[client] Hand off dialed connections to the sweeper atomically

WrapDialContext and WrapConn registered the dial and the connection
independently, so a sweep landing between the dial finishing and WrapConn
cancelled only the dial registration: the connection dialed on the old
network entered the fresh registry and survived the network change.

Replace the pair with a Dial handle. Sweep marks pending dials under the
sweeper mutex, and WrapConn decides under the same mutex: a swept dial's
connection is closed and ErrSwept returned, so the caller redials on the
new network; otherwise the connection transfers to the registry with no
window in between.

client/android/client.go

@@ -24,8 +24,9 @@ import (
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	"github.com/netbirdio/netbird/client/internal/routemanager"
 	"github.com/netbirdio/netbird/client/internal/stdnet"
-	"github.com/netbirdio/netbird/client/mdm"
 	"github.com/netbirdio/netbird/client/net"
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/formatter"
 	"github.com/netbirdio/netbird/route"
@@ -33,11 +34,6 @@ import (
 	types "github.com/netbirdio/netbird/upload-server/types"
 )
 
-// ConnectionListener export internal Listener for mobile
-type ConnectionListener interface {
-	peer.Listener
-}
-
 // TunAdapter export internal TunAdapter for mobile
 type TunAdapter interface {
 	device.TunAdapter
@@ -78,19 +74,18 @@ type Client struct {
 	deviceName            string
 	uiVersion             string
 	networkChangeListener listener.NetworkChangeListener
+	// netState outlives engine restarts: it mirrors the OS connectivity, not
+	// the engine lifecycle. Run and RunWithoutLogin inject it into each new
+	// ConnectClient, which distributes it to every reconnection loop.
+	netState *netstate.State
+
+	// sweeper also outlives engine restarts; NotifyNetworkChange sweeps it.
+	sweeper *netsweep.Sweeper
 
 	stateMu       sync.RWMutex
 	connectClient *internal.ConnectClient
 	config        *profilemanager.Config
 	cacheDir      string
-
-	// mdmLoader holds the per-Client MDM policy source. Set by
-	// SetMDMPolicyFetcher (called from the Kotlin side). Each Run
-	// passes this loader to the resolved Config so applyMDMPolicy
-	// picks up the active overlay. Nil means "MDM enforcement off
-	// for this Client".
-	mdmLoader *mdm.Loader
-
 	// Identifies the running profile for the SSO login hint; see profile_state.go.
 	cfgPath string
 
@@ -157,9 +152,28 @@ func NewClient(androidSDKVersion int, deviceName string, uiVersion string, tunAd
 		recorder:              peer.NewRecorder(""),
 		ctxCancelLock:         &sync.Mutex{},
 		networkChangeListener: networkChangeListener,
+		netState:              netstate.New(),
+		sweeper:               netsweep.New(),
 	}
 }
 
+// SetNetworkAvailable feeds OS-reported network availability into the client.
+// While unavailable, the internal reconnect loops suspend their attempts and
+// the connection listener reports NoNetwork instead of Connecting; when
+// availability returns, the loops resume immediately with a fresh backoff.
+func (c *Client) SetNetworkAvailable(available bool) {
+	c.netState.Set(available)
+	c.recorder.SetNetworkAvailable(available)
+}
+
+// NotifyNetworkChange cuts the management, signal and relay connections
+// after the OS switched networks, so the reconnect loops redial immediately
+// on the new one. The engine and the TUN device stay untouched.
+func (c *Client) NotifyNetworkChange() {
+	n := c.sweeper.Sweep()
+	log.Infof("network change: swept %d connections", n)
+}
+
 // Run start the internal client. It is a blocker function
 func (c *Client) Run(platformFiles PlatformFiles, urlOpener URLOpener, isAndroidTV bool, dns *DNSList, dnsReadyListener DnsReadyListener, envList *EnvList) error {
 	exportEnvList(envList)
@@ -176,7 +190,6 @@ func (c *Client) Run(platformFiles PlatformFiles, urlOpener URLOpener, isAndroid
 	if err != nil {
 		return err
 	}
-	c.applyMDMOverlay(cfg)
 	c.recorder.UpdateManagementAddress(cfg.ManagementURL.String())
 	c.recorder.UpdateRosenpass(cfg.RosenpassEnabled, cfg.RosenpassPermissive)
 
@@ -198,7 +211,8 @@ func (c *Client) Run(platformFiles PlatformFiles, urlOpener URLOpener, isAndroid
 	}
 	// todo do not throw error in case of cancelled context
 	ctx = internal.CtxInitState(ctx)
-	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
+	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder,
+		internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
 	c.setState(cfg, cacheDir, cfgFile, connectClient)
 	// This path runs the interactive SSO flow, so reaching here means the peer
 	// is authenticated again — release the latch Status() reports from. Clear
@@ -226,7 +240,6 @@ func (c *Client) RunWithoutLogin(platformFiles PlatformFiles, dns *DNSList, dnsR
 	if err != nil {
 		return err
 	}
-	c.applyMDMOverlay(cfg)
 	c.recorder.UpdateManagementAddress(cfg.ManagementURL.String())
 	c.recorder.UpdateRosenpass(cfg.RosenpassEnabled, cfg.RosenpassPermissive)
 
@@ -240,7 +253,8 @@ func (c *Client) RunWithoutLogin(platformFiles PlatformFiles, dns *DNSList, dnsR
 
 	// todo do not throw error in case of cancelled context
 	ctx = internal.CtxInitState(ctx)
-	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
+	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder,
+		internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
 	c.setState(cfg, cacheDir, cfgFile, connectClient)
 	return connectClient.RunOnAndroid(c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, slices.Clone(dns.items), dnsReadyListener, stateFile, cacheDir)
 }
@@ -302,7 +316,6 @@ func (c *Client) DebugBundle(platformFiles PlatformFiles, anonymize bool) (strin
 		if err != nil {
 			return "", fmt.Errorf("load config: %w", err)
 		}
-		c.applyMDMOverlay(cfg)
 		cacheDir = platformFiles.CacheDir()
 	}
 
@@ -525,7 +538,7 @@ func (c *Client) OnUpdatedHostDNS(list *DNSList) error {
 
 // SetConnectionListener set the network connection listener
 func (c *Client) SetConnectionListener(listener ConnectionListener) {
-	c.recorder.SetConnectionListener(listener)
+	c.recorder.SetConnectionListener(connectionListenerAdapter{listener})
 }
 
 // RemoveConnectionListener remove connection listener

client/android/connection_listener.go

@@ -0,0 +1,41 @@
+//go:build android
+
+package android
+
+import (
+	"github.com/netbirdio/netbird/client/internal/peer"
+)
+
+// Client state values delivered via ConnectionListener.OnStateChanged,
+// re-exported as basic constants so gomobile emits them into the generated
+// Java bindings. They mirror peer.ClientState*: append-only, never reorder.
+const (
+	ClientStateDisconnected  = int(peer.ClientStateDisconnected)
+	ClientStateConnected     = int(peer.ClientStateConnected)
+	ClientStateConnecting    = int(peer.ClientStateConnecting)
+	ClientStateDisconnecting = int(peer.ClientStateDisconnecting)
+	ClientStateNoNetwork     = int(peer.ClientStateNoNetwork)
+)
+
+// ConnectionListener export internal Listener for mobile. It mirrors
+// peer.Listener with OnStateChanged taking a plain int (one of the
+// ClientState* constants), because gomobile cannot bind named types.
+type ConnectionListener interface {
+	OnStateChanged(state int)
+	OnConnected()
+	OnDisconnected()
+	OnConnecting()
+	OnDisconnecting()
+	OnAddressChanged(string, string)
+	OnPeersListChanged(int)
+}
+
+// connectionListenerAdapter adapts the gomobile-facing ConnectionListener to
+// peer.Listener, converting the typed state to the int the binding carries.
+type connectionListenerAdapter struct {
+	ConnectionListener
+}
+
+func (a connectionListenerAdapter) OnStateChanged(state peer.ClientState) {
+	a.ConnectionListener.OnStateChanged(int(state))
+}

client/android/mdm.go

@@ -1,80 +0,0 @@
-//go:build android
-
-package android
-
-import (
-	"encoding/json"
-
-	log "github.com/sirupsen/logrus"
-
-	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
-)
-
-// PolicyFetcher is the mobile-side bridge for the MDM managed-config
-// snapshot. The native layer (Kotlin) implements this and registers
-// the instance per Client via Client.SetMDMPolicyFetcher. Every
-// invocation of fetchJSON must read the current RestrictionsManager
-// state and return the result as a JSON-encoded map[string]any string.
-//
-// JSON is used because gomobile does not support map[string]any
-// crossing the JNI boundary — the adapter on the Go side parses the
-// string back into the map[string]any expected by mdm.Loader.
-//
-// Return value contract:
-//   - "" (empty)         : interpreted as "no MDM source / no managed keys"
-//   - "{}"               : managed config explicitly empty
-//   - "{...}"            : JSON object with key/value pairs
-//   - malformed JSON     : logged and treated as empty
-type PolicyFetcher interface {
-	FetchJSON() string
-}
-
-// jsonFetcherAdapter wraps a gomobile-exposed PolicyFetcher into the
-// internal mdm.PolicyFetcher interface, taking care of JSON decoding
-// on every Fetch.
-type jsonFetcherAdapter struct {
-	inner PolicyFetcher
-}
-
-func (a *jsonFetcherAdapter) Fetch() map[string]any {
-	raw := a.inner.FetchJSON()
-	if raw == "" {
-		return nil
-	}
-	var out map[string]any
-	if err := json.Unmarshal([]byte(raw), &out); err != nil {
-		log.Warnf("MDM mobile fetcher: invalid JSON payload from native: %v", err)
-		return nil
-	}
-	return out
-}
-
-// SetMDMPolicyFetcher registers the native-provided MDM policy fetcher
-// on this Client. Call once from the gomobile-init code (Kotlin
-// Application.onCreate or Service onCreate) before invoking Run /
-// RunWithoutLogin. Passing nil disables MDM enforcement on this
-// Client.
-//
-// The fetcher is held as a *mdm.Loader instance on the Client (no
-// package-level state) — multiple Clients in the same process get
-// independent Loaders, and tests can inject fakes per Client.
-func (c *Client) SetMDMPolicyFetcher(p PolicyFetcher) {
-	if p == nil {
-		c.mdmLoader = mdm.NewLoader(nil)
-		return
-	}
-	c.mdmLoader = mdm.NewLoader(&jsonFetcherAdapter{inner: p})
-}
-
-// applyMDMOverlay applies the Client-held MDM Loader's current policy
-// on top of the just-read Config. Called immediately after every
-// UpdateOrCreateConfig — profilemanager's apply() initialises the
-// policy to empty and leaves overlay responsibility to the lifecycle
-// owner. No-op when no fetcher was registered.
-func (c *Client) applyMDMOverlay(cfg *profilemanager.Config) {
-	if cfg == nil || c.mdmLoader == nil {
-		return
-	}
-	cfg.ApplyMDMPolicy(c.mdmLoader.Load())
-}

