e2e/agentnetwork/vllm_test.go

@@ -106,13 +106,11 @@ func TestVLLMProvider(t *testing.T) {
 	t.Cleanup(func() { _ = cl.Terminate(context.Background()) })
 
 	require.NoError(t, cl.WaitConnected(ctx, 90*time.Second), "client must connect to management")
-	// Resolve first: the DNS lookup triggers the lazy-connection warm-up, waking
-	// the proxy peer so WaitProxyPeer then observes it connected.
-	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
-	require.NoError(t, err, "resolve endpoint to proxy IP")
 	if err := cl.WaitProxyPeer(ctx, 180*time.Second); err != nil {
 		t.Fatalf("client did not see the proxy peer: %v\n=== proxy logs ===\n%s", err, px.Logs(context.Background()))
 	}
+	proxyIP, err := cl.ResolveProxyIP(ctx, settings.Endpoint)
+	require.NoError(t, err, "resolve endpoint to proxy IP")
 
 	before, _ := srv.ListAccessLogs(ctx)
 	sessionID := "e2e-session-vllm"

e2e/harness/client.go

@@ -231,17 +231,6 @@ func (cl *Client) Vertex(ctx context.Context, endpoint, proxyIP, project, region
 	return cl.post(ctx, endpoint, proxyIP, path, body, withSessionID(nil, sessionID))
 }
 
-// VertexCountTokens issues the Anthropic-on-Vertex token-count POST over the
-// tunnel. The URL carries the literal "count-tokens" pseudo-model and the real
-// model travels in the body — the one Vertex shape whose model is body-routed,
-// sent by Claude Code before each request. The proxy must resolve the body
-// model for the guardrail allowlist and routing.
-func (cl *Client) VertexCountTokens(ctx context.Context, endpoint, proxyIP, project, region, model string) (int, string, error) {
-	path := fmt.Sprintf("/v1/projects/%s/locations/%s/publishers/anthropic/models/count-tokens:rawPredict", project, region)
-	body := fmt.Sprintf(`{"model":%q,"messages":[{"role":"user","content":"hi"}]}`, model)
-	return cl.post(ctx, endpoint, proxyIP, path, body, nil)
-}
-
 // Bedrock issues a native AWS Bedrock InvokeModel POST over the tunnel. The
 // model id is carried in the request path (/model/{id}/invoke), so the proxy
 // routes by path; the body uses the bedrock anthropic_version rather than a

proxy/Dockerfile.multistage

@@ -14,7 +14,6 @@ COPY proxy ./proxy
 COPY route ./route
 COPY shared ./shared
 COPY sharedsock ./sharedsock
-COPY trustedproxy ./trustedproxy
 COPY upload-server ./upload-server
 COPY util ./util
 COPY version ./version

proxy/internal/llm/vertex_model.go

@@ -1,17 +0,0 @@
-package llm
-
-import "strings"
-
-// NormalizeVertexModel strips the "@version" suffix from a Google Vertex AI
-// publisher model id so it matches the catalog/pricing key, e.g.
-// "claude-sonnet-4-5@20250929" -> "claude-sonnet-4-5". It is the single source
-// of truth shared by the request parser (which normalizes the request model
-// from the URL path), the router (which normalizes the operator's registered
-// Vertex model ids), and the guardrail (which normalizes allowlist entries) so
-// both spellings of the same model compare equal.
-func NormalizeVertexModel(modelID string) string {
-	if at := strings.Index(modelID, "@"); at >= 0 {
-		return modelID[:at]
-	}
-	return modelID
-}

proxy/internal/llm/vertex_model_test.go

@@ -1,23 +0,0 @@
-package llm
-
-import (
-	"testing"
-
-	"github.com/stretchr/testify/require"
-)
-
-func TestNormalizeVertexModel(t *testing.T) {
-	cases := map[string]string{
-		"claude-sonnet-4-5@20250929": "claude-sonnet-4-5",
-		"claude-opus-4-6@20250514":   "claude-opus-4-6",
-		// Bare ids (the form Vertex resolves to the default version) pass through.
-		"claude-opus-4-6": "claude-opus-4-6",
-		// Numeric version aliases used by Google publisher models.
-		"text-embedding-005@001": "text-embedding-005",
-		// A leading "@" yields the empty string; callers treat that as "no model".
-		"@cf/meta/llama-3-8b": "",
-	}
-	for in, want := range cases {
-		require.Equal(t, want, NormalizeVertexModel(in), "normalize %q", in)
-	}
-}

proxy/internal/middleware/builtin/llm_guardrail/middleware.go

@@ -13,7 +13,6 @@ import (
 	"context"
 	"unicode/utf8"
 
-	"github.com/netbirdio/netbird/proxy/internal/llm"
 	"github.com/netbirdio/netbird/proxy/internal/middleware"
 )
 
@@ -163,15 +162,6 @@ func (m *Middleware) modelInAllowlist(model string) bool {
 		if allowed == normalised {
 			return true
 		}
-		// Vertex model ids are often stored with their "@version" suffix (e.g.
-		// "claude-opus-4-6@20250514") — the id the Google console documents —
-		// while the request parser strips the suffix before the model reaches
-		// the guardrail. Accept an entry whose version-stripped form matches so
-		// both spellings of the same model pass one allowlist. Entries without
-		// an "@" are untouched, so non-Vertex matching stays exact.
-		if v := llm.NormalizeVertexModel(allowed); v != "" && v != allowed && v == normalised {
-			return true
-		}
 	}
 	return false
 }

proxy/internal/middleware/builtin/llm_guardrail/middleware_test.go

@@ -70,27 +70,6 @@ func TestAllowlistMatchAllows(t *testing.T) {
 	assert.Equal(t, middleware.DecisionAllow, out.Decision, "model in allowlist must be allowed")
 }
 
-// A Vertex allowlist entry stored with its "@version" suffix must match the
-// request model, which reaches the guardrail with the suffix already stripped
-// by the request parser.
-func TestAllowlistVertexVersionedEntryMatchesStrippedModel(t *testing.T) {
-	mw := New(Config{ModelAllowlist: []string{"claude-opus-4-6@20250514"}})
-	out, err := mw.Invoke(context.Background(), newInput(
-		middleware.KV{Key: middleware.KeyLLMModel, Value: "claude-opus-4-6"},
-	))
-	require.NoError(t, err)
-	assert.Equal(t, middleware.DecisionAllow, out.Decision,
-		"@version allowlist entry must match the version-stripped request model")
-
-	// The stripped form must not over-match a different model.
-	out, err = mw.Invoke(context.Background(), newInput(
-		middleware.KV{Key: middleware.KeyLLMModel, Value: "claude-opus-4-8"},
-	))
-	require.NoError(t, err)
-	assert.Equal(t, middleware.DecisionDeny, out.Decision,
-		"a different model must stay denied")
-}
-
 func TestAllowlistMissDenies(t *testing.T) {
 	mw := New(Config{ModelAllowlist: []string{"gpt-4o"}})
 	out, err := mw.Invoke(context.Background(), newInput(

proxy/internal/middleware/builtin/llm_request_parser/guardrail_allowlist_test.go

@@ -104,96 +104,3 @@ func TestModelAllowlist_URLRoutedProviders(t *testing.T) {
 		})
 	}
 }
-
-// TestModelAllowlist_VertexRequestShapes replays the Vertex request shapes an
-// Anthropic SDK client (Claude Code) actually sends — the model travels in the
-// URL path, optionally without an "@version" suffix, at the global location —
-// against allowlists holding either the bare catalog id or the raw versioned
-// id. The URLs mirror the shape of a customer-reported request (unversioned
-// model, global location, a project id that itself contains "claude").
-func TestModelAllowlist_VertexRequestShapes(t *testing.T) {
-	const (
-		opusBare      = "/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/claude-opus-4-6:rawPredict"
-		opusBareSSE   = "/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/claude-opus-4-6:streamRawPredict"
-		countTokens   = "/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/count-tokens:rawPredict"
-		messagesBody  = `{"anthropic_version":"vertex-2023-10-16","messages":[{"role":"user","content":"hi"}]}`
-		countOpusBody = `{"model":"claude-opus-4-6","messages":[{"role":"user","content":"hi"}]}`
-	)
-
-	tests := []struct {
-		name      string
-		url       string
-		body      string
-		allowlist []string
-		decision  middleware.Decision
-		denyCode  string
-	}{
-		{
-			name:      "unversioned model allowed by bare catalog entry",
-			url:       opusBare,
-			body:      messagesBody,
-			allowlist: []string{"claude-opus-4-6"},
-			decision:  middleware.DecisionAllow,
-		},
-		{
-			name:      "unversioned model allowed by @version allowlist entry",
-			url:       opusBare,
-			body:      messagesBody,
-			allowlist: []string{"claude-opus-4-6@20250514"},
-			decision:  middleware.DecisionAllow,
-		},
-		{
-			name:      "streaming action allowed the same as rawPredict",
-			url:       opusBareSSE,
-			body:      messagesBody,
-			allowlist: []string{"claude-opus-4-6"},
-			decision:  middleware.DecisionAllow,
-		},
-		{
-			// The original customer report: a Sonnet-only allowlist must block
-			// an Opus request regardless of the id carrying a version suffix.
-			name:      "unversioned model outside the allowlist denied",
-			url:       opusBare,
-			body:      messagesBody,
-			allowlist: []string{"claude-sonnet-4-5"},
-			decision:  middleware.DecisionDeny,
-			denyCode:  "llm_policy.model_blocked",
-		},
-		{
-			name:      "count-tokens resolves the body model and passes when allowed",
-			url:       countTokens,
-			body:      countOpusBody,
-			allowlist: []string{"claude-opus-4-6"},
-			decision:  middleware.DecisionAllow,
-		},
-		{
-			name:      "count-tokens with a disallowed body model denied",
-			url:       countTokens,
-			body:      countOpusBody,
-			allowlist: []string{"claude-sonnet-4-5"},
-			decision:  middleware.DecisionDeny,
-			denyCode:  "llm_policy.model_blocked",
-		},
-		{
-			// No model in the body: the pseudo-model stays and fails closed.
-			name:      "count-tokens without a body model fails closed",
-			url:       countTokens,
-			body:      messagesBody,
-			allowlist: []string{"claude-opus-4-6"},
-			decision:  middleware.DecisionDeny,
-			denyCode:  "llm_policy.model_blocked",
-		},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			out := runParserGuardrail(t, tt.url, []byte(tt.body), tt.allowlist)
-			assert.Equal(t, tt.decision, out.Decision, "unexpected decision for %s", tt.name)
-			if tt.decision == middleware.DecisionDeny {
-				require.NotNil(t, out.DenyReason, "deny reason must be set for %s", tt.name)
-				assert.Equal(t, 403, out.DenyStatus, "deny status must be 403 for %s", tt.name)
-				assert.Equal(t, tt.denyCode, out.DenyReason.Code, "deny code for %s", tt.name)
-			}
-		})
-	}
-}