client/cmd/login.go

@@ -17,7 +17,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/auth"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
 	nbnet "github.com/netbirdio/netbird/client/net"
 	"github.com/netbirdio/netbird/client/proto"
 	"github.com/netbirdio/netbird/client/server"
@@ -333,11 +332,6 @@ func doForegroundLogin(ctx context.Context, cmd *cobra.Command, setupKey string,
 	if err != nil {
 		return fmt.Errorf("read config file %s: %v", configFilePath, err)
 	}
-	// CLI standalone login: profilemanager no longer auto-applies MDM,
-	// so layer in the OS-native policy here. Desktop builds construct
-	// a Loader with no fetcher — the build-tagged loadPlatform reads
-	// the registry/plist directly.
-	config.ApplyMDMPolicy(mdm.NewLoader(nil).Load())
 
 	// Mirror runInForegroundMode: recover residual state (DNS, firewall,
 	// ssh config, legacy routing) from a previous unclean shutdown and

client/cmd/up.go

@@ -21,7 +21,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
 	"github.com/netbirdio/netbird/client/proto"
 	nbnet "github.com/netbirdio/netbird/client/net"
 	"github.com/netbirdio/netbird/client/server"
@@ -229,10 +228,6 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command, activeProf *pr
 	if err != nil {
 		return fmt.Errorf("get config file: %v", err)
 	}
-	// CLI foreground path runs without the daemon Server: layer in the
-	// active MDM policy explicitly so a forced ManagementURL / PSK /
-	// other managed key actually takes effect on this run.
-	config.ApplyMDMPolicy(mdm.NewLoader(nil).Load())
 
 	_, _ = profilemanager.UpdateOldManagementURL(ctx, config, configFilePath)
 

client/embed/embed.go

@@ -21,7 +21,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal/auth"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
 	sshcommon "github.com/netbirdio/netbird/client/ssh"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/shared/management/domain"
@@ -216,10 +215,6 @@ func New(opts Options) (*Client, error) {
 	if err != nil {
 		return nil, fmt.Errorf("create config: %w", err)
 	}
-	// Embedded path runs without the daemon Server: apply the active
-	// MDM policy explicitly so a forced ManagementURL / PSK / other
-	// managed key takes effect on this embedded engine instance.
-	config.ApplyMDMPolicy(mdm.NewLoader(nil).Load())
 
 	if opts.PrivateKey != "" {
 		config.PrivateKey = opts.PrivateKey

client/grpc/dialer_generic.go

@@ -16,28 +16,47 @@ import (
 	"google.golang.org/grpc"
 
 	nbnet "github.com/netbirdio/netbird/client/net"
+	"github.com/netbirdio/netbird/client/netsweep"
 )
 
 func WithCustomDialer(_ bool, _ string) grpc.DialOption {
+	return grpc.WithContextDialer(dialContext)
+}
+
+// WithSweeper dials like WithCustomDialer but registers connections and
+// dials with the sweeper. Append it after WithCustomDialer: gRPC applies
+// dial options in order, so the later context dialer wins.
+func WithSweeper(sweeper *netsweep.Sweeper) grpc.DialOption {
 	return grpc.WithContextDialer(func(ctx context.Context, addr string) (net.Conn, error) {
-		if runtime.GOOS == "linux" {
-			currentUser, err := user.Current()
-			if err != nil {
-				return nil, status.Errorf(codes.FailedPrecondition, "failed to get current user: %v", err)
-			}
+		dial := sweeper.StartDial(ctx)
+		defer dial.Release()
 
-			// the custom dialer requires root permissions which are not required for use cases run as non-root
-			if currentUser.Uid != "0" {
-				log.Debug("Not running as root, using standard dialer")
-				dialer := &net.Dialer{}
-				return dialer.DialContext(ctx, "tcp", addr)
-			}
-		}
-
-		conn, err := nbnet.NewDialer().DialContext(ctx, "tcp", addr)
+		conn, err := dialContext(dial.Ctx(), addr)
 		if err != nil {
-			return nil, fmt.Errorf("nbnet.NewDialer().DialContext: %w", err)
+			return nil, err
 		}
-		return conn, nil
+		return dial.WrapConn(conn)
 	})
 }
+
+func dialContext(ctx context.Context, addr string) (net.Conn, error) {
+	if runtime.GOOS == "linux" {
+		currentUser, err := user.Current()
+		if err != nil {
+			return nil, status.Errorf(codes.FailedPrecondition, "failed to get current user: %v", err)
+		}
+
+		// the custom dialer requires root permissions which are not required for use cases run as non-root
+		if currentUser.Uid != "0" {
+			log.Debug("Not running as root, using standard dialer")
+			dialer := &net.Dialer{}
+			return dialer.DialContext(ctx, "tcp", addr)
+		}
+	}
+
+	conn, err := nbnet.NewDialer().DialContext(ctx, "tcp", addr)
+	if err != nil {
+		return nil, fmt.Errorf("nbnet.NewDialer().DialContext: %w", err)
+	}
+	return conn, nil
+}

client/grpc/dialer_js.go

@@ -3,6 +3,7 @@ package grpc
 import (
 	"google.golang.org/grpc"
 
+	"github.com/netbirdio/netbird/client/netsweep"
 	"github.com/netbirdio/netbird/util/wsproxy/client"
 )
 
@@ -11,3 +12,8 @@ import (
 func WithCustomDialer(tlsEnabled bool, component string) grpc.DialOption {
 	return client.WithWebSocketDialer(tlsEnabled, component)
 }
+
+// WithSweeper is a no-op on WASM/JS: there is no network change signal.
+func WithSweeper(_ *netsweep.Sweeper) grpc.DialOption {
+	return grpc.EmptyDialOption{}
+}

client/internal/connect.go

@@ -38,6 +38,8 @@ import (
 	"github.com/netbirdio/netbird/client/internal/updater"
 	"github.com/netbirdio/netbird/client/internal/updater/installer"
 	nbnet "github.com/netbirdio/netbird/client/net"
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	cProto "github.com/netbirdio/netbird/client/proto"
 	"github.com/netbirdio/netbird/client/ssh"
 	sshconfig "github.com/netbirdio/netbird/client/ssh/config"
@@ -70,18 +72,42 @@ type ConnectClient struct {
 	updateManager *updater.Manager
 
 	persistSyncResponse bool
+
+	// netState gates every reconnection loop on OS-reported network
+	// availability. Nil (the default) disables gating; mobile platforms
+	// inject it via WithNetworkState.
+	netState *netstate.State
+
+	// sweeper cuts the management, signal and relay connections on network
+	// change; nil disables it.
+	sweeper *netsweep.Sweeper
+}
+
+// ConnectClientOption configures optional ConnectClient behavior.
+type ConnectClientOption func(*ConnectClient)
+
+// WithNetworkState injects the OS network availability state that gates every
+// reconnection loop; without it gating is disabled.
+func WithNetworkState(netState *netstate.State) ConnectClientOption {
+	return func(c *ConnectClient) { c.netState = netState }
+}
+
+// WithSweeper injects the network change sweeper.
+func WithSweeper(sweeper *netsweep.Sweeper) ConnectClientOption {
+	return func(c *ConnectClient) { c.sweeper = sweeper }
 }
 
 func NewConnectClient(
 	ctx context.Context,
 	config *profilemanager.Config,
 	statusRecorder *peer.Status,
+	opts ...ConnectClientOption,
 ) *ConnectClient {
 	// Derive the run context here so Stop owns the cancel that unblocks the run
 	// loop. runCancel is set once at construction, so Stop can call it without
 	// racing the run loop's startup. Callers therefore need not cancel before Stop.
 	runCtx, runCancel := context.WithCancel(ctx)
-	return &ConnectClient{
+	c := &ConnectClient{
 		ctx:            runCtx,
 		runCancel:      runCancel,
 		runExited:      make(chan struct{}),
@@ -89,6 +115,10 @@ func NewConnectClient(
 		statusRecorder: statusRecorder,
 		engineMutex:    sync.Mutex{},
 	}
+	for _, opt := range opts {
+		opt(c)
+	}
+	return c
 }
 
 func (c *ConnectClient) SetUpdateManager(um *updater.Manager) {
@@ -274,6 +304,13 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 			return nil
 		}
 
+		// suspend connection attempts while the OS reports no usable network
+		if waited, err := c.netState.Wait(c.ctx); err != nil {
+			return nil
+		} else if waited {
+			backOff.Reset()
+		}
+
 		state.Set(StatusConnecting)
 
 		engineCtx, cancel := context.WithCancel(c.ctx)
@@ -285,7 +322,8 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 		}()
 
 		log.Debugf("connecting to the Management service %s", c.config.ManagementURL.Host)
-		mgmClient, err := mgm.NewClient(engineCtx, c.config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled)
+		mgmClient, err := mgm.NewClient(engineCtx, c.config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled,
+			mgm.WithNetworkState(c.netState), mgm.WithSweeper(c.sweeper))
 		if err != nil {
 			// On daemon shutdown / Down() the parent context is cancelled
 			// and the dial fails with "context canceled". Wrapping that
@@ -360,7 +398,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 		}()
 
 		// with the global Netbird config in hand connect (just a connection, no stream yet) Signal
-		signalClient, err := connectToSignal(engineCtx, loginResp.GetNetbirdConfig(), myPrivateKey)
+		signalClient, err := connectToSignal(engineCtx, loginResp.GetNetbirdConfig(), myPrivateKey, c.netState, c.sweeper)
 		if err != nil {
 			log.Error(err)
 			return wrapErr(err)
@@ -396,7 +434,8 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 			engineConfig.StateDir = filepath.Dir(path)
 		}
 
-		relayManager := relayClient.NewManager(engineCtx, relayURLs, myPrivateKey.PublicKey().String(), engineConfig.MTU)
+		relayManager := relayClient.NewManager(engineCtx, relayURLs, myPrivateKey.PublicKey().String(), engineConfig.MTU,
+			relayClient.WithNetworkState(c.netState), relayClient.WithSweeper(c.sweeper))
 		c.statusRecorder.SetRelayMgr(relayManager)
 		if len(relayURLs) > 0 {
 			if token != nil {
@@ -424,6 +463,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 			UpdateManager:  c.updateManager,
 			ClientMetrics:  c.clientMetrics,
 			MetricsCtx:     c.ctx,
+			NetState:       c.netState,
 		}, mobileDependency)
 		engine.SetSyncResponsePersistence(c.persistSyncResponse)
 		c.engine = engine
@@ -480,6 +520,16 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 	// status stream stuck at Connecting.
 	err = backoff.Retry(operation, backoff.WithContext(backOff, c.ctx))
 	if err != nil {
+		// Once the client context is cancelled backoff.WithContext surfaces the
+		// bare context error, and any attempt torn down mid-flight reports the
+		// same. That cancellation is the caller asking us to stop (Stop, Down or
+		// an engine restart), so exit cleanly instead of handing back a failure
+		// the caller would have to distinguish from a real one.
+		if c.ctx.Err() != nil && errors.Is(err, context.Canceled) {
+			log.Info("exiting client retry loop, context cancelled")
+			return nil
+		}
+
 		log.Debugf("exiting client retry loop due to unrecoverable error: %s", err)
 		if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.PermissionDenied) {
 			state.Set(StatusNeedsLogin)
@@ -673,7 +723,7 @@ func selectMTU(localMTU uint16, peerMTU int32) uint16 {
 }
 
 // connectToSignal creates Signal Service client and established a connection
-func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourPrivateKey wgtypes.Key) (*signal.GrpcClient, error) {
+func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourPrivateKey wgtypes.Key, netState *netstate.State, sweeper *netsweep.Sweeper) (*signal.GrpcClient, error) {
 	var sigTLSEnabled bool
 	if wtConfig.Signal.Protocol == mgmProto.HostConfig_HTTPS {
 		sigTLSEnabled = true
@@ -681,7 +731,8 @@ func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourP
 		sigTLSEnabled = false
 	}
 
-	signalClient, err := signal.NewClient(ctx, wtConfig.Signal.Uri, ourPrivateKey, sigTLSEnabled)
+	signalClient, err := signal.NewClient(ctx, wtConfig.Signal.Uri, ourPrivateKey, sigTLSEnabled,
+		signal.WithNetworkState(netState), signal.WithSweeper(sweeper))
 	if err != nil {
 		log.Errorf("error while connecting to the Signal Exchange Service %s: %s", wtConfig.Signal.Uri, err)
 		return nil, gstatus.Errorf(codes.FailedPrecondition, "failed connecting to Signal Service : %s", err)

client/internal/engine.go

@@ -58,6 +58,7 @@ import (
 	"github.com/netbirdio/netbird/client/internal/syncstore"
 	"github.com/netbirdio/netbird/client/internal/updater"
 	"github.com/netbirdio/netbird/client/jobexec"
+	"github.com/netbirdio/netbird/client/netstate"
 	cProto "github.com/netbirdio/netbird/client/proto"
 	"github.com/netbirdio/netbird/client/system"
 	nbdns "github.com/netbirdio/netbird/dns"
@@ -180,6 +181,9 @@ type EngineServices struct {
 	UpdateManager  *updater.Manager
 	ClientMetrics  *metrics.ClientMetrics
 	MetricsCtx     context.Context
+	// NetState gates the reconnection loops on OS-reported network
+	// availability; nil disables gating.
+	NetState *netstate.State
 }
 
 // Engine is a mechanism responsible for reacting on Signal and Management stream events and managing connections to the remote peers.
@@ -203,6 +207,10 @@ type Engine struct {
 	config    *EngineConfig
 	mobileDep MobileDependency
 
+	// netState gates the peer reconnection guards on OS-reported network
+	// availability; nil disables gating.
+	netState *netstate.State
+
 	// STUNs is a list of STUN servers used by ICE
 	STUNs []*stun.URI
 	// TURNs is a list of STUN servers used by ICE
@@ -336,6 +344,7 @@ func NewEngine(
 		syncMsgMux:         &sync.Mutex{},
 		config:             config,
 		mobileDep:          mobileDep,
+		netState:           services.NetState,
 		STUNs:              []*stun.URI{},
 		TURNs:              []*stun.URI{},
 		networkSerial:      0,
@@ -1891,7 +1900,8 @@ func (e *Engine) createPeerConn(pubKey string, allowedIPs []netip.Prefix, agentV
 			Addr:           e.getRosenpassAddr(),
 			PermissiveMode: e.config.RosenpassPermissive,
 		},
-		ICEConfig: e.createICEConfig(),
+		ICEConfig:    e.createICEConfig(),
+		NetworkState: e.netState,
 	}
 
 	serviceDependencies := peer.ServiceDependencies{

client/internal/peer/conn.go

@@ -26,6 +26,7 @@ import (
 	"github.com/netbirdio/netbird/client/internal/portforward"
 	"github.com/netbirdio/netbird/client/internal/rosenpass"
 	"github.com/netbirdio/netbird/client/internal/stdnet"
+	"github.com/netbirdio/netbird/client/netstate"
 	"github.com/netbirdio/netbird/route"
 	relayClient "github.com/netbirdio/netbird/shared/relay/client"
 )
@@ -93,6 +94,10 @@ type ConnConfig struct {
 
 	// ICEConfig ICE protocol configuration
 	ICEConfig icemaker.Config
+
+	// NetworkState gates the reconnection guard on OS-reported network
+	// availability; nil disables gating.
+	NetworkState *netstate.State
 }
 
 type Conn struct {
@@ -254,7 +259,7 @@ func (conn *Conn) open(engineCtx context.Context, firstPacket []byte) error {
 		conn.handshaker.AddICEListener(conn.workerICE.OnNewOffer)
 	}
 
-	conn.guard = guard.NewGuard(conn.Log, conn.isConnectedOnAllWay, conn.config.Timeout, conn.srWatcher)
+	conn.guard = guard.NewGuard(conn.Log, conn.isConnectedOnAllWay, conn.config.Timeout, conn.srWatcher, conn.config.NetworkState)
 
 	conn.wg.Add(1)
 	go func() {

client/internal/peer/guard/guard.go

@@ -6,6 +6,8 @@ import (
 
 	"github.com/cenkalti/backoff/v4"
 	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/netstate"
 )
 
 // ConnStatus represents the connection state as seen by the guard.
@@ -31,20 +33,26 @@ type connStatusFunc func() ConnStatus
 // - Relayed connection disconnected
 // - ICE candidate changes
 type Guard struct {
-	log                     *log.Entry
-	isConnectedOnAllWay     connStatusFunc
-	timeout                 time.Duration
-	srWatcher               *SRWatcher
+	log                 *log.Entry
+	isConnectedOnAllWay connStatusFunc
+	timeout             time.Duration
+	srWatcher           *SRWatcher
+	// netState gates reconnect attempts on OS-reported network availability;
+	// nil disables gating.
+	netState                *netstate.State
 	relayedConnDisconnected chan struct{}
 	iCEConnDisconnected     chan struct{}
 }
 
-func NewGuard(log *log.Entry, isConnectedFn connStatusFunc, timeout time.Duration, srWatcher *SRWatcher) *Guard {
+// NewGuard creates a reconnection guard for a peer connection. A nil netState
+// disables network availability gating.
+func NewGuard(log *log.Entry, isConnectedFn connStatusFunc, timeout time.Duration, srWatcher *SRWatcher, netState *netstate.State) *Guard {
 	return &Guard{
 		log:                     log,
 		isConnectedOnAllWay:     isConnectedFn,
 		timeout:                 timeout,
 		srWatcher:               srWatcher,
+		netState:                netState,
 		relayedConnDisconnected: make(chan struct{}, 1),
 		iCEConnDisconnected:     make(chan struct{}, 1),
 	}
@@ -99,6 +107,12 @@ func (g *Guard) reconnectLoopWithRetry(ctx context.Context, callback func()) {
 	for {
 		select {
 		case <-tickerChannel:
+			// skip attempts while the OS reports no usable network; the ticker
+			// keeps running so other events remain responsive, and the guard
+			// resumes via the signal/relay reconnect events once network returns
+			if !g.netState.IsOnline() {
+				continue
+			}
 			switch g.isConnectedOnAllWay() {
 			case ConnStatusConnected:
 				// all good, nothing to do

client/internal/peer/guard/guard_leak_test.go

@@ -15,7 +15,7 @@ import (
 
 func newTestGuard(status connStatusFunc) *Guard {
 	srw := NewSRWatcher(nil, nil, nil, ice.Config{})
-	return NewGuard(log.WithField("test", "guard"), status, 50*time.Millisecond, srw)
+	return NewGuard(log.WithField("test", "guard"), status, 50*time.Millisecond, srw, nil)
 }
 