proxy/internal/middleware/builtin/llm_request_parser/middleware.go

@@ -253,7 +253,9 @@ func parseVertexPath(reqPath string) (vertexRequest, bool) {
 	if c := strings.LastIndex(rest, ":"); c >= 0 {
 		model, action = rest[:c], rest[c+1:]
 	}
-	model = llm.NormalizeVertexModel(model)
+	if at := strings.Index(model, "@"); at >= 0 {
+		model = model[:at]
+	}
 	if model == "" {
 		return vertexRequest{}, false
 	}
@@ -274,44 +276,24 @@ func vertexPublisherVendor(publisher string) string {
 	}
 }
 
-// vertexCountTokensModel is the pseudo-model id of the Vertex token-count
-// endpoint (…/models/count-tokens:rawPredict). It is the one Vertex request
-// shape whose real model travels in the JSON body rather than the URL path.
-const vertexCountTokensModel = "count-tokens"
-
 // invokeVertex emits the model/vendor/session/prompt for a Vertex publisher
 // request, using the publisher's parser to read the (vendor-native) body.
 func (m middlewareImpl) invokeVertex(in *middleware.Input, vx vertexRequest) *middleware.Output {
 	out := &middleware.Output{Decision: middleware.DecisionAllow}
 	vendor := vertexPublisherVendor(vx.publisher)
 
-	var parser llm.Parser
-	if vendor != "" {
-		parser, _ = llm.ParserByName(vendor)
-	}
-
-	model := vx.model
-	// Token counting carries the literal "count-tokens" pseudo-model in the URL
-	// and the real model in the body (Claude Code issues one such call per
-	// request). Resolve the body model so the guardrail allowlist and the
-	// router evaluate the actual model — otherwise every count-tokens call
-	// under an allowlist denies as model_blocked. A body the parser cannot
-	// read keeps the pseudo-model, which fails closed downstream.
-	if model == vertexCountTokensModel && parser != nil {
-		if facts, err := parser.ParseRequest(in.Body); err == nil && facts.Model != "" {
-			if bodyModel := llm.NormalizeVertexModel(facts.Model); bodyModel != "" {
-				model = bodyModel
-			}
-		}
-	}
-
 	md := []middleware.KV{}
 	if vendor != "" {
 		md = append(md, middleware.KV{Key: middleware.KeyLLMProvider, Value: vendor})
 	}
-	md = append(md, middleware.KV{Key: middleware.KeyLLMModel, Value: model})
+	md = append(md, middleware.KV{Key: middleware.KeyLLMModel, Value: vx.model})
 	md = append(md, middleware.KV{Key: middleware.KeyLLMStream, Value: strconv.FormatBool(vx.stream)})
 
+	var parser llm.Parser
+	if vendor != "" {
+		parser, _ = llm.ParserByName(vendor)
+	}
+
 	sessionID := sessionIDFromHeaders(in.Headers)
 	if sessionID == "" && parser != nil {
 		sessionID = parser.ExtractSessionID(in.Body)

proxy/internal/middleware/builtin/llm_router/middleware.go

@@ -564,14 +564,6 @@ func routeClaimsModel(route ProviderRoute, model string) bool {
 		if route.Bedrock && llm.NormalizeBedrockModel(candidate) == model {
 			return true
 		}
-		// Vertex request models likewise reach the router with the "@version"
-		// suffix already stripped by the parser, while the operator may register
-		// the raw versioned id the Google console documents (e.g.
-		// "claude-opus-4-6@20250514"). Normalize the candidate so both spellings
-		// of the same model match.
-		if route.Vertex && llm.NormalizeVertexModel(candidate) == model {
-			return true
-		}
 	}
 	return false
 }

proxy/internal/middleware/builtin/llm_router/vertex_route_test.go

@@ -1,64 +0,0 @@
-package llm_router
-
-import (
-	"context"
-	"testing"
-
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-
-	"github.com/netbirdio/netbird/proxy/internal/middleware"
-)
-
-// TestRouteClaimsModel_VertexNormalizesCandidate guards the Vertex analog of
-// the native-Bedrock routing gap: the request model reaches the router already
-// normalized (the parser strips the "@version" suffix from the URL path), so a
-// provider registered with the raw versioned id must still match.
-func TestRouteClaimsModel_VertexNormalizesCandidate(t *testing.T) {
-	route := ProviderRoute{Vertex: true, Models: []string{"claude-opus-4-6@20250514"}}
-	assert.True(t, routeClaimsModel(route, "claude-opus-4-6"),
-		"raw @version Vertex model must match the normalized request model")
-	assert.False(t, routeClaimsModel(route, "claude-haiku-4-5"),
-		"a model outside the provider's list must not match")
-
-	bare := ProviderRoute{Vertex: true, Models: []string{"claude-opus-4-6"}}
-	assert.True(t, routeClaimsModel(bare, "claude-opus-4-6"),
-		"a bare catalog id keeps matching exactly")
-
-	// Non-Vertex routes keep exact matching for "@"-suffixed candidates.
-	direct := ProviderRoute{Models: []string{"claude-opus-4-6@20250514"}}
-	assert.False(t, routeClaimsModel(direct, "claude-opus-4-6"),
-		"non-Vertex routes must not normalize @version candidates")
-}
-
-// TestRouter_VertexUnversionedModelRoutes replays the customer-reported request
-// end to end through the router: an unversioned model id at the global location
-// (…/models/claude-opus-4-6:rawPredict) must route on a Vertex provider whose
-// models were registered with the raw "@version" ids.
-func TestRouter_VertexUnversionedModelRoutes(t *testing.T) {
-	route := vertexRoute()
-	route.Models = []string{"claude-opus-4-6@20250514", "claude-sonnet-4-5@20250929"}
-	mw := New(Config{Providers: []ProviderRoute{route}})
-
-	in := pathRoutedInput(
-		"/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/claude-opus-4-6:rawPredict",
-		"anthropic",
-		"claude-opus-4-6",
-	)
-	out, err := mw.Invoke(context.Background(), in)
-	require.NoError(t, err)
-	assert.Equal(t, middleware.DecisionAllow, out.Decision,
-		"unversioned request model must route on a provider registered with @version ids")
-
-	denied := pathRoutedInput(
-		"/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/claude-haiku-4-5:rawPredict",
-		"anthropic",
-		"claude-haiku-4-5",
-	)
-	out, err = mw.Invoke(context.Background(), denied)
-	require.NoError(t, err)
-	assert.Equal(t, middleware.DecisionDeny, out.Decision,
-		"a model outside the provider's registered list must still deny")
-	require.NotNil(t, out.DenyReason)
-	assert.Equal(t, denyCodeNotRoutable, out.DenyReason.Code)
-}