 // countBackoffTickerGoroutines returns how many goroutines are currently sitting

client/internal/peer/listener.go

@@ -1,11 +1,40 @@
 package peer
 
+// ClientState identifies the client connection state delivered via
+// Listener.OnStateChanged.
+type ClientState int
+
+// Client states. The numeric values cross the gomobile boundary (the mobile
+// bindings re-export them as integer constants), so they are a wire format:
+// append new states at the end, never reorder or insert.
+const (
+	ClientStateDisconnected ClientState = iota
+	ClientStateConnected
+	ClientStateConnecting
+	ClientStateDisconnecting
+	// ClientStateNoNetwork is an overlay state: it is never stored as the
+	// last notification, only derived from ClientStateConnecting while the
+	// OS reports no usable network (see notifier.effectiveState).
+	ClientStateNoNetwork
+)
+
 // Listener is a callback type about the NetBird network connection state
 type Listener interface {
+	// OnStateChanged reports every client state transition. New states are
+	// delivered only through this callback; the per-state callbacks below
+	// are kept for compatibility and will be removed once all consumers
+	// have migrated.
+	OnStateChanged(state ClientState)
+
+	// Deprecated: consume OnStateChanged instead.
 	OnConnected()
+	// Deprecated: consume OnStateChanged instead.
 	OnDisconnected()
+	// Deprecated: consume OnStateChanged instead.
 	OnConnecting()
+	// Deprecated: consume OnStateChanged instead.
 	OnDisconnecting()
+
 	OnAddressChanged(string, string)
 	OnPeersListChanged(int)
 }

client/internal/peer/notifier.go

@@ -4,31 +4,57 @@ import (
 	"sync"
 )
 
-const (
-	stateDisconnected = iota
-	stateConnected
-	stateConnecting
-	stateDisconnecting
-)
-
 type notifier struct {
 	serverStateLock    sync.Mutex
 	listenersLock      sync.Mutex
 	listener           Listener
 	currentClientState bool
-	lastNotification   int
+	lastNotification   ClientState
 	lastNumberOfPeers  int
 	lastFqdnAddress    string
 	lastIPAddress      string
+	networkAvailable   bool
 }
 
 func newNotifier() *notifier {
-	return &notifier{}
+	return &notifier{
+		networkAvailable: true,
+	}
+}
+
+// effectiveState maps the computed state to what listeners should see:
+// while the OS reports no usable network, "Connecting" would be a lie —
+// connection attempts are suspended — so it is reported as NoNetwork.
+// Caller must hold serverStateLock.
+func (n *notifier) effectiveState(state ClientState) ClientState {
+	if !n.networkAvailable && state == ClientStateConnecting {
+		return ClientStateNoNetwork
+	}
+	return state
+}
+
+// setNetworkAvailable records the OS network availability and re-notifies
+// the listener when the flag flips the effective state (Connecting <->
+// NoNetwork).
+func (n *notifier) setNetworkAvailable(available bool) {
+	n.serverStateLock.Lock()
+	if n.networkAvailable == available {
+		n.serverStateLock.Unlock()
+		return
+	}
+	previous := n.effectiveState(n.lastNotification)
+	n.networkAvailable = available
+	current := n.effectiveState(n.lastNotification)
+	n.serverStateLock.Unlock()
+
+	if previous != current {
+		n.notify(current)
+	}
 }
 
 func (n *notifier) setListener(listener Listener) {
 	n.serverStateLock.Lock()
-	lastNotification := n.lastNotification
+	lastNotification := n.effectiveState(n.lastNotification)
 	numOfPeers := n.lastNumberOfPeers
 	fqdnAddress := n.lastFqdnAddress
 	address := n.lastIPAddress
@@ -61,43 +87,45 @@ func (n *notifier) updateServerStates(mgmState bool, signalState bool) {
 	}
 
 	n.lastNotification = calculatedState
+	effective := n.effectiveState(calculatedState)
 	n.serverStateLock.Unlock()
 
-	n.notify(calculatedState)
+	n.notify(effective)
 }
 
 func (n *notifier) clientStart() {
 	n.serverStateLock.Lock()
 	n.currentClientState = true
-	n.lastNotification = stateConnecting
+	n.lastNotification = ClientStateConnecting
+	effective := n.effectiveState(ClientStateConnecting)
 	n.serverStateLock.Unlock()
 
-	n.notify(stateConnecting)
+	n.notify(effective)
 }
 
 func (n *notifier) clientStop() {
 	n.serverStateLock.Lock()
 	n.currentClientState = false
-	n.lastNotification = stateDisconnected
+	n.lastNotification = ClientStateDisconnected
 	n.serverStateLock.Unlock()
 
-	n.notify(stateDisconnected)
+	n.notify(ClientStateDisconnected)
 }
 
 func (n *notifier) clientTearDown() {
 	n.serverStateLock.Lock()
 	n.currentClientState = false
-	n.lastNotification = stateDisconnecting
+	n.lastNotification = ClientStateDisconnecting
 	n.serverStateLock.Unlock()
 
-	n.notify(stateDisconnecting)
+	n.notify(ClientStateDisconnecting)
 }
 
-func (n *notifier) isServerStateChanged(newState int) bool {
+func (n *notifier) isServerStateChanged(newState ClientState) bool {
 	return n.lastNotification != newState
 }
 
-func (n *notifier) notify(state int) {
+func (n *notifier) notify(state ClientState) {
 	n.listenersLock.Lock()
 	listener := n.listener
 	n.listenersLock.Unlock()
@@ -109,20 +137,20 @@ func (n *notifier) notify(state int) {
 	notifyListener(listener, state)
 }
 
-func (n *notifier) calculateState(managementConn, signalConn bool) int {
+func (n *notifier) calculateState(managementConn, signalConn bool) ClientState {
 	if managementConn && signalConn {
-		return stateConnected
+		return ClientStateConnected
 	}
 
 	if !managementConn && !signalConn && !n.currentClientState {
-		return stateDisconnected
+		return ClientStateDisconnected
 	}
 
-	if n.lastNotification == stateDisconnecting {
-		return stateDisconnecting
+	if n.lastNotification == ClientStateDisconnecting {
+		return ClientStateDisconnecting
 	}
 
-	return stateConnecting
+	return ClientStateConnecting
 }
 
 func (n *notifier) peerListChanged(numOfPeers int) {
@@ -159,15 +187,19 @@ func (n *notifier) localAddressChanged(fqdn, address string) {
 	listener.OnAddressChanged(fqdn, address)
 }
 
-func notifyListener(l Listener, state int) {
+func notifyListener(l Listener, state ClientState) {
+	// legacy per-state callbacks; NoNetwork is delivered only via
+	// OnStateChanged below
 	switch state {
-	case stateDisconnected:
+	case ClientStateDisconnected:
 		l.OnDisconnected()
-	case stateConnected:
+	case ClientStateConnected:
 		l.OnConnected()
-	case stateConnecting:
+	case ClientStateConnecting:
 		l.OnConnecting()
-	case stateDisconnecting:
+	case ClientStateDisconnecting:
 		l.OnDisconnecting()
 	}
+
+	l.OnStateChanged(state)
 }

client/internal/peer/notifier_test.go

@@ -6,29 +6,32 @@ import (
 )
 
 type mocListener struct {
-	lastState int
+	lastState ClientState
 	wg        sync.WaitGroup
 	peersWg   sync.WaitGroup
 	peers     int
 }
 
 func (l *mocListener) OnConnected() {
-	l.lastState = stateConnected
+	l.lastState = ClientStateConnected
 	l.wg.Done()
 }
 func (l *mocListener) OnDisconnected() {
-	l.lastState = stateDisconnected
+	l.lastState = ClientStateDisconnected
 	l.wg.Done()
 }
 func (l *mocListener) OnConnecting() {
-	l.lastState = stateConnecting
+	l.lastState = ClientStateConnecting
 	l.wg.Done()
 }
 func (l *mocListener) OnDisconnecting() {
-	l.lastState = stateDisconnecting
+	l.lastState = ClientStateDisconnecting
 	l.wg.Done()
 }
 
+func (l *mocListener) OnStateChanged(state ClientState) {
+
+}
 func (l *mocListener) OnAddressChanged(host, addr string) {
 
 }
@@ -57,15 +60,15 @@ func Test_notifier_serverState(t *testing.T) {
 
 	type scenario struct {
 		name        string
-		expected    int
+		expected    ClientState
 		mgmState    bool
 		signalState bool
 	}
 	scenarios := []scenario{
-		{"connected", stateConnected, true, true},
-		{"mgm down", stateConnecting, false, true},
-		{"signal down", stateConnecting, true, false},
-		{"disconnected", stateDisconnected, false, false},
+		{"connected", ClientStateConnected, true, true},
+		{"mgm down", ClientStateConnecting, false, true},
+		{"signal down", ClientStateConnecting, true, false},
+		{"disconnected", ClientStateDisconnected, false, false},
 	}
 
 	for _, tt := range scenarios {
@@ -85,7 +88,7 @@ func Test_notifier_SetListener(t *testing.T) {
 	listener.setPeersWaiter()
 
 	n := newNotifier()
-	n.lastNotification = stateConnecting
+	n.lastNotification = ClientStateConnecting
 	n.setListener(listener)
 	listener.wait()
 	listener.waitPeers()
@@ -99,7 +102,7 @@ func Test_notifier_RemoveListener(t *testing.T) {
 	listener.setWaiter()
 	listener.setPeersWaiter()
 	n := newNotifier()
-	n.lastNotification = stateConnecting
+	n.lastNotification = ClientStateConnecting
 	n.setListener(listener)
 	// setListener replays cached state on a goroutine; wait for both the state
 	// and peers callbacks to finish so we don't race on listener.peers.

client/internal/peer/status.go

@@ -1211,6 +1211,12 @@ func (d *Status) ClientTeardown() {
 	d.notifyStateChange()
 }
 
+// SetNetworkAvailable records the OS-reported network availability; while
+// unavailable, listeners see NoNetwork instead of Connecting.
+func (d *Status) SetNetworkAvailable(available bool) {
+	d.notifier.setNetworkAvailable(available)
+}
+
 // SetConnectionListener set a listener to the notifier
 func (d *Status) SetConnectionListener(listener Listener) {
 	d.notifier.setListener(listener)

client/internal/profilemanager/config.go

@@ -58,6 +58,10 @@ var DefaultInterfaceBlacklist = []string{
 	"Tailscale", "tailscale", "docker", "veth", "br-", "lo",
 }
 
+// loadMDMPolicy is the package-level indirection used by apply() to read the
+// active MDM policy. Tests override this to inject a fake policy.
+var loadMDMPolicy = mdm.LoadPolicy
+
 // ConfigInput carries configuration changes to the client
 type ConfigInput struct {
 	ManagementURL                 string
@@ -182,27 +186,14 @@ type Config struct {
 