proxy/internal/proxy/agent_network_vertex_guardrail_realstack_test.go

@@ -1,291 +0,0 @@
-package proxy_test
-
-import (
-	"context"
-	"net"
-	"net/http"
-	"net/http/httptest"
-	"net/url"
-	"runtime"
-	"strings"
-	"sync/atomic"
-	"testing"
-
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-	"google.golang.org/grpc"
-	"google.golang.org/grpc/credentials/insecure"
-	"google.golang.org/grpc/test/bufconn"
-
-	"github.com/netbirdio/netbird/management/internals/modules/agentnetwork"
-	agentNetworkTypes "github.com/netbirdio/netbird/management/internals/modules/agentnetwork/types"
-	rpservice "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/service"
-	mgmtgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
-	"github.com/netbirdio/netbird/management/server/store"
-	"github.com/netbirdio/netbird/proxy/internal/middleware"
-	"github.com/netbirdio/netbird/proxy/internal/middleware/bodytap"
-	mwbuiltin "github.com/netbirdio/netbird/proxy/internal/middleware/builtin"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/cost_meter"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_guardrail"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_identity_inject"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_limit_check"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_limit_record"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_request_parser"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_response_parser"
-	_ "github.com/netbirdio/netbird/proxy/internal/middleware/builtin/llm_router"
-	"github.com/netbirdio/netbird/proxy/internal/proxy"
-	nbproxytypes "github.com/netbirdio/netbird/proxy/internal/types"
-	"github.com/netbirdio/netbird/shared/management/proto"
-
-	log "github.com/sirupsen/logrus"
-)
-
-// TestReverseProxy_VertexGuardrail_ModelAllowlist drives Anthropic-on-Vertex
-// requests (model in the URL path, not the body — the shape Claude Code sends)
-// through the full synthesized middleware chain against an in-process
-// management stack and a fake Vertex upstream. It reproduces the customer
-// report end to end: with a guardrail allowlisting only Sonnet, an Opus request
-// must be denied by the guardrail (model_blocked) before it reaches the
-// upstream, and the allowed Sonnet request must reach it.
-//
-// Two provider shapes are covered because they exercise different code:
-//
-//   - "catch_all": provider registered with no models (the historical Vertex
-//     default). The router waves every model through, so the guardrail is the
-//     only thing that can block Opus — the direct regression guard for "the
-//     disallowed model is never blocked".
-//   - "versioned_models": provider registered with the raw "@version" model ids
-//     the Google console documents. The router must normalize the registered id
-//     to route the version-stripped request model (otherwise even Sonnet denies
-//     as model_not_routable), and the guardrail must still block Opus.
-//
-// The upstream-hit counter proves whether a denied request actually
-// short-circuited or leaked through to Google.
-func TestReverseProxy_VertexGuardrail_ModelAllowlist(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("sqlite store not supported on Windows")
-	}
-
-	cases := []struct {
-		name           string
-		providerModels []agentNetworkTypes.ProviderModel
-	}{
-		{name: "catch_all", providerModels: nil},
-		{name: "versioned_models", providerModels: []agentNetworkTypes.ProviderModel{
-			{ID: "claude-sonnet-4-5@20250929"},
-			{ID: "claude-opus-4-6@20250514"},
-		}},
-	}
-	for _, tc := range cases {
-		tc := tc
-		t.Run(tc.name, func(t *testing.T) {
-			runVertexGuardrailCase(t, tc.providerModels)
-		})
-	}
-}
-
-func runVertexGuardrailCase(t *testing.T, providerModels []agentNetworkTypes.ProviderModel) {
-	t.Helper()
-
-	const (
-		testAccountID = "acct-vertex-guard-1"
-		testAdminUser = "user-admin-1"
-		adminGroupID  = "grp-admins"
-		providerID    = "prov-vertex-test"
-		guardrailID   = "ainguard-sonnet-only"
-		cluster       = "test.proxy.local"
-		subdomain     = "vertexguard"
-	)
-	testLogger := log.New()
-	testLogger.SetLevel(log.PanicLevel)
-
-	ctx := context.Background()
-
-	// Fake Vertex upstream: any request that reaches it is a guardrail miss for
-	// the disallowed model.
-	var upstreamHits atomic.Int64
-	upstreamBody := []byte(`{"id":"msg_x","type":"message","role":"assistant","model":"claude","content":[{"type":"text","text":"pong"}],"usage":{"input_tokens":5,"output_tokens":2}}`)
-	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		upstreamHits.Add(1)
-		w.Header().Set("Content-Type", "application/json")
-		_, _ = w.Write(upstreamBody)
-	}))
-	t.Cleanup(upstream.Close)
-
-	// In-process management gRPC (bufconn) over a real sqlite store + manager.
-	st, cleanup, err := store.NewTestStoreFromSQL(ctx, "", t.TempDir())
-	require.NoError(t, err, "real sqlite test store must come up")
-	t.Cleanup(cleanup)
-
-	anMgr := agentnetwork.NewManager(st, nil, nil, nil)
-	server := &mgmtgrpc.ProxyServiceServer{}
-	server.SetAgentNetworkLimitsService(anMgr)
-
-	lis := bufconn.Listen(1024 * 1024)
-	srv := grpc.NewServer()
-	proto.RegisterProxyServiceServer(srv, server)
-	go func() { _ = srv.Serve(lis) }()
-	t.Cleanup(srv.Stop)
-
-	conn, err := grpc.NewClient("passthrough:///bufnet",
-		grpc.WithContextDialer(func(_ context.Context, _ string) (net.Conn, error) { return lis.Dial() }),
-		grpc.WithTransportCredentials(insecure.NewCredentials()),
-	)
-	require.NoError(t, err)
-	t.Cleanup(func() { _ = conn.Close() })
-	mgmtClient := proto.NewProxyServiceClient(conn)
-
-	require.NoError(t, st.SaveAgentNetworkSettings(ctx, &agentNetworkTypes.Settings{
-		AccountID: testAccountID,
-		Cluster:   cluster,
-		Subdomain: subdomain,
-	}))
-	require.NoError(t, st.SaveAgentNetworkProvider(ctx, &agentNetworkTypes.Provider{
-		ID:          providerID,
-		AccountID:   testAccountID,
-		ProviderID:  "vertex_ai_api",
-		Name:        "vertex-guard-test",
-		UpstreamURL: upstream.URL,
-		// A static bearer (not "keyfile::…") so the router injects a static auth
-		// header instead of minting a GCP OAuth token at request time — the mint
-		// needs network egress and would deny before the guardrail runs, masking
-		// the guardrail decision this test exists to observe.
-		APIKey:            "static-vertex-token",
-		Enabled:           true,
-		Models:            providerModels,
-		SessionPrivateKey: "priv",
-		SessionPublicKey:  "pub",
-	}))
-	// Guardrail allowlisting ONLY Sonnet.
-	require.NoError(t, st.SaveAgentNetworkGuardrail(ctx, &agentNetworkTypes.Guardrail{
-		ID:        guardrailID,
-		AccountID: testAccountID,
-		Name:      "sonnet-only",
-		Checks: agentNetworkTypes.GuardrailChecks{
-			ModelAllowlist: agentNetworkTypes.GuardrailModelAllowlist{
-				Enabled: true,
-				Models:  []string{"claude-sonnet-4-5"},
-			},
-		},
-	}))
-	require.NoError(t, st.SaveAgentNetworkPolicy(ctx, &agentNetworkTypes.Policy{
-		ID:                     "ainpol-vertex-guard",
-		AccountID:              testAccountID,
-		Name:                   "admins-vertex",
-		Enabled:                true,
-		SourceGroups:           []string{adminGroupID},
-		DestinationProviderIDs: []string{providerID},
-		GuardrailIDs:           []string{guardrailID},
-	}))
-
-	services, err := agentnetwork.SynthesizeServices(ctx, st, testAccountID)
-	require.NoError(t, err)
-	require.Len(t, services, 1, "exactly one synth service expected")
-	synthSvc := services[0]
-	require.NotEmpty(t, synthSvc.Targets, "synth target must exist")
-
-	mwbuiltin.Configure(ctx, t.TempDir(), nil, testLogger, mgmtClient)
-	registry := mwbuiltin.DefaultRegistry()
-	mwMetrics, err := middleware.NewMetrics(nil)
-	require.NoError(t, err)
-	mwMgr := middleware.NewManager(0, mwMetrics, testLogger)
-	mwMgr.SetResolver(middleware.NewResolver(registry))
-
-	specs := make([]middleware.Spec, 0, len(synthSvc.Targets[0].Options.Middlewares))
-	for _, mw := range synthSvc.Targets[0].Options.Middlewares {
-		var slot middleware.Slot
-		switch mw.Slot {
-		case rpservice.MiddlewareSlotOnRequest:
-			slot = middleware.SlotOnRequest
-		case rpservice.MiddlewareSlotOnResponse:
-			slot = middleware.SlotOnResponse
-		case rpservice.MiddlewareSlotTerminal:
-			slot = middleware.SlotTerminal
-		default:
-			t.Fatalf("unknown middleware slot %q on %s", mw.Slot, mw.ID)
-		}
-		specs = append(specs, middleware.Spec{
-			ID:        mw.ID,
-			Slot:      slot,
-			Enabled:   mw.Enabled,
-			FailMode:  middleware.FailOpen,
-			Timeout:   middleware.DefaultTimeout,
-			RawConfig: append([]byte(nil), mw.ConfigJSON...),
-			CanMutate: mw.CanMutate,
-		})
-	}
-
-	serviceIDStr := synthSvc.ID
-	require.NoError(t, mwMgr.Rebuild(serviceIDStr, []middleware.PathTargetBinding{{
-		ServiceID: serviceIDStr,
-		PathID:    "/",
-		Specs:     specs,
-	}}))
-
-	upstreamURL, err := url.Parse(upstream.URL)
-	require.NoError(t, err)
-
-	rp := proxy.NewReverseProxy(http.DefaultTransport, "auto", nil, testLogger, proxy.WithMiddlewareManager(mwMgr))
-	rp.AddMapping(proxy.Mapping{
-		ID:        nbproxytypes.ServiceID(serviceIDStr),
-		AccountID: nbproxytypes.AccountID(testAccountID),
-		Host:      synthSvc.Domain,
-		Paths: map[string]*proxy.PathTarget{
-			"/": {
-				URL:            upstreamURL,
-				DirectUpstream: true,
-				AgentNetwork:   true,
-				Middlewares:    specs,
-				CaptureConfig: &bodytap.Config{
-					MaxRequestBytes:  1 << 20,
-					MaxResponseBytes: 1 << 20,
-					ContentTypes:     []string{"application/json", "text/event-stream"},
-				},
-			},
-		},
-	})
-
-	// vertexBody carries no "model" field — the model lives in the URL path,
-	// exactly as an Anthropic SDK client (Claude Code) posts to Vertex.
-	const vertexBody = `{"anthropic_version":"vertex-2023-10-16","max_tokens":64,"messages":[{"role":"user","content":"Reply with exactly: pong"}]}`
-
-	send := func(t *testing.T, model string) (int, int64, string) {
-		t.Helper()
-		before := upstreamHits.Load()
-		path := "/v1/projects/corp-gcp-it-all-claude/locations/global/publishers/anthropic/models/" + model + ":rawPredict"
-		req := httptest.NewRequest("POST", "https://"+synthSvc.Domain+path, strings.NewReader(vertexBody))
-		req.Host = synthSvc.Domain
-		req.Header.Set("Content-Type", "application/json")
-
-		cd := proxy.NewCapturedData("req-" + model)
-		cd.SetServiceID(nbproxytypes.ServiceID(serviceIDStr))
-		cd.SetAccountID(nbproxytypes.AccountID(testAccountID))
-		cd.SetUserID(testAdminUser)
-		cd.SetUserGroups([]string{adminGroupID})
-		cd.SetAuthMethod("tunnel_peer")
-		req = req.WithContext(proxy.WithCapturedData(req.Context(), cd))
-
-		w := httptest.NewRecorder()
-		rp.ServeHTTP(w, req)
-		return w.Code, upstreamHits.Load() - before, w.Body.String()
-	}
-
-	// The disallowed model (Opus) MUST be denied by the guardrail (model_blocked)
-	// before the upstream — this is the customer-reported bug (Opus let through
-	// under a Sonnet-only allowlist). The deny code distinguishes a guardrail
-	// block from a router model_not_routable, which would deny for the wrong
-	// reason.
-	t.Run("opus_denied_by_guardrail", func(t *testing.T) {
-		code, hits, body := send(t, "claude-opus-4-6")
-		assert.Equal(t, http.StatusForbidden, code, "Opus must be denied under a Sonnet-only allowlist; body=%s", body)
-		assert.Contains(t, body, "llm_policy.model_blocked", "denial must come from the guardrail allowlist, not routing; body=%s", body)
-		assert.Equal(t, int64(0), hits, "a denied request must never reach the Vertex upstream")
-	})
-
-	// The allowed model (Sonnet) passes the guardrail and reaches the upstream.
-	t.Run("sonnet_allowed", func(t *testing.T) {
-		code, hits, body := send(t, "claude-sonnet-4-5")
-		assert.Equal(t, http.StatusOK, code, "Sonnet is allowlisted and must be served; body=%s", body)
-		assert.Equal(t, int64(1), hits, "the allowed request must reach the Vertex upstream exactly once")
-	})
-}

shared/relay/client/fallback_opener.go

@@ -0,0 +1,176 @@
+package client
+
+import (
+	"context"
+	"errors"
+	"net"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+)
+
+const (
+	raceTotalTimeout  = 40 * time.Second
+	raceFallbackDelay = 10 * time.Second
+)
+
+type raceAttempt struct {
+	conn net.Conn
+	err  error
+}
+
+type raceOutcome struct {
+	conn net.Conn
+	err  error
+	done bool
+}
+
+type connRace struct {
+	opener            *FallbackOpener
+	peerKey           string
+	remoteRelayServer RelayServer
+	preferForeign     bool
+
+	raceCtx         context.Context
+	otherCtx        context.Context
+	cancelPreferred context.CancelFunc
+	cancelOther     context.CancelFunc
+	results         chan raceAttempt
+	fallbackTimer   *time.Timer
+
+	otherStarted bool
+	settled      int
+	lastErr      error
+}
+
+type FallbackOpener struct {
+	home         *Client
+	foreignStore *ForeignRelaysStore
+
+	fallbackDelay time.Duration
+	totalTimeout  time.Duration
+	// openFn performs a single attempt. It is overridable in tests; when nil the
+	// real home/foreign dispatch in open is used.
+	openFn func(ctx context.Context, peerKey string, remoteRelayServer RelayServer, foreign bool) raceAttempt
+}
+
+func NewFallbackOpener(home *Client, foreignStore *ForeignRelaysStore) *FallbackOpener {
+	return &FallbackOpener{
+		home:          home,
+		foreignStore:  foreignStore,
+		fallbackDelay: raceFallbackDelay,
+		totalTimeout:  raceTotalTimeout,
+	}
+}
+
+func (r *FallbackOpener) Run(ctx context.Context, peerKey string, remoteRelayServer RelayServer, preferForeign bool) (net.Conn, error) {
+	raceCtx, cancel := context.WithTimeout(ctx, r.totalTimeout)
+	defer cancel()
+
+	preferredCtx, cancelPreferred := context.WithCancel(raceCtx)
+	otherCtx, cancelOther := context.WithCancel(raceCtx)
+
+	race := &connRace{
+		opener:            r,
+		peerKey:           peerKey,
+		remoteRelayServer: remoteRelayServer,
+		preferForeign:     preferForeign,
+		raceCtx:           raceCtx,
+		otherCtx:          otherCtx,
+		cancelPreferred:   cancelPreferred,
+		cancelOther:       cancelOther,
+		results:           make(chan raceAttempt, 2),
+		fallbackTimer:     time.NewTimer(r.fallbackDelay),
+	}
+	defer race.fallbackTimer.Stop()
+
+	go func() {
+		race.results <- r.open(preferredCtx, peerKey, remoteRelayServer, preferForeign)
+	}()
+
+	for {
+		select {
+		case <-race.fallbackTimer.C:
+			race.startOther()
+		case res := <-race.results:
+			if o := race.handleResult(res); o.done {
+				return o.conn, o.err
+			}
+		case <-raceCtx.Done():
+			return race.onTimeout()
+		}
+	}
+}
+
+func (c *connRace) startOther() {
+	if c.otherStarted {
+		return
+	}
+	c.otherStarted = true
+	c.fallbackTimer.Stop()
+	go func() {
+		c.results <- c.opener.open(c.otherCtx, c.peerKey, c.remoteRelayServer, !c.preferForeign)
+	}()
+}
+
+func (c *connRace) handleResult(res raceAttempt) raceOutcome {
+	if (res.err == nil && res.conn != nil) || errors.Is(res.err, ErrConnAlreadyExists) {
+		c.settled++
+		c.stop()
+		return raceOutcome{conn: res.conn, err: res.err, done: true}
+	}
+
+	c.lastErr = res.err
+	c.settled++
+	if !c.otherStarted {
+		c.startOther()
+		return raceOutcome{}
+	}
+	if c.settled == 2 {
+		c.cancelPreferred()
+		c.cancelOther()
+		return raceOutcome{err: c.lastErr, done: true}
+	}
+	return raceOutcome{}
+}
+
+func (c *connRace) onTimeout() (net.Conn, error) {
+	c.stop()
+	if c.lastErr != nil {
+		return nil, c.lastErr
+	}
+	return nil, c.raceCtx.Err()
+}
+
+func (c *connRace) stop() {
+	c.cancelPreferred()
+	c.cancelOther()
+	go c.opener.drainLoser(c.results, c.settled, c.otherStarted)
+}
+
+func (r *FallbackOpener) open(ctx context.Context, peerKey string, remoteRelayServer RelayServer, foreign bool) raceAttempt {
+	if r.openFn != nil {
+		return r.openFn(ctx, peerKey, remoteRelayServer, foreign)
+	}
+	if foreign {
+		conn, err := r.foreignStore.OpenConn(ctx, peerKey, remoteRelayServer)
+		return raceAttempt{conn: conn, err: err}
+	}
+	conn, err := r.home.OpenConn(ctx, peerKey)
+	return raceAttempt{conn: conn, err: err}
+}
+
+func (r *FallbackOpener) drainLoser(results chan raceAttempt, settled int, otherStarted bool) {
+	started := 1
+	if otherStarted {
+		started = 2
+	}
+	for i := settled; i < started; i++ {
+		res := <-results
+		if res.conn != nil {
+			if err := res.conn.Close(); err != nil {
+				log.Debugf("failed to close losing relay connection: %v", err)
+			}
+		}
+	}
+}