 	MTU uint16
 
-	// policy is the MDM policy that produced the currently-set values
-	// for any MDM-enforced fields. Set by ApplyMDMPolicy on every
-	// invocation. Never persisted to disk. Callers query enforcement
-	// state via Policy() and the mdm.Policy API (HasKey, ManagedKeys,
-	// IsEmpty).
+	// policy is the MDM policy that produced the currently-set values for
+	// any MDM-enforced fields. Set by applyMDMPolicy at the tail of apply()
+	// and reset on every apply() invocation. Never persisted to disk.
+	// Callers query enforcement state via Policy() and the mdm.Policy API
+	// (HasKey, ManagedKeys, IsEmpty).
 	policy *mdm.Policy `json:"-"`
 }
 
-// ApplyMDMPolicy overlays the supplied MDM Policy on top of the
-// currently resolved Config values. Idempotent — pass an empty Policy
-// to clear any prior overlay. The lifecycle owner (Server.getConfig
-// on desktop, the Client.Run path on mobile) calls this with
-// loader.Load() once the per-process Loader is known; the Config
-// itself holds no reference to the Loader.
-func (config *Config) ApplyMDMPolicy(policy *mdm.Policy) {
-	if config == nil {
-		return
-	}
-	config.applyMDMPolicy(policy)
-}
-
 // Policy returns the MDM policy applied to this Config. Returns a non-nil
 // empty Policy when MDM enforcement is inactive; callers can always invoke
 // HasKey / ManagedKeys / IsEmpty without a nil check.
@@ -659,11 +650,9 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
 		updated = true
 	}
 
-	// Initialise the MDM overlay to "no enforcement" so Config.Policy()
-	// never returns a stale or nil policy on a freshly applied Config.
-	// Lifecycle owners that want to enforce a real MDM policy invoke
-	// Config.ApplyMDMPolicy(loader.Load()) after this returns.
-	config.applyMDMPolicy(mdm.NewPolicy(nil))
+	// MDM is the last override layer: any key present in the policy
+	// supersedes defaults, on-disk config, env vars and CLI input.
+	config.applyMDMPolicy(loadMDMPolicy())
 
 	return updated, nil
 }

client/internal/profilemanager/config_mdm_test.go

@@ -10,58 +10,24 @@ import (
 	"github.com/netbirdio/netbird/client/mdm"
 )
 
-// fakeFetcher implements mdm.PolicyFetcher returning a pre-set policy
-// map. Test helper used to construct a Loader without touching the OS
-// or any package-level state.
-type fakeFetcher struct{ values map[string]any }
-
-func (f *fakeFetcher) Fetch() map[string]any { return f.values }
-
-// loaderFor builds an mdm.Loader whose loadPlatform returns the
-// supplied Policy's underlying values.
-func loaderFor(policy *mdm.Policy) *mdm.Loader {
-	if policy == nil || policy.IsEmpty() {
-		return mdm.NewLoader(&fakeFetcher{values: nil})
-	}
-	values := make(map[string]any)
-	for _, k := range policy.ManagedKeys() {
-		if v, ok := policy.GetString(k); ok {
-			values[k] = v
-			continue
-		}
-		if v, ok := policy.GetBool(k); ok {
-			values[k] = v
-			continue
-		}
-		if v, ok := policy.GetInt(k); ok {
-			values[k] = v
-			continue
-		}
-		if v, ok := policy.GetStringSlice(k); ok {
-			values[k] = v
-		}
-	}
-	return mdm.NewLoader(&fakeFetcher{values: values})
-}
-
-// configWithMDM is the test convenience that builds a Config via
-// UpdateOrCreateConfig and overlays the supplied MDM policy on top —
-// mirrors the production pattern (Server.getConfig / Client.applyMDMOverlay)
-// where the Loader lives outside Config and the apply step is driven
-// by the lifecycle owner.
-func configWithMDM(t *testing.T, input ConfigInput, policy *mdm.Policy) *Config {
+// withMDMPolicy temporarily overrides the package-level loadMDMPolicy hook so
+// apply() observes the supplied Policy. The original loader is restored at
+// test cleanup.
+func withMDMPolicy(t *testing.T, policy *mdm.Policy) {
 	t.Helper()
-	cfg, err := UpdateOrCreateConfig(input)
-	require.NoError(t, err)
-	require.NotNil(t, cfg)
-	cfg.ApplyMDMPolicy(loaderFor(policy).Load())
-	return cfg
+	prev := loadMDMPolicy
+	loadMDMPolicy = func() *mdm.Policy { return policy }
+	t.Cleanup(func() { loadMDMPolicy = prev })
 }
 
 func TestApply_MDMEmpty_NoEnforcement(t *testing.T) {
-	cfg := configWithMDM(t, ConfigInput{
+	withMDMPolicy(t, mdm.NewPolicy(nil))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
 		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
-	}, mdm.NewPolicy(nil))
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
 
 	assert.True(t, cfg.Policy().IsEmpty(), "no MDM source ⇒ empty Policy")
 	assert.False(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
@@ -73,15 +39,18 @@ func TestApply_MDMEmpty_NoEnforcement(t *testing.T) {
 
 func TestApply_MDMOnly_OverridesDefaults(t *testing.T) {
 	const mdmURL = "https://corp.mdm.example.com:443"
-
-	cfg := configWithMDM(t, ConfigInput{
-		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
-	}, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL:       mdmURL,
 		mdm.KeyDisableClientRoutes: true,
 		mdm.KeyBlockInbound:        true,
 	}))
 
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
 	assert.Equal(t, mdmURL, cfg.ManagementURL.String())
 	assert.True(t, cfg.DisableClientRoutes)
 	assert.True(t, cfg.BlockInbound)
@@ -96,25 +65,33 @@ func TestApply_MDMBeatsCLIInput(t *testing.T) {
 	const mdmURL = "https://mdm.example.com:443"
 	const cliURL = "https://cli.example.com:443"
 
-	cfg := configWithMDM(t, ConfigInput{
-		ConfigPath:    filepath.Join(t.TempDir(), "config.json"),
-		ManagementURL: cliURL,
-	}, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL: mdmURL,
 	}))
 
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath:    filepath.Join(t.TempDir(), "config.json"),
+		ManagementURL: cliURL,
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
 	// MDM wins over CLI-supplied management URL.
 	assert.Equal(t, mdmURL, cfg.ManagementURL.String())
 	assert.True(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
 }
 
 func TestApply_MDMInvalidURL_KeepsPreviousValue(t *testing.T) {
-	cfg := configWithMDM(t, ConfigInput{
-		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
-	}, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL: "not-a-url",
 	}))
 
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
 	// Invalid MDM URL is logged and skipped: default URL stays in place
 	// to keep the client functional.
 	assert.Equal(t, DefaultManagementURL, cfg.ManagementURL.String())
@@ -129,20 +106,24 @@ func TestApply_MDMBoolKeysOverrideOnDiskValue(t *testing.T) {
 	tmp := filepath.Join(t.TempDir(), "config.json")
 
 	// Seed without MDM.
-	configWithMDM(t, ConfigInput{
+	withMDMPolicy(t, mdm.NewPolicy(nil))
+	_, err := UpdateOrCreateConfig(ConfigInput{
 		ConfigPath:          tmp,
 		DisableClientRoutes: boolPtr(false),
 		RosenpassEnabled:    boolPtr(false),
-	}, mdm.NewPolicy(nil))
+	})
+	require.NoError(t, err)
 
 	// Now enable MDM enforcement for these keys.
-	cfg := configWithMDM(t, ConfigInput{
-		ConfigPath: tmp,
-	}, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyDisableClientRoutes: true,
 		mdm.KeyRosenpassEnabled:    true,
 	}))
 
+	cfg, err := UpdateOrCreateConfig(ConfigInput{ConfigPath: tmp})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
 	assert.True(t, cfg.DisableClientRoutes, "MDM override should flip on-disk false to true")
 	assert.True(t, cfg.RosenpassEnabled)
 	assert.True(t, cfg.Policy().HasKey(mdm.KeyDisableClientRoutes))
@@ -164,12 +145,16 @@ func TestApply_MDMLazyConnection(t *testing.T) {
 	}
 	for _, c := range cases {
 		t.Run(c.name, func(t *testing.T) {
-			cfg := configWithMDM(t, ConfigInput{
-				ConfigPath: filepath.Join(t.TempDir(), "config.json"),
-			}, mdm.NewPolicy(map[string]any{
+			withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 				mdm.KeyLazyConnection: c.raw,
 			}))
 
+			cfg, err := UpdateOrCreateConfig(ConfigInput{
+				ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+			})
+			require.NoError(t, err)
+			require.NotNil(t, cfg)
+
 			assert.Equal(t, c.want, cfg.LazyConnection)
 			assert.True(t, cfg.Policy().HasKey(mdm.KeyLazyConnection))
 		})
@@ -179,12 +164,16 @@ func TestApply_MDMLazyConnection(t *testing.T) {
 func TestApply_MDMPreSharedKeyRedactionSentinelRejected(t *testing.T) {
 	const maskSentinel = "**********"
 
-	cfg := configWithMDM(t, ConfigInput{
-		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
-	}, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyPreSharedKey: maskSentinel,
 	}))
 
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
 	// Mask sentinel must not be persisted as the actual PSK.
 	assert.NotEqual(t, maskSentinel, cfg.PreSharedKey)
 	// Key still marked managed so user writes are still rejected.

client/ios/NetBirdSDK/client.go

@@ -21,7 +21,8 @@ import (
 	"github.com/netbirdio/netbird/client/internal/listener"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/formatter"
 	"github.com/netbirdio/netbird/route"
@@ -29,11 +30,6 @@ import (
 	types "github.com/netbirdio/netbird/upload-server/types"
 )
 