shared/relay/client/fallback_opener_test.go

@@ -0,0 +1,450 @@
+package client
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+	"sync"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// The FallbackOpener race is driven by a single goroutine (Run's select loop)
+// with worker goroutines that communicate only through the buffered results
+// channel and the two cancel contexts. These tests exercise that state machine
+// in isolation via an injected openFn, so no relay server or network is needed.
+// Timing is scaled down through the fallbackDelay/totalTimeout fields.
+
+// raceFakeConn tracks whether Close was called. Only Close is exercised by the
+// race logic (drainLoser closes losers; Run returns the winner untouched).
+type raceFakeConn struct {
+	net.Conn
+	label  string
+	closed atomic.Bool
+}
+
+func (c *raceFakeConn) Close() error {
+	c.closed.Store(true)
+	return nil
+}
+
+// raceAttemptScript describes how a single scripted attempt behaves.
+type raceAttemptScript struct {
+	delay time.Duration
+	conn  *raceFakeConn // non-nil => the attempt succeeds and returns this conn
+	err   error         // returned when conn is nil
+	// ignoreCtx makes the attempt complete after delay even if its context is
+	// cancelled. It models an OpenConn that produced a real connection right as
+	// the race cancelled it - exactly the case drainLoser must clean up.
+	ignoreCtx bool
+}
+
+// fakeOpener replaces FallbackOpener.open. Scripts are keyed by the foreign
+// flag, so which script is "preferred" depends on the preferForeign argument
+// passed to Run.
+type fakeOpener struct {
+	mu      sync.Mutex
+	scripts map[bool]raceAttemptScript
+	calls   []bool // foreign flag of each open() invocation, in order
+}
+
+func (f *fakeOpener) open(ctx context.Context, _ string, _ RelayServer, foreign bool) raceAttempt {
+	f.mu.Lock()
+	f.calls = append(f.calls, foreign)
+	s, ok := f.scripts[foreign]
+	f.mu.Unlock()
+	if !ok {
+		return raceAttempt{err: fmt.Errorf("no script for foreign=%v", foreign)}
+	}
+
+	if s.delay > 0 {
+		timer := time.NewTimer(s.delay)
+		defer timer.Stop()
+		if s.ignoreCtx {
+			<-timer.C
+		} else {
+			select {
+			case <-timer.C:
+			case <-ctx.Done():
+				return raceAttempt{err: ctx.Err()}
+			}
+		}
+	}
+
+	if s.conn != nil {
+		return raceAttempt{conn: s.conn}
+	}
+	return raceAttempt{err: s.err}
+}
+
+func (f *fakeOpener) callCount() int {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	return len(f.calls)
+}
+
+func (f *fakeOpener) firstCallForeign(t *testing.T) bool {
+	t.Helper()
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	require.NotEmpty(t, f.calls, "expected at least one open attempt")
+	return f.calls[0]
+}
+
+func newTestOpener(f *fakeOpener, fallbackDelay, totalTimeout time.Duration) *FallbackOpener {
+	o := NewFallbackOpener(nil, nil)
+	o.openFn = f.open
+	o.fallbackDelay = fallbackDelay
+	o.totalTimeout = totalTimeout
+	return o
+}
+
+const (
+	// controller prefers the home relay, i.e. preferForeign == false.
+	preferHome    = false
+	preferForeign = true
+)
+
+var errAttempt = errors.New("attempt failed")
+
+// The preferred attempt wins before the fallback timer fires, so the other
+// attempt is never started.
+func TestFallbackOpener_PreferredWinsImmediately(t *testing.T) {
+	homeConn := &raceFakeConn{label: "home"}
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {conn: homeConn},                            // preferred (home): instant success
+		true:  {delay: 5 * time.Second, conn: foreignConn}, // would never finish in time
+	}}
+	o := newTestOpener(f, 40*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.NoError(t, err)
+	require.Same(t, homeConn, conn)
+	assert.Equal(t, 1, f.callCount(), "other attempt must not start when preferred wins first")
+	assert.False(t, f.firstCallForeign(t), "home must be tried first when preferring home")
+	assert.False(t, foreignConn.closed.Load())
+}
+
+// preferForeign flips which relay is tried first.
+func TestFallbackOpener_PreferForeignRoutesForeignFirst(t *testing.T) {
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		true:  {conn: foreignConn}, // preferred (foreign): instant success
+		false: {delay: 5 * time.Second, conn: &raceFakeConn{}},
+	}}
+	o := newTestOpener(f, 40*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferForeign)
+
+	require.NoError(t, err)
+	require.Same(t, foreignConn, conn)
+	assert.Equal(t, 1, f.callCount())
+	assert.True(t, f.firstCallForeign(t), "foreign must be tried first when preferring foreign")
+}
+
+// ErrConnAlreadyExists counts as success: Run returns it and does not start the
+// other attempt.
+func TestFallbackOpener_ErrConnAlreadyExistsIsSuccess(t *testing.T) {
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {err: ErrConnAlreadyExists},
+		true:  {delay: 5 * time.Second, conn: &raceFakeConn{}},
+	}}
+	o := newTestOpener(f, 40*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.ErrorIs(t, err, ErrConnAlreadyExists)
+	assert.Nil(t, conn)
+	assert.Equal(t, 1, f.callCount(), "other attempt must not start on ErrConnAlreadyExists")
+}
+
+// A preferred failure starts the other attempt immediately, without waiting for
+// the fallback timer.
+func TestFallbackOpener_PreferredFailsStartsOtherBeforeTimer(t *testing.T) {
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {err: errAttempt}, // preferred fails instantly
+		true:  {delay: 5 * time.Millisecond, conn: foreignConn},
+	}}
+	fallbackDelay := 500 * time.Millisecond
+	o := newTestOpener(f, fallbackDelay, 2*time.Second)
+
+	start := time.Now()
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+	elapsed := time.Since(start)
+
+	require.NoError(t, err)
+	require.Same(t, foreignConn, conn)
+	assert.Equal(t, 2, f.callCount())
+	assert.Less(t, elapsed, fallbackDelay/2, "fallback must not wait for the timer after a preferred failure")
+}
+
+// When the preferred attempt is slow, the fallback timer starts the other
+// attempt and its success wins.
+func TestFallbackOpener_TimerStartsOtherWhenPreferredSlow(t *testing.T) {
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {delay: 5 * time.Second}, // preferred hangs until cancelled
+		true:  {delay: 5 * time.Millisecond, conn: foreignConn},
+	}}
+	fallbackDelay := 40 * time.Millisecond
+	o := newTestOpener(f, fallbackDelay, 2*time.Second)
+
+	start := time.Now()
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+	elapsed := time.Since(start)
+
+	require.NoError(t, err)
+	require.Same(t, foreignConn, conn)
+	assert.Equal(t, 2, f.callCount())
+	assert.GreaterOrEqual(t, elapsed, fallbackDelay, "other must not start before the fallback timer fires")
+}
+
+// Both attempts fail: Run returns the last error and tries both relays.
+func TestFallbackOpener_BothFail(t *testing.T) {
+	errOther := errors.New("other failed")
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {err: errAttempt},
+		true:  {err: errOther},
+	}}
+	o := newTestOpener(f, 40*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.Error(t, err)
+	assert.Nil(t, conn)
+	assert.ErrorIs(t, err, errOther, "the most recent error should be surfaced")
+	assert.Equal(t, 2, f.callCount())
+}
+
+// When both attempts succeed, drainLoser must close the losing connection so it
+// is not leaked. Here the preferred attempt wins and the foreign loser - which
+// produced a real conn despite being cancelled - is closed.
+func TestFallbackOpener_DoubleSuccessClosesLoser(t *testing.T) {
+	homeConn := &raceFakeConn{label: "home"}
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {delay: 30 * time.Millisecond, conn: homeConn},                     // preferred wins
+		true:  {delay: 80 * time.Millisecond, conn: foreignConn, ignoreCtx: true}, // loser yields a conn after cancel
+	}}
+	o := newTestOpener(f, 15*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.NoError(t, err)
+	require.Same(t, homeConn, conn)
+	assert.Equal(t, 2, f.callCount())
+	assert.False(t, homeConn.closed.Load(), "the winning connection must not be closed")
+	require.Eventually(t, foreignConn.closed.Load, time.Second, 5*time.Millisecond,
+		"the losing connection must be closed by drainLoser")
+}
+
+// Winner selection is purely by result arrival order, not by preference: when
+// the non-preferred attempt returns first it wins even though home was
+// preferred. This is the mechanism behind the split-relay concern - two peers
+// racing independently have no shared tie-break, so under adversarial timing
+// they can settle on different relays. Documented here as current behavior.
+func TestFallbackOpener_FasterOtherWinsDespitePreference(t *testing.T) {
+	homeConn := &raceFakeConn{label: "home"}
+	foreignConn := &raceFakeConn{label: "foreign"}
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {delay: 60 * time.Millisecond, conn: homeConn, ignoreCtx: true}, // preferred but slower
+		true:  {delay: 5 * time.Millisecond, conn: foreignConn},                // other is faster
+	}}
+	o := newTestOpener(f, 15*time.Millisecond, 2*time.Second)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.NoError(t, err)
+	require.Same(t, foreignConn, conn, "the first successful attempt wins regardless of preference")
+	require.Eventually(t, homeConn.closed.Load, time.Second, 5*time.Millisecond,
+		"the slower preferred attempt becomes the loser and is closed")
+}
+
+// The whole race is bounded by totalTimeout. With no attempt succeeding or
+// failing, Run returns the deadline error.
+func TestFallbackOpener_TotalTimeout(t *testing.T) {
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {delay: 5 * time.Second},
+		true:  {delay: 5 * time.Second},
+	}}
+	o := newTestOpener(f, 20*time.Millisecond, 80*time.Millisecond)
+
+	conn, err := o.Run(context.Background(), "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.Error(t, err)
+	assert.Nil(t, conn)
+	assert.ErrorIs(t, err, context.DeadlineExceeded)
+}
+
+// Cancelling the caller's context aborts the race promptly with the cancel
+// error, even before the fallback timer would fire.
+func TestFallbackOpener_ParentContextCanceled(t *testing.T) {
+	f := &fakeOpener{scripts: map[bool]raceAttemptScript{
+		false: {delay: 5 * time.Second},
+		true:  {delay: 5 * time.Second},
+	}}
+	// fallbackDelay large so the timer never fires; only the parent cancel ends the race.
+	o := newTestOpener(f, 5*time.Second, 5*time.Second)
+
+	ctx, cancel := context.WithCancel(context.Background())
+	go func() {
+		time.Sleep(20 * time.Millisecond)
+		cancel()
+	}()
+
+	start := time.Now()
+	conn, err := o.Run(ctx, "peer", RelayServer{Addr: "srv"}, preferHome)
+
+	require.Error(t, err)
+	assert.Nil(t, conn)
+	assert.ErrorIs(t, err, context.Canceled)
+	assert.Less(t, time.Since(start), time.Second, "must return shortly after the parent context is cancelled")
+	assert.Equal(t, 1, f.callCount(), "the other attempt must not start")
+}
+
+// rendezvous models the relay-level requirement that a relayed connection is
+// established only once BOTH peers subscribe to the same relay server. arrive
+// records a peer's presence on a relay and returns a channel that closes when
+// the second peer arrives, so an attempt can only complete after a real
+// rendezvous - the same coupling the production code depends on.
+type rendezvous struct {
+	mu       sync.Mutex
+	arrivals map[string]int
+	gates    map[string]chan struct{}
+}
+
+func newRendezvous() *rendezvous {
+	return &rendezvous{arrivals: map[string]int{}, gates: map[string]chan struct{}{}}
+}
+
+func (r *rendezvous) arrive(relay string) <-chan struct{} {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	g, ok := r.gates[relay]
+	if !ok {
+		g = make(chan struct{})
+		r.gates[relay] = g
+	}
+	r.arrivals[relay]++
+	if r.arrivals[relay] == 2 {
+		close(g)
+	}
+	return g
+}
+
+// splitPeer is one peer's view of the two relays. relayFor maps the foreign
+// flag to a relay name; postDelay is how long after the rendezvous that peer's
+// OpenConn takes to return (its per-relay subscribe latency). Different values
+// per peer model the asymmetric timing that triggers finding #1.
+type splitPeer struct {
+	rv        *rendezvous
+	relayFor  map[bool]string
+	postDelay map[string]time.Duration
+}
+
+func (p *splitPeer) open(ctx context.Context, _ string, _ RelayServer, foreign bool) raceAttempt {
+	relay := p.relayFor[foreign]
+
+	select {
+	case <-p.rv.arrive(relay):
+	case <-ctx.Done():
+		return raceAttempt{err: ctx.Err()}
+	}
+
+	timer := time.NewTimer(p.postDelay[relay])
+	defer timer.Stop()
+	select {
+	case <-timer.C:
+	case <-ctx.Done():
+		return raceAttempt{err: ctx.Err()}
+	}
+	return raceAttempt{conn: &raceFakeConn{label: relay}}
+}
+
+// TestFallbackOpener_SplitRelaySelection reproduces finding #1: the two peers
+// run FallbackOpener.Run independently with no shared tie-break, so the winner
+// is chosen purely by local result-arrival order. Under an adversarial - but
+// self-consistent - timing profile they settle on DIFFERENT relays.
+//
+// Both peers prefer relayA (the controller's home). The split needs each peer's
+// preferred relayA to be slow enough that both start their fallback (so both
+// relays actually rendezvous), and then each peer's fast path to be a different
+// relay:
+//   - peerA: relayA slow (abandoned), relayB fast  -> peerA wins relayB
+//   - peerB: relayA fast (wins),      relayB slow  -> peerB wins relayA
+//
+// Each winner then cancels its attempt on the relay the OTHER peer actually
+// kept, leaving two half-open relayed connections that were both reported as
+// successful. When a deterministic cross-peer tie-break is added to fix this,
+// invert the assertion below to require convergence.
+func TestFallbackOpener_SplitRelaySelection(t *testing.T) {
+	const (
+		relayA = "relayA" // controller's home relay; both peers prefer it
+		relayB = "relayB" // non-controller's home relay
+	)
+	rv := newRendezvous()
+
+	peerA := &splitPeer{
+		rv:       rv,
+		relayFor: map[bool]string{false: relayA, true: relayB}, // home=relayA
+		postDelay: map[string]time.Duration{
+			relayA: 500 * time.Millisecond, // preferred but slow -> abandoned
+			relayB: 10 * time.Millisecond,  // fallback is fast -> peerA wins relayB
+		},
+	}
+	peerB := &splitPeer{
+		rv:       rv,
+		relayFor: map[bool]string{false: relayB, true: relayA}, // home=relayB
+		postDelay: map[string]time.Duration{
+			relayA: 80 * time.Millisecond,  // preferred, wins - but only after starting fallback
+			relayB: 500 * time.Millisecond, // fallback (home) is slow -> abandoned
+		},
+	}
+
+	fallbackDelay := 30 * time.Millisecond
+	newPeerOpener := func(p *splitPeer) *FallbackOpener {
+		o := NewFallbackOpener(nil, nil)
+		o.openFn = p.open
+		o.fallbackDelay = fallbackDelay
+		o.totalTimeout = 5 * time.Second
+		return o
+	}
+	oA := newPeerOpener(peerA)
+	oB := newPeerOpener(peerB)
+
+	type result struct {
+		conn net.Conn
+		err  error
+	}
+	var ra, rb result
+	var wg sync.WaitGroup
+	wg.Add(2)
+	go func() {
+		defer wg.Done()
+		ra.conn, ra.err = oA.Run(context.Background(), "peerB", RelayServer{Addr: relayB}, preferHome)
+	}()
+	go func() {
+		defer wg.Done()
+		rb.conn, rb.err = oB.Run(context.Background(), "peerA", RelayServer{Addr: relayA}, preferForeign)
+	}()
+	wg.Wait()
+
+	require.NoError(t, ra.err)
+	require.NoError(t, rb.err)
+	aRelay := ra.conn.(*raceFakeConn).label
+	bRelay := rb.conn.(*raceFakeConn).label
+	t.Logf("peerA settled on %s, peerB settled on %s", aRelay, bRelay)
+
+	assert.Equal(t, aRelay, bRelay,
+		"peers selected different relays with no cross-peer tie-break")
+	assert.Equal(t, relayB, aRelay, "peerA abandoned its slow preferred relay and won the fallback")
+	assert.Equal(t, relayA, bRelay, "peerB won its preferred relay after starting the fallback")
+}

shared/relay/client/foreign_relays_store.go

@@ -0,0 +1,155 @@
+package client
+
+import (
+	"context"
+	"net"
+	"sync"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/sync/singleflight"
+
+	relayAuth "github.com/netbirdio/netbird/shared/relay/auth/hmac"
+)
+
+type foreignRelay struct {
+	client  *Client
+	created time.Time
+	inUse   int
+}
+
+type ForeignRelaysStore struct {
+	mu      sync.RWMutex
+	clients map[string]*foreignRelay
+
+	group singleflight.Group
+
+	ctx                  context.Context
+	tokenStore           *relayAuth.TokenStore
+	peerID               string
+	mtu                  uint16
+	transportFallback    *transportFallback
+	onDisconnect         func(string)
+	keepUnusedServerTime time.Duration
+}
+
+func NewForeignRelaysStore(ctx context.Context, tokenStore *relayAuth.TokenStore, peerID string, mtu uint16, transportFallback *transportFallback, onDisconnect func(string), keepUnusedServerTime time.Duration) *ForeignRelaysStore {
+	return &ForeignRelaysStore{
+		clients:              make(map[string]*foreignRelay),
+		ctx:                  ctx,
+		tokenStore:           tokenStore,
+		peerID:               peerID,
+		mtu:                  mtu,
+		transportFallback:    transportFallback,
+		onDisconnect:         onDisconnect,
+		keepUnusedServerTime: keepUnusedServerTime,
+	}
+}
+
+func (f *ForeignRelaysStore) OpenConn(ctx context.Context, peerKey string, remoteRelayServer RelayServer) (net.Conn, error) {
+	fr, err := f.acquire(remoteRelayServer)
+	if err != nil {
+		return nil, err
+	}
+	defer f.release(fr)
+
+	return fr.client.OpenConn(ctx, peerKey)
+}
+
+func (f *ForeignRelaysStore) acquire(remoteRelayServer RelayServer) (*foreignRelay, error) {
+	f.mu.Lock()
+	if fr, ok := f.clients[remoteRelayServer.Addr]; ok {
+		fr.inUse++
+		f.mu.Unlock()
+		return fr, nil
+	}
+	f.mu.Unlock()
+
+	v, err, _ := f.group.Do(remoteRelayServer.Addr, func() (any, error) {
+		f.mu.RLock()
+		fr, ok := f.clients[remoteRelayServer.Addr]
+		f.mu.RUnlock()
+		if ok {
+			return fr, nil
+		}
+
+		relayClient := NewClientWithServerIP(remoteRelayServer.Addr, remoteRelayServer.IP, f.tokenStore, f.peerID, f.mtu)
+		relayClient.SetTransportFallback(f.transportFallback)
+		if err := relayClient.Connect(f.ctx); err != nil {
+			return nil, err
+		}
+		relayClient.SetOnDisconnectListener(f.onDisconnect)
+
+		f.mu.Lock()
+		fr = &foreignRelay{client: relayClient, created: time.Now()}
+		f.clients[remoteRelayServer.Addr] = fr
+		f.mu.Unlock()
+		return fr, nil
+	})
+	if err != nil {
+		return nil, err
+	}
+
+	fr := v.(*foreignRelay)
+	f.mu.Lock()
+	if cur, ok := f.clients[remoteRelayServer.Addr]; !ok || cur != fr {
+		f.mu.Unlock()
+		return f.acquire(remoteRelayServer)
+	}
+	fr.inUse++
+	f.mu.Unlock()
+	return fr, nil
+}
+
+func (f *ForeignRelaysStore) release(fr *foreignRelay) {
+	f.mu.Lock()
+	fr.inUse--
+	f.mu.Unlock()
+}
+
+func (f *ForeignRelaysStore) evict(serverAddress string) {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	if _, ok := f.clients[serverAddress]; ok {
+		delete(f.clients, serverAddress)
+		log.Debugf("evicted disconnected foreign relay client: %s", serverAddress)
+	}
+}
+
+func (f *ForeignRelaysStore) cleanupUnused() {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+
+	for addr, fr := range f.clients {
+		if time.Since(fr.created) <= f.keepUnusedServerTime {
+			continue
+		}
+		if fr.inUse > 0 {
+			continue
+		}
+		if fr.client.HasConns() {
+			continue
+		}
+		fr.client.SetOnDisconnectListener(nil)
+		go func() {
+			_ = fr.client.Close()
+		}()
+		log.Debugf("clean up unused relay server connection: %s", addr)
+		delete(f.clients, addr)
+	}
+}
+
+func (f *ForeignRelaysStore) states() []RelayConnState {
+	f.mu.RLock()
+	clients := make([]*Client, 0, len(f.clients))
+	for _, fr := range f.clients {
+		clients = append(clients, fr.client)
+	}
+	f.mu.RUnlock()
+
+	states := make([]RelayConnState, 0, len(clients))
+	for _, c := range clients {
+		states = append(states, relayConnState(c))
+	}
+	return states
+}

shared/relay/client/manager.go

@@ -22,27 +22,6 @@ var (
 	ErrRelayClientNotConnected = fmt.Errorf("relay client not connected")
 )
 
-// RelayTrack hold the relay clients for the foreign relay servers.
-// With the mutex can ensure we can open new connection in case the relay connection has been established with
-// the relay server.
-type RelayTrack struct {
-	sync.RWMutex
-	relayClient *Client
-	err         error
-	created     time.Time
-	// ready is closed once the dial started by openConnVia finishes (relayClient
-	// or err is set). Callers reusing a track wait on this instead of the track
-	// lock, so the dial never runs under rt.Lock.
-	ready chan struct{}
-}
-
-func NewRelayTrack() *RelayTrack {
-	return &RelayTrack{
-		created: time.Now(),
-		ready:   make(chan struct{}),
-	}
-}
-
 type OnServerCloseListener func()
 
 // ManagerOption configures a Manager at construction time.
@@ -59,6 +38,11 @@ type RelayConnState struct {
 	Err error
 }
 
+type RelayServer struct {
+	Addr string
+	IP   netip.Addr
+}
+
 // WithMaxBackoffInterval caps the exponential backoff between reconnect
 // attempts to the home relay. A non-positive value keeps the default.
 func WithMaxBackoffInterval(d time.Duration) ManagerOption {
@@ -83,8 +67,7 @@ type Manager struct {
 	relayClientMu  sync.RWMutex
 	reconnectGuard *Guard
 
-	relayClients      map[string]*RelayTrack
-	relayClientsMutex sync.RWMutex
+	foreign *ForeignRelaysStore
 
 	onDisconnectedListeners map[string]*list.List
 	onReconnectedListenerFn func()
@@ -120,7 +103,6 @@ func NewManager(ctx context.Context, serverURLs []string, peerID string, mtu uin
 			ConnectionTimeout: defaultConnectionTimeout,
 			TransportFallback: tf,
 		},
-		relayClients:            make(map[string]*RelayTrack),
 		onDisconnectedListeners: make(map[string]*list.List),
 		cleanupInterval:         relayCleanupInterval,
 		keepUnusedServerTime:    keepUnusedServerTime,
@@ -128,6 +110,7 @@ func NewManager(ctx context.Context, serverURLs []string, peerID string, mtu uin
 	for _, opt := range opts {
 		opt(m)
 	}
+	m.foreign = NewForeignRelaysStore(ctx, tokenStore, peerID, mtu, tf, m.onServerDisconnected, m.keepUnusedServerTime)
 	m.serverPicker.ServerURLs.Store(serverURLs)
 	m.reconnectGuard = NewGuard(m.serverPicker, m.maxBackoffInterval)
 	return m
@@ -159,40 +142,26 @@ func (m *Manager) Serve() error {
 	return err
 }
 