-// ConnectionListener export internal Listener for mobile
-type ConnectionListener interface {
-	peer.Listener
-}
-
 // RouteListener export internal RouteListener for mobile
 type NetworkChangeListener interface {
 	listener.NetworkChangeListener
@@ -80,16 +76,15 @@ type Client struct {
 	onHostDnsFn           func([]string)
 	dnsManager            dns.IosDnsManager
 	loginComplete         bool
+	// netState outlives engine restarts: it mirrors the OS connectivity, not
+	// the engine lifecycle. Run injects it into each new ConnectClient, which
+	// distributes it to every reconnection loop.
+	netState *netstate.State
+	// sweeper also outlives engine restarts; NotifyNetworkChange sweeps it.
+	sweeper *netsweep.Sweeper
 	// preloadedConfig holds config loaded from JSON (used on tvOS where file writes are blocked)
 	preloadedConfig *profilemanager.Config
 
-	// mdmLoader holds the per-Client MDM policy source. Set by
-	// SetMDMPolicyFetcher (called from the Swift side at extension
-	// init). Each Run passes this loader to the resolved Config so
-	// applyMDMPolicy picks up the active overlay. Nil means "MDM
-	// enforcement off for this Client".
-	mdmLoader *mdm.Loader
-
 	stateMu       sync.RWMutex
 	connectClient *internal.ConnectClient
 	config        *profilemanager.Config
@@ -109,6 +104,8 @@ func NewClient(cfgFile, stateFile, cacheDir, logFilePath, deviceName string, osV
 		ctxCancelLock:         &sync.Mutex{},
 		networkChangeListener: networkChangeListener,
 		dnsManager:            dnsManager,
+		netState:              netstate.New(),
+		sweeper:               netsweep.New(),
 	}
 }
 
@@ -150,7 +147,6 @@ func (c *Client) Run(fd int32, interfaceName string, envList *EnvList) error {
 		if err != nil {
 			return err
 		}
-		c.applyMDMOverlay(cfg)
 	}
 	c.recorder.UpdateManagementAddress(cfg.ManagementURL.String())
 	c.recorder.UpdateRosenpass(cfg.RosenpassEnabled, cfg.RosenpassPermissive)
@@ -185,7 +181,8 @@ func (c *Client) Run(fd int32, interfaceName string, envList *EnvList) error {
 	c.onHostDnsFn = func([]string) {}
 	cfg.WgIface = interfaceName
 
-	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
+	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder,
+		internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
 	c.setState(cfg, connectClient)
 	// Persist the latest sync response so DebugBundle can include the network
 	// map. On iOS this is backed by disk to keep it out of the constrained
@@ -194,6 +191,24 @@ func (c *Client) Run(fd int32, interfaceName string, envList *EnvList) error {
 	return connectClient.RunOniOS(fd, c.networkChangeListener, c.dnsManager, c.stateFile, c.cacheDir, c.logFilePath)
 }
 
+// SetNetworkAvailable feeds OS-reported network availability into the client
+// (e.g. from NWPathMonitor). While unavailable, the internal reconnect loops
+// suspend their attempts and the connection listener reports NoNetwork
+// instead of Connecting; when availability returns, the loops resume
+// immediately with a fresh backoff.
+func (c *Client) SetNetworkAvailable(available bool) {
+	c.netState.Set(available)
+	c.recorder.SetNetworkAvailable(available)
+}
+
+// NotifyNetworkChange cuts the management, signal and relay connections
+// after the OS switched networks, so the reconnect loops redial immediately
+// on the new one. The engine and the TUN device stay untouched.
+func (c *Client) NotifyNetworkChange() {
+	n := c.sweeper.Sweep()
+	log.Infof("network change: swept %d connections", n)
+}
+
 // Stop the internal client and free the resources
 func (c *Client) Stop() {
 	c.ctxCancelLock.Lock()
@@ -229,7 +244,6 @@ func (c *Client) DebugBundle(anonymize bool) (string, error) {
 				return "", fmt.Errorf("load config: %w", err)
 			}
 		}
-		c.applyMDMOverlay(cfg)
 	}
 
 	deps := debug.GeneratorDependencies{
@@ -330,7 +344,7 @@ func (c *Client) GetStatusDetails() *StatusDetails {
 
 // SetConnectionListener set the network connection listener
 func (c *Client) SetConnectionListener(listener ConnectionListener) {
-	c.recorder.SetConnectionListener(listener)
+	c.recorder.SetConnectionListener(connectionListenerAdapter{listener})
 }
 
 // RemoveConnectionListener remove connection listener
@@ -379,7 +393,6 @@ func (c *Client) IsLoginRequired() bool {
 			// If we can't load config, assume login is required
 			return true
 		}
-		c.applyMDMOverlay(cfg)
 	}
 
 	if cfg == nil {
@@ -428,7 +441,6 @@ func (c *Client) LoginForMobile() string {
 		log.Errorf("LoginForMobile: failed to load config: %v", err)
 		return fmt.Sprintf("failed to load config: %v", err)
 	}
-	c.applyMDMOverlay(cfg)
 
 	oAuthFlow, err := auth.NewOAuthFlow(ctx, cfg, false, false, "")
 	if err != nil {

client/ios/NetBirdSDK/connection_listener.go

@@ -0,0 +1,43 @@
+//go:build ios
+
+package NetBirdSDK
+
+import (
+	"github.com/netbirdio/netbird/client/internal/peer"
+)
+
+// Client state values, re-exported as basic constants so gomobile emits them
+// into the generated bindings. They mirror peer.ClientState*: append-only,
+// never reorder.
+const (
+	ClientStateDisconnected  = int(peer.ClientStateDisconnected)
+	ClientStateConnected     = int(peer.ClientStateConnected)
+	ClientStateConnecting    = int(peer.ClientStateConnecting)
+	ClientStateDisconnecting = int(peer.ClientStateDisconnecting)
+	ClientStateNoNetwork     = int(peer.ClientStateNoNetwork)
+)
+
+// ConnectionListener export internal Listener for mobile.
+//
+// It intentionally lacks OnStateChanged for now: adding a method to a gomobile
+// interface breaks every Swift implementation, so the iOS app keeps building
+// against the legacy per-state callbacks. A follow-up will extend it together
+// with the app.
+type ConnectionListener interface {
+	OnConnected()
+	OnDisconnected()
+	OnConnecting()
+	OnDisconnecting()
+	OnAddressChanged(string, string)
+	OnPeersListChanged(int)
+}
+
+// connectionListenerAdapter adapts the gomobile-facing ConnectionListener to
+// peer.Listener.
+type connectionListenerAdapter struct {
+	ConnectionListener
+}
+
+// OnStateChanged is dropped on iOS until the app adopts the state callback;
+// the legacy per-state callbacks continue to fire.
+func (a connectionListenerAdapter) OnStateChanged(peer.ClientState) {}

client/ios/NetBirdSDK/mdm.go

@@ -1,82 +0,0 @@
-//go:build ios
-
-package NetBirdSDK
-
-import (
-	"encoding/json"
-
-	log "github.com/sirupsen/logrus"
-
-	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	"github.com/netbirdio/netbird/client/mdm"
-)
-
-// PolicyFetcher is the mobile-side bridge for the MDM managed-config
-// snapshot. The native layer (Swift) implements this and registers
-// the instance per Client via Client.SetMDMPolicyFetcher. Every
-// invocation of fetchJSON must read the current
-// UserDefaults.standard.dictionary(forKey: "com.apple.configuration.managed")
-// and return the result as a JSON-encoded map[string]any string.
-//
-// JSON is used because gomobile does not support map[string]any
-// crossing the Objective-C boundary — the adapter on the Go side
-// parses the string back into the map[string]any expected by
-// mdm.Loader.
-//
-// Return value contract:
-//   - "" (empty)         : interpreted as "no MDM source / no managed keys"
-//   - "{}"               : managed config explicitly empty
-//   - "{...}"            : JSON object with key/value pairs
-//   - malformed JSON     : logged and treated as empty
-type PolicyFetcher interface {
-	FetchJSON() string
-}
-
-// jsonFetcherAdapter wraps a gomobile-exposed PolicyFetcher into the
-// internal mdm.PolicyFetcher interface, taking care of JSON decoding
-// on every Fetch.
-type jsonFetcherAdapter struct {
-	inner PolicyFetcher
-}
-
-func (a *jsonFetcherAdapter) Fetch() map[string]any {
-	raw := a.inner.FetchJSON()
-	if raw == "" {
-		return nil
-	}
-	var out map[string]any
-	if err := json.Unmarshal([]byte(raw), &out); err != nil {
-		log.Warnf("MDM mobile fetcher: invalid JSON payload from native: %v", err)
-		return nil
-	}
-	return out
-}
-
-// SetMDMPolicyFetcher registers the native-provided MDM policy fetcher
-// on this Client. Call once from the gomobile-init code (Swift
-// AppDelegate / PacketTunnelProvider.startTunnel) before invoking Run.
-// Passing nil disables MDM enforcement on this Client.
-//
-// The fetcher is held as a *mdm.Loader instance on the Client (no
-// package-level state) — multiple Clients in the same process get
-// independent Loaders, and tests can inject fakes per Client.
-func (c *Client) SetMDMPolicyFetcher(p PolicyFetcher) {
-	if p == nil {
-		c.mdmLoader = mdm.NewLoader(nil)
-		return
-	}
-	c.mdmLoader = mdm.NewLoader(&jsonFetcherAdapter{inner: p})
-}
-
-// applyMDMOverlay applies the Client-held MDM Loader's current policy
-// on top of the just-read Config. Called immediately after every
-// UpdateOrCreateConfig / DirectUpdateOrCreateConfig — profilemanager's
-// apply() initialises the policy to empty and leaves overlay
-// responsibility to the lifecycle owner. No-op when no fetcher was
-// registered.
-func (c *Client) applyMDMOverlay(cfg *profilemanager.Config) {
-	if cfg == nil || c.mdmLoader == nil {
-		return
-	}
-	cfg.ApplyMDMPolicy(c.mdmLoader.Load())
-}

client/mdm/policy.go

@@ -104,48 +104,16 @@ func NewPolicy(values map[string]any) *Policy {
 	return &Policy{values: values}
 }
 
-// PolicyFetcher is implemented by mobile platforms (Android / iOS) that
-// push the OS-managed configuration into the Go runtime instead of
-// having Go read an on-disk source directly. Desktop platforms ignore
-// this interface — Loader.loadPlatform on windows/darwin reads the
-// registry / plist on its own. A Loader constructed with a non-nil
-// fetcher delegates to it on mobile; passing nil disables MDM
-// enforcement (loadPlatform returns nil values).
-type PolicyFetcher interface {
-	Fetch() map[string]any
-}
-
-// Loader is the DI-friendly entry point for reading the active MDM
-// policy. Construct one at the daemon's lifecycle owner (Server on
-// desktop, gomobile-exposed bridge on mobile) and pass it to anything
-// that needs to read MDM state (the reload ticker, profilemanager's
-// Config). Each callsite has the Loader handed in instead of looking
-// up package-level state.
-type Loader struct {
-	fetcher PolicyFetcher
-}
-
-// NewLoader constructs a Loader. The fetcher is consulted only on
-// mobile builds (ios || android); on desktop it is unused but accepted
-// to keep a single constructor signature across platforms — pass nil
-// on desktop.
-func NewLoader(f PolicyFetcher) *Loader {
-	return &Loader{fetcher: f}
-}
-
-// Load reads the platform-native MDM configuration and returns a
-// Policy. Returns an empty (but non-nil) Policy when no source is
-// present, the source is empty, or the platform is unsupported.
+// LoadPolicy reads the platform-native MDM configuration. Returns an
+// empty (but non-nil) Policy when no source is present, the source is
+// empty, or the platform is unsupported.
 //
 // Diagnostic logging differentiates the three states:
 //   - source absent / unsupported platform: trace log only
 //   - source present, zero keys:             info "MDM enrolled (no managed keys)"
 //   - source present, N keys:                info "MDM enrolled with N managed keys: [...]"
-func (l *Loader) Load() *Policy {
-	if l == nil {
-		return &Policy{values: map[string]any{}}
-	}
-	values, err := l.loadPlatform()
+func LoadPolicy() *Policy {
+	values, err := loadPlatformPolicy()
 	if err != nil {
 		log.Tracef("MDM policy load: %v", err)
 		return &Policy{values: map[string]any{}}

client/mdm/policy_darwin.go

@@ -25,10 +25,8 @@ import (
 // writable plist, as a defense against tampered installs.
 const policyPlistPath = "/Library/Managed Preferences/io.netbird.client.plist"
 
-// loadPlatform reads the MDM-managed configuration from the macOS
-// managed-preferences plist at policyPlistPath. The Loader's fetcher
-// field is unused on this platform — the plist is the authoritative
-// source. Returns:
+// loadPlatformPolicy reads the MDM-managed configuration from the macOS
+// managed-preferences plist at policyPlistPath. Returns:
 //   - (nil, nil)  when the plist is absent (device not MDM-enrolled for
 //     NetBird, or admin has not yet pushed a payload)
 //   - (map, nil)  with N entries when N managed values are present
@@ -41,13 +39,7 @@ const policyPlistPath = "/Library/Managed Preferences/io.netbird.client.plist"
 // skipped so a stray entry in the payload does not block startup.
 // Native plist value types map naturally onto the Policy accessor
 // expectations (GetString / GetBool / GetInt / GetStringSlice).
-func (l *Loader) loadPlatform() (map[string]any, error) {
-	// Honour the injected fetcher when present so tests (and any
-	// future non-macOS MDM channel) can short-circuit the plist read
-	// with a scripted policy.
-	if l != nil && l.fetcher != nil {
-		return l.fetcher.Fetch(), nil
-	}
+func loadPlatformPolicy() (map[string]any, error) {
 	f, err := os.Open(policyPlistPath)
 	if err != nil {
 		if errors.Is(err, fs.ErrNotExist) {

client/mdm/policy_mobile.go

@@ -2,14 +2,13 @@
 
 package mdm
 
-// loadPlatform reads the OS-managed configuration via the native
-// PolicyFetcher injected at Loader construction. Returns
-// (nil, nil) — the platform-absent sentinel that Loader.Load treats as
-// "no MDM source present" — when no fetcher was provided.
-func (l *Loader) loadPlatform() (map[string]any, error) {
-	if l == nil || l.fetcher == nil {
-		//nolint:nilnil // (nil, nil) is the documented platform-absent sentinel; see Loader.Load.
-		return nil, nil
-	}
-	return l.fetcher.Fetch(), nil
+// loadPlatformPolicy is unused on mobile: the native layer (Swift on iOS,
+// Kotlin/Java on Android) reads the OS managed-config store and pushes the
+// resulting dictionary in-process via a gomobile entry point that lands in
+// Phase 5 / Phase 6. The stub keeps the package compilable for mobile
+// builds and returns (nil, nil) — the platform-absent sentinel that
+// LoadPolicy in policy.go treats as "no MDM source present".
+func loadPlatformPolicy() (map[string]any, error) {
+	//nolint:nilnil // (nil, nil) is the documented platform-absent sentinel; see LoadPolicy.
+	return nil, nil
 }

client/mdm/policy_other.go

@@ -2,17 +2,13 @@
 
 package mdm
 
-// loadPlatform reads the MDM policy on platforms without a native MDM
-// channel (Linux, FreeBSD). When no fetcher was injected the policy is
-// (nil, nil) — the platform-absent sentinel that Loader.Load treats as
-// "MDM enforcement disabled". A non-nil fetcher takes precedence: it
-// is the test-seam used by unit tests to inject a scripted policy
-// without touching the OS, and the same hook supports any future
-// non-mobile OS that grows an out-of-band MDM channel.
-func (l *Loader) loadPlatform() (map[string]any, error) {
-	if l != nil && l.fetcher != nil {
-		return l.fetcher.Fetch(), nil
-	}
-	//nolint:nilnil // (nil, nil) is the documented platform-absent sentinel; see Loader.Load.
+// loadPlatformPolicy returns no policy on platforms without an MDM channel
+// (Linux, FreeBSD). MDM enforcement is off and the client behaves as if
+// the feature did not exist. Returns (nil, nil) — the platform-absent
+// sentinel the caller (LoadPolicy in policy.go) treats as "no MDM
+// source present"; an error here would just translate to the same
+// outcome with an extra log line.
+func loadPlatformPolicy() (map[string]any, error) {
+	//nolint:nilnil // (nil, nil) is the documented platform-absent sentinel; see LoadPolicy.
 	return nil, nil
 }

client/mdm/policy_test.go

@@ -155,12 +155,10 @@ func TestPolicy_GetStringSlice(t *testing.T) {
 	})
 }
 
-func TestLoader_NilFetcherReturnsEmpty(t *testing.T) {
-	// Loader.Load with no fetcher (desktop construction) must degrade
-	// gracefully and never return nil; on linux loadPlatform is a stub
-	// returning (nil, nil), and Load is expected to translate that
-	// into a non-nil empty Policy.
-	p := NewLoader(nil).Load()
+func TestLoadPolicy_PlatformStubReturnsEmpty(t *testing.T) {
+	// loadPlatformPolicy is a stub on every OS for Phase 1. LoadPolicy must
+	// degrade gracefully and never return nil.
+	p := LoadPolicy()
 	require.NotNil(t, p)
 	assert.True(t, p.IsEmpty())
 	assert.Empty(t, p.ManagedKeys())

client/mdm/policy_windows.go

@@ -61,10 +61,8 @@ func readRegistryValue(k registry.Key, name, canonical string, out map[string]an
 	}
 }
 
-// loadPlatform reads the MDM-managed configuration from the Windows
-// registry under HKLM\Software\Policies\NetBird. The Loader's fetcher
-// field is unused on this platform — the registry is the
-// authoritative source. Returns:
+// loadPlatformPolicy reads the MDM-managed configuration from the
+// Windows registry under HKLM\Software\Policies\NetBird. Returns:
 //   - (nil, nil)  when the key is absent (device not MDM-enrolled for NetBird)
 //   - (map, nil)  with N entries when N managed values are set (N may be 0)
 //   - (nil, err)  on open / enumerate registry errors
@@ -72,13 +70,7 @@ func readRegistryValue(k registry.Key, name, canonical string, out map[string]an
 // Per-value type coercion + skip-on-error is delegated to
 // readRegistryValue. Unknown value names are logged and skipped so a
 // malformed deployment does not block startup.
-func (l *Loader) loadPlatform() (map[string]any, error) {
-	// Honour the injected fetcher when present so tests (and any
-	// future non-Windows MDM channel) can short-circuit the registry
-	// read with a scripted policy.
-	if l != nil && l.fetcher != nil {
-		return l.fetcher.Fetch(), nil
-	}
+func loadPlatformPolicy() (map[string]any, error) {
 	k, err := registry.OpenKey(registry.LOCAL_MACHINE, policyRegistryPath, registry.QUERY_VALUE)
 	if err != nil {
 		if errors.Is(err, registry.ErrNotExist) {

client/mdm/ticker.go

@@ -15,33 +15,33 @@ import (
 // instead, hence anticipating the ticker mechanism entirely.
 const DefaultReloadInterval = 1 * time.Minute
 
-// Ticker periodically re-reads the OS-native MDM policy via the
-// injected Loader and invokes the onChange callback (supplied to Run)
-// whenever the observed Policy diverges from the last observation
-// (added / removed / changed keys). Launch with Run from a goroutine;
-// cancel the supplied context to stop.
+// policyLoader is the indirection through which the ticker reads the
+// OS-native policy, both for the initial observation and on every tick.
+// Production points it at LoadPolicy; tests in this package override it to
+// feed a scripted sequence of policies without touching the real OS store.
+var policyLoader = LoadPolicy
+
+// Ticker periodically re-reads the OS-native MDM policy via LoadPolicy and
+// invokes the onChange callback (supplied to Run) whenever the observed
+// Policy diverges from the last observation (added / removed / changed
+// keys). Launch with Run from a goroutine; cancel the supplied context
+// to stop.
 type Ticker struct {
 	interval time.Duration
-	loader   *Loader
 	prev     *Policy
 }
 