-// OpenConn opens a connection to the given peer key. If the peer is on the same relay server, the connection will be
-// established via the relay server. If the peer is on a different relay server, the manager will establish a new
-// connection to the relay server. It returns back with a net.Conn what represent the remote peer connection.
-//
-// serverIP, when valid and serverAddress is foreign, is used as a dial target if the FQDN-based dial fails.
-// Ignored for the local home-server path. TLS verification still uses the FQDN via SNI.
-func (m *Manager) OpenConn(ctx context.Context, serverAddress, peerKey string, serverIP netip.Addr) (net.Conn, error) {
+func (m *Manager) OpenConn(ctx context.Context, remoteRelayServer RelayServer, peerKey string, preferForeign bool) (net.Conn, error) {
 	m.relayClientMu.RLock()
-	defer m.relayClientMu.RUnlock()
+	relayClient := m.relayClient
+	m.relayClientMu.RUnlock()
 
-	if m.relayClient == nil {
+	if relayClient == nil {
 		return nil, ErrRelayClientNotConnected
 	}
 
-	foreign, err := m.isForeignServer(serverAddress)
+	foreign, err := m.isForeignServer(relayClient, remoteRelayServer.Addr)
 	if err != nil {
 		return nil, err
 	}
 
-	var (
-		netConn net.Conn
-	)
 	if !foreign {
-		log.Debugf("open peer connection via permanent server: %s", peerKey)
-		netConn, err = m.relayClient.OpenConn(ctx, peerKey)
-	} else {
-		log.Debugf("open peer connection via foreign server: %s", serverAddress)
-		netConn, err = m.openConnVia(ctx, serverAddress, peerKey, serverIP)
-	}
-	if err != nil {
-		return nil, err
+		return relayClient.OpenConn(ctx, peerKey)
 	}
 
-	return netConn, err
+	opener := NewFallbackOpener(relayClient, m.foreign)
+	return opener.Run(ctx, peerKey, remoteRelayServer, preferForeign)
 }
 
 // Ready returns true if the home Relay client is connected to the relay server.
@@ -223,7 +192,7 @@ func (m *Manager) AddCloseListener(serverAddress string, onClosedListener OnServ
 		return ErrRelayClientNotConnected
 	}
 
-	foreign, err := m.isForeignServer(serverAddress)
+	foreign, err := m.isForeignServer(m.relayClient, serverAddress)
 	if err != nil {
 		return err
 	}
@@ -287,26 +256,7 @@ func (m *Manager) RelayStates() []RelayConnState {
 		states = append(states, st)
 	}
 
-	// Snapshot the tracks, then query each outside the map lock: a track can be
-	// held by an in-progress Connect, and blocking on it must not stall other
-	// relay operations.
-	m.relayClientsMutex.RLock()
-	tracks := make([]*RelayTrack, 0, len(m.relayClients))
-	for _, rt := range m.relayClients {
-		tracks = append(tracks, rt)
-	}
-	m.relayClientsMutex.RUnlock()
-
-	// Only connected foreign relays carry state; a failed connect is evicted
-	// immediately (openConnVia), so there is no error state to surface.
-	for _, rt := range tracks {
-		rt.RLock()
-		rc := rt.relayClient
-		rt.RUnlock()
-		if rc != nil {
-			states = append(states, relayConnState(rc))
-		}
-	}
+	states = append(states, m.foreign.states()...)
 
 	return states
 }
@@ -327,76 +277,6 @@ func (m *Manager) UpdateToken(token *relayAuth.Token) error {
 	return m.tokenStore.UpdateToken(token)
 }
 
-func (m *Manager) openConnVia(ctx context.Context, serverAddress, peerKey string, serverIP netip.Addr) (net.Conn, error) {
-	// check if already has a connection to the desired relay server
-	m.relayClientsMutex.RLock()
-	rt, ok := m.relayClients[serverAddress]
-	m.relayClientsMutex.RUnlock()
-	if ok {
-		return m.openConnOnTrack(ctx, rt, peerKey)
-	}
-
-	// if not, establish a new connection but check it again (because changed the lock type) before starting the
-	// connection
-	m.relayClientsMutex.Lock()
-	rt, ok = m.relayClients[serverAddress]
-	if ok {
-		m.relayClientsMutex.Unlock()
-		return m.openConnOnTrack(ctx, rt, peerKey)
-	}
-
-	// Publish the track and release the map lock BEFORE dialing, so the dial does
-	// not run under rt.Lock (which would block RelayStates and the cleanup loop
-	// for the full dial). Concurrent callers find this track and wait on rt.ready.
-	rt = NewRelayTrack()
-	m.relayClients[serverAddress] = rt
-	m.relayClientsMutex.Unlock()
-
-	relayClient := NewClientWithServerIP(serverAddress, serverIP, m.tokenStore, m.peerID, m.mtu)
-	relayClient.SetTransportFallback(m.transportFallback)
-	err := relayClient.Connect(m.ctx)
-	if err != nil {
-		rt.Lock()
-		rt.err = err
-		rt.Unlock()
-		close(rt.ready)
-		m.relayClientsMutex.Lock()
-		delete(m.relayClients, serverAddress)
-		m.relayClientsMutex.Unlock()
-		return nil, err
-	}
-	// if connection closed then delete the relay client from the list
-	relayClient.SetOnDisconnectListener(m.onServerDisconnected)
-	rt.Lock()
-	rt.relayClient = relayClient
-	rt.Unlock()
-	close(rt.ready)
-
-	return relayClient.OpenConn(ctx, peerKey)
-}
-
-// openConnOnTrack opens a peer connection through an existing relay track,
-// waiting for the dial started by another openConnVia call to finish. It waits
-// on rt.ready rather than the track lock, so it neither holds nor contends the
-// track lock across the dial.
-func (m *Manager) openConnOnTrack(ctx context.Context, rt *RelayTrack, peerKey string) (net.Conn, error) {
-	select {
-	case <-rt.ready:
-	case <-ctx.Done():
-		return nil, ctx.Err()
-	}
-
-	rt.RLock()
-	defer rt.RUnlock()
-	if rt.err != nil {
-		return nil, rt.err
-	}
-	if rt.relayClient == nil {
-		return nil, ErrRelayClientNotConnected
-	}
-	return rt.relayClient.OpenConn(ctx, peerKey)
-}
-
 func (m *Manager) onServerConnected() {
 	m.listenerLock.Lock()
 	defer m.listenerLock.Unlock()
@@ -422,21 +302,12 @@ func (m *Manager) onServerDisconnected(serverAddress string) {
 	m.relayClientMu.Unlock()
 
 	if !isHome {
-		m.evictForeignRelay(serverAddress)
+		m.foreign.evict(serverAddress)
 	}
 
 	m.notifyOnDisconnectListeners(serverAddress)
 }
 
-func (m *Manager) evictForeignRelay(serverAddress string) {
-	m.relayClientsMutex.Lock()
-	defer m.relayClientsMutex.Unlock()
-	if _, ok := m.relayClients[serverAddress]; ok {
-		delete(m.relayClients, serverAddress)
-		log.Debugf("evicted disconnected foreign relay client: %s", serverAddress)
-	}
-}
-
 func (m *Manager) listenGuardEvent(ctx context.Context) {
 	for {
 		select {
@@ -459,8 +330,8 @@ func (m *Manager) storeClient(client *Client) {
 	m.relayClient.SetOnDisconnectListener(m.onServerDisconnected)
 }
 
-func (m *Manager) isForeignServer(address string) (bool, error) {
-	rAddr, err := m.relayClient.ServerInstanceURL()
+func (m *Manager) isForeignServer(relayClient *Client, address string) (bool, error) {
+	rAddr, err := relayClient.ServerInstanceURL()
 	if err != nil {
 		return false, fmt.Errorf("relay client not connected")
 	}
@@ -475,50 +346,11 @@ func (m *Manager) startCleanupLoop() {
 		case <-m.ctx.Done():
 			return
 		case <-ticker.C:
-			m.cleanUpUnusedRelays()
+			m.foreign.cleanupUnused()
 		}
 	}
 }
 
-func (m *Manager) cleanUpUnusedRelays() {
-	m.relayClientsMutex.Lock()
-	defer m.relayClientsMutex.Unlock()
-
-	for addr, rt := range m.relayClients {
-		rt.Lock()
-		// if the connection failed to the server the relay client will be nil
-		// but the instance will be kept in the relayClients until the next locking
-		if rt.err != nil {
-			rt.Unlock()
-			continue
-		}
-
-		// dial still in progress (openConnVia publishes the track before Connect
-		// completes and no longer holds rt.Lock during it), nothing to clean up.
-		if rt.relayClient == nil {
-			rt.Unlock()
-			continue
-		}
-
-		if time.Since(rt.created) <= m.keepUnusedServerTime {
-			rt.Unlock()
-			continue
-		}
-
-		if rt.relayClient.HasConns() {
-			rt.Unlock()
-			continue
-		}
-		rt.relayClient.SetOnDisconnectListener(nil)
-		go func() {
-			_ = rt.relayClient.Close()
-		}()
-		log.Debugf("clean up unused relay server connection: %s", addr)
-		delete(m.relayClients, addr)
-		rt.Unlock()
-	}
-}
-
 func (m *Manager) addListener(serverAddress string, onClosedListener OnServerCloseListener) {
 	m.listenerLock.Lock()
 	defer m.listenerLock.Unlock()

shared/relay/client/manager_cleanup_test.go

@@ -2,17 +2,14 @@ package client
 
 import (
 	"context"
-	"net/netip"
 	"testing"
 	"time"
-
-	"github.com/stretchr/testify/require"
 )
 
 // TestCleanUpUnusedRelays_DoesNotBlockOnRealHangingDial drives a real, hanging foreign
-// relay dial and asserts cleanUpUnusedRelays does not stall behind it.
+// relay dial and asserts the foreign store cleanup does not stall behind it.
 func TestCleanUpUnusedRelays_DoesNotBlockOnRealHangingDial(t *testing.T) {
-	serverAddr := stallingRelayListener(t)
+	serverAddr, accepted := stallingRelayListener(t)
 
 	mCtx, mCancel := context.WithCancel(context.Background())
 	t.Cleanup(mCancel)
@@ -22,39 +19,32 @@ func TestCleanUpUnusedRelays_DoesNotBlockOnRealHangingDial(t *testing.T) {
 	dialDone := make(chan struct{})
 	go func() {
 		defer close(dialDone)
-		_, _ = m.openConnVia(mCtx, serverAddr, "peerKey", netip.Addr{})
+		_, _ = m.foreign.OpenConn(mCtx, "peerKey", RelayServer{Addr: serverAddr})
 	}()
 
-	// The track appears in the map once the dial is in flight.
-	require.Eventually(t, func() bool {
-		m.relayClientsMutex.RLock()
-		defer m.relayClientsMutex.RUnlock()
-		_, ok := m.relayClients[serverAddr]
-		return ok
-	}, 5*time.Second, 5*time.Millisecond, "relay dial did not start")
+	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.cleanUpUnusedRelays()
+		m.foreign.cleanupUnused()
 	}()
 
 	select {
 	case <-cleanupDone:
 	case <-time.After(2 * time.Second):
-		t.Fatal("cleanUpUnusedRelays blocked on an in-progress relay dial while holding the relay map lock")
+		t.Fatal("cleanupUnused blocked on an in-progress relay dial")
 	}
 
-	m.relayClientsMutex.RLock()
-	_, stillTracked := m.relayClients[serverAddr]
-	m.relayClientsMutex.RUnlock()
-	require.True(t, stillTracked, "an in-progress relay dial must not be evicted by cleanup")
-
 	// Release the hanging dial so the goroutine can exit cleanly.
 	mCancel()
 	select {
 	case <-dialDone:
 	case <-time.After(5 * time.Second):
-		t.Fatal("openConnVia did not return after context cancellation")
+		t.Fatal("foreign OpenConn did not return after context cancellation")
 	}
 }

shared/relay/client/manager_relaystates_test.go

@@ -3,7 +3,6 @@ package client
 import (
 	"context"
 	"net"
-	"net/netip"
 	"sync"
 	"testing"
 	"time"
@@ -13,13 +12,16 @@ import (
 
 // stallingRelayListener accepts TCP connections and holds them open without ever
 // responding, so a relay handshake dialed against it blocks until its context is
-// cancelled. It returns the "rel://host:port" URL to dial.
-func stallingRelayListener(t *testing.T) string {
+// 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() {
@@ -31,6 +33,10 @@ func stallingRelayListener(t *testing.T) string {
 			mu.Lock()
 			conns = append(conns, c)
 			mu.Unlock()
+			select {
+			case accepted <- struct{}{}:
+			default:
+			}
 		}
 	}()
 	t.Cleanup(func() {
@@ -42,14 +48,14 @@ func stallingRelayListener(t *testing.T) string {
 		mu.Unlock()
 	})
 
-	return "rel://" + ln.Addr().String()
+	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
-// relay dial.
+// foreign relay dial.
 func TestRelayStates_DoesNotBlockOnRealHangingDial(t *testing.T) {
-	serverAddr := stallingRelayListener(t)
+	serverAddr, accepted := stallingRelayListener(t)
 
 	mCtx, mCancel := context.WithCancel(context.Background())
 	t.Cleanup(mCancel)
@@ -59,15 +65,14 @@ func TestRelayStates_DoesNotBlockOnRealHangingDial(t *testing.T) {
 	dialDone := make(chan struct{})
 	go func() {
 		defer close(dialDone)
-		_, _ = m.openConnVia(mCtx, serverAddr, "peerKey", netip.Addr{})
+		_, _ = m.foreign.OpenConn(mCtx, "peerKey", RelayServer{Addr: serverAddr})
 	}()
 
-	require.Eventually(t, func() bool {
-		m.relayClientsMutex.RLock()
-		defer m.relayClientsMutex.RUnlock()
-		_, ok := m.relayClients[serverAddr]
-		return ok
-	}, 5*time.Second, 5*time.Millisecond, "relay dial did not start")
+	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() {
@@ -86,6 +91,6 @@ func TestRelayStates_DoesNotBlockOnRealHangingDial(t *testing.T) {
 	select {
 	case <-dialDone:
 	case <-time.After(5 * time.Second):
-		t.Fatal("openConnVia did not return after context cancellation")
+		t.Fatal("foreign OpenConn did not return after context cancellation")
 	}
 }

shared/relay/client/manager_serverip_test.go

@@ -3,7 +3,6 @@ package client
 import (
 	"context"
 	"io"
-	"net/netip"
 	"testing"
 	"time"
 
@@ -85,7 +84,7 @@ func TestManager_ForeignRelayServerIP(t *testing.T) {
 	t.Run("no server IP, dial fails", func(t *testing.T) {
 		dialCtx, dialCancel := context.WithTimeout(ctx, 5*time.Second)
 		defer dialCancel()
-		_, err := mgrAlice.OpenConn(dialCtx, brokenFQDN, "bob", netip.Addr{})
+		_, err := mgrAlice.OpenConn(dialCtx, RelayServer{Addr: brokenFQDN}, "bob", true)
 		if err == nil {
 			t.Fatalf("expected OpenConn to fail without server IP, got success")
 		}
@@ -95,7 +94,7 @@ func TestManager_ForeignRelayServerIP(t *testing.T) {
 		// Bob waits for Alice's incoming peer connection on his side.
 		bobSideCh := make(chan error, 1)
 		go func() {
-			conn, err := mgrBob.OpenConn(ctx, bobRealAddr, "alice", netip.Addr{})
+			conn, err := mgrBob.OpenConn(ctx, RelayServer{Addr: bobRealAddr}, "alice", false)
 			if err != nil {
 				bobSideCh <- err
 				return
@@ -113,7 +112,7 @@ func TestManager_ForeignRelayServerIP(t *testing.T) {
 			bobSideCh <- nil
 		}()
 
-		aliceConn, err := mgrAlice.OpenConn(ctx, brokenFQDN, "bob", bobAdvertisedIP)
+		aliceConn, err := mgrAlice.OpenConn(ctx, RelayServer{Addr: brokenFQDN, IP: bobAdvertisedIP}, "bob", true)
 		if err != nil {
 			t.Fatalf("alice OpenConn with server IP: %s", err)
 		}

shared/relay/client/manager_test.go

@@ -3,7 +3,6 @@ package client
 import (
 	"context"
 	"fmt"
-	"net/netip"
 	"testing"
 	"time"
 
@@ -106,11 +105,11 @@ func TestForeignConn(t *testing.T) {
 	if err != nil {
 		t.Fatalf("failed to get relay address: %s", err)
 	}
-	connAliceToBob, err := clientAlice.OpenConn(ctx, bobsSrvAddr, "bob", netip.Addr{})
+	connAliceToBob, err := clientAlice.OpenConn(ctx, RelayServer{Addr: bobsSrvAddr}, "bob", true)
 	if err != nil {
 		t.Fatalf("failed to bind channel: %s", err)
 	}
-	connBobToAlice, err := clientBob.OpenConn(ctx, bobsSrvAddr, "alice", netip.Addr{})
+	connBobToAlice, err := clientBob.OpenConn(ctx, RelayServer{Addr: bobsSrvAddr}, "alice", false)
 	if err != nil {
 		t.Fatalf("failed to bind channel: %s", err)
 	}
@@ -210,7 +209,7 @@ func TestForeginConnClose(t *testing.T) {
 	if err != nil {
 		t.Fatalf("failed to serve manager: %s", err)
 	}
-	conn, err := mgr.OpenConn(ctx, toURL(srvCfg2)[0], "bob", netip.Addr{})
+	conn, err := mgr.OpenConn(ctx, RelayServer{Addr: toURL(srvCfg2)[0]}, "bob", true)
 	if err != nil {
 		t.Fatalf("failed to bind channel: %s", err)
 	}
@@ -302,7 +301,7 @@ func TestForeignAutoClose(t *testing.T) {
 	}
 
 	t.Log("open connection to another peer")
-	if _, err = mgr.OpenConn(ctx, foreignServerURL, "anotherpeer", netip.Addr{}); err == nil {
+	if _, err = mgr.OpenConn(ctx, RelayServer{Addr: foreignServerURL}, "anotherpeer", true); err == nil {
 		t.Fatalf("should have failed to open connection to another peer")
 	}
 
@@ -372,7 +371,7 @@ func TestAutoReconnect(t *testing.T) {
 	if err != nil {
 		t.Errorf("failed to get relay address: %s", err)
 	}
-	conn, err := clientAlice.OpenConn(ctx, ra, "bob", netip.Addr{})
+	conn, err := clientAlice.OpenConn(ctx, RelayServer{Addr: ra}, "bob", false)
 	if err != nil {
 		t.Errorf("failed to bind channel: %s", err)
 	}
@@ -392,7 +391,7 @@ func TestAutoReconnect(t *testing.T) {
 	}
 
 	log.Infof("reopent the connection")
-	_, err = clientAlice.OpenConn(ctx, ra, "bob", netip.Addr{})
+	_, err = clientAlice.OpenConn(ctx, RelayServer{Addr: ra}, "bob", false)
 	if err != nil {
 		t.Errorf("failed to open channel: %s", err)
 	}
@@ -454,7 +453,7 @@ func TestNotifierDoubleAdd(t *testing.T) {
 		t.Fatalf("failed to serve manager: %s", err)
 	}
 
-	conn1, err := clientAlice.OpenConn(ctx, clientAlice.ServerURLs()[0], "bob", netip.Addr{})
+	conn1, err := clientAlice.OpenConn(ctx, RelayServer{Addr: clientAlice.ServerURLs()[0]}, "bob", false)
 	if err != nil {
 		t.Fatalf("failed to bind channel: %s", err)
 	}