 // NewTicker constructs a Ticker that will re-read the OS-native policy
-// every reloadInterval once Run is called. The Loader is injected so
-// the ticker doesn't depend on any package-level state — production
-// passes the daemon-owned Loader, tests pass a fake Loader (built with
-// a fake PolicyFetcher).
-//
-// The initial snapshot is populated by calling loader.Load() at
+// every reloadInterval once Run is called.
+// The initial snapshot is populated by calling policyLoader at
 // construction time so the first tick only fires
 // onChange when the policy actually changed since boot — without
 // this baseline the first tick would report every currently-managed
 // key as "added" and trigger a spurious engine restart.
-func NewTicker(reloadInterval time.Duration, loader *Loader) *Ticker {
+func NewTicker(reloadInterval time.Duration) *Ticker {
 	return &Ticker{
 		interval: reloadInterval,
-		loader:   loader,
-		prev:     loader.Load(),
+		prev:     policyLoader(),
 	}
 }
 
@@ -58,7 +58,7 @@ func (t *Ticker) Run(ctx context.Context, onChange func(prev, curr *Policy) erro
 			log.Info("MDM policy reload ticker stopped")
 			return
 		case <-tk.C:
-			curr := t.loader.Load()
+			curr := policyLoader()
 			if policiesEqual(t.prev, curr) {
 				continue
 			}

client/mdm/ticker_test.go

@@ -13,40 +13,28 @@ import (
 // testReloadInterval for speeding up the ticker cadence under `go test`
 const testReloadInterval = 1 * time.Second
 
-// fakePolicyFetcher implements PolicyFetcher returning a scripted
-// policy map. Goroutine-safe so the test can mutate the script while
-// the ticker is observing it.
-type fakePolicyFetcher struct {
-	mu     sync.Mutex
-	values map[string]any
-}
-
-func (f *fakePolicyFetcher) Fetch() map[string]any {
-	f.mu.Lock()
-	defer f.mu.Unlock()
-	if f.values == nil {
-		return nil
-	}
-	out := make(map[string]any, len(f.values))
-	for k, v := range f.values {
-		out[k] = v
-	}
-	return out
-}
-
-func (f *fakePolicyFetcher) set(values map[string]any) {
-	f.mu.Lock()
-	defer f.mu.Unlock()
-	f.values = values
+// withPolicyLoader overrides the package-level policyLoader for the duration
+// of the test so the ticker observes a scripted policy instead of the real
+// OS-native store. The original loader is restored on cleanup.
+func withPolicyLoader(t *testing.T, fn func() *Policy) {
+	t.Helper()
+	prev := policyLoader
+	policyLoader = fn
+	t.Cleanup(func() { policyLoader = prev })
 }
 
 func TestTicker_FiresOnChangeWithDelta(t *testing.T) {
-	fetcher := &fakePolicyFetcher{} // initial observation: empty (no enforcement)
-	loader := NewLoader(fetcher)
+	var mu sync.Mutex
+	current := NewPolicy(nil) // initial observation: empty (no enforcement)
+	withPolicyLoader(t, func() *Policy {
+		mu.Lock()
+		defer mu.Unlock()
+		return current
+	})
 
 	type change struct{ prev, curr *Policy }
 	changes := make(chan change, 1)
-	tk := NewTicker(testReloadInterval, loader)
+	tk := NewTicker(testReloadInterval)
 	require.Equal(t, testReloadInterval, tk.interval)
 
 	ctx, cancel := context.WithCancel(context.Background())
@@ -61,13 +49,15 @@ func TestTicker_FiresOnChangeWithDelta(t *testing.T) {
 		})
 		close(done)
 	}()
-	// Stop Run and wait for it to exit before returning, so the test
-	// goroutine doesn't race the still-running ticker.
+	// Stop Run and wait for it to exit before returning, so the policyLoader
+	// restore in t.Cleanup can't race the ticker goroutine still reading it.
 	defer func() { cancel(); <-done }()
 
-	// Flip the OS-observed policy from empty to one managed key. The
-	// next tick must detect the diff and invoke onChange.
-	fetcher.set(map[string]any{KeyManagementURL: "https://mdm.example.com:443"})
+	// Flip the OS-observed policy from empty to one managed key. The next
+	// tick must detect the diff and invoke onChange.
+	mu.Lock()
+	current = NewPolicy(map[string]any{KeyManagementURL: "https://mdm.example.com:443"})
+	mu.Unlock()
 
 	select {
 	case c := <-changes:
@@ -79,11 +69,12 @@ func TestTicker_FiresOnChangeWithDelta(t *testing.T) {
 }
 
 func TestTicker_NoCallbackWhenPolicyUnchanged(t *testing.T) {
-	fetcher := &fakePolicyFetcher{values: map[string]any{KeyBlockInbound: true}}
-	loader := NewLoader(fetcher)
+	withPolicyLoader(t, func() *Policy {
+		return NewPolicy(map[string]any{KeyBlockInbound: true})
+	})
 
 	fired := make(chan struct{}, 1)
-	tk := NewTicker(testReloadInterval, loader)
+	tk := NewTicker(testReloadInterval)
 
 	ctx, cancel := context.WithCancel(context.Background())
 	done := make(chan struct{})
@@ -99,8 +90,8 @@ func TestTicker_NoCallbackWhenPolicyUnchanged(t *testing.T) {
 	}()
 	defer func() { cancel(); <-done }()
 
-	// Over ~2 ticks at the 1s test cadence the policy never changes,
-	// so the diff guard must suppress the callback entirely.
+	// Over ~2 ticks at the 1s test cadence the policy never changes, so the
+	// diff guard must suppress the callback entirely.
 	select {
 	case <-fired:
 		t.Fatal("onChange fired despite an unchanged policy")

client/netstate/netstate.go

@@ -0,0 +1,93 @@
+// Package netstate tracks OS-reported network availability for the client.
+//
+// A State instance is owned by the platform integration (e.g. the Android or
+// iOS bindings, fed from ConnectivityManager callbacks or NWPathMonitor) and
+// is injected into the connection retry loops (management, signal, relay,
+// peer guards and the top-level connect loop), which consult it to avoid
+// burning CPU and battery on reconnect attempts while the device has no
+// network at all (e.g. airplane mode), and to reset their backoff as soon as
+// the network returns.
+//
+// Consumers hold a *State that may be nil — every non-mobile platform leaves
+// it unset. The read methods are safe on a nil receiver: they report online
+// and never block, so consumers behave as if this package did not exist.
+package netstate
+
+import (
+	"context"
+	"sync"
+
+	log "github.com/sirupsen/logrus"
+)
+
+// State holds the OS-reported network availability. The zero value is not
+// usable; create instances with New.
+type State struct {
+	mu      sync.Mutex
+	online  bool
+	changed chan struct{}
+}
+
+// New creates a State that starts online.
+func New() *State {
+	return &State{
+		online:  true,
+		changed: make(chan struct{}),
+	}
+}
+
+// Set records whether the OS reports any usable network. Transitions wake up
+// all Wait callers immediately.
+func (s *State) Set(online bool) {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	if s.online == online {
+		return
+	}
+	s.online = online
+	close(s.changed)
+	s.changed = make(chan struct{})
+	log.Infof("OS network availability changed: online=%t", online)
+}
+
+// IsOnline reports whether the OS reports at least one usable network. On a
+// nil receiver — no State injected — it reports online.
+func (s *State) IsOnline() bool {
+	if s == nil {
+		return true
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	return s.online
+}
+
+// Wait blocks while the network is offline. It reports whether it had to
+// wait, so callers can reset their backoff after an outage. It returns early
+// with the context error when ctx is done. On a nil receiver — no State
+// injected — it returns immediately.
+func (s *State) Wait(ctx context.Context) (bool, error) {
+	if s == nil {
+		return false, nil
+	}
+	waited := false
+	for {
+		s.mu.Lock()
+		if s.online {
+			s.mu.Unlock()
+			return waited, nil
+		}
+		ch := s.changed
+		s.mu.Unlock()
+
+		if !waited {
+			waited = true
+			log.Debugf("network is offline, pausing connection attempts")
+		}
+
+		select {
+		case <-ctx.Done():
+			return waited, ctx.Err()
+		case <-ch:
+		}
+	}
+}

client/netstate/netstate_test.go

@@ -0,0 +1,170 @@
+package netstate
+
+import (
+	"context"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestNewStateIsOnline(t *testing.T) {
+	assert.True(t, New().IsOnline(), "a fresh State should start online")
+}
+
+func TestSetTogglesOnlineState(t *testing.T) {
+	s := New()
+
+	s.Set(false)
+	assert.False(t, s.IsOnline(), "state should be offline after Set(false)")
+
+	s.Set(true)
+	assert.True(t, s.IsOnline(), "state should be online after Set(true)")
+}
+
+func TestWaitReturnsImmediatelyWhenOnline(t *testing.T) {
+	s := New()
+
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second)
+	defer cancel()
+
+	waited, err := s.Wait(ctx)
+	require.NoError(t, err)
+	assert.False(t, waited, "Wait should not block when the network is online")
+}
+
+func TestWaitBlocksUntilOnline(t *testing.T) {
+	s := New()
+	s.Set(false)
+
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	result := make(chan bool, 1)
+	go func() {
+		waited, err := s.Wait(ctx)
+		if err != nil {
+			result <- false
+			return
+		}
+		result <- waited
+	}()
+
+	// Verify Wait is actually blocking while offline
+	select {
+	case <-result:
+		t.Fatal("Wait should block while the network is offline")
+	case <-time.After(100 * time.Millisecond):
+	}
+
+	s.Set(true)
+
+	select {
+	case waited := <-result:
+		assert.True(t, waited, "Wait should report that it had to wait for the network")
+	case <-time.After(2 * time.Second):
+		t.Fatal("Wait should return promptly after the network becomes available")
+	}
+}
+
+func TestWaitReturnsOnContextCancel(t *testing.T) {
+	s := New()
+	s.Set(false)
+
+	ctx, cancel := context.WithCancel(context.Background())
+
+	result := make(chan error, 1)
+	go func() {
+		_, err := s.Wait(ctx)
+		result <- err
+	}()
+
+	cancel()
+
+	select {
+	case err := <-result:
+		assert.ErrorIs(t, err, context.Canceled)
+	case <-time.After(2 * time.Second):
+		t.Fatal("Wait should return promptly after context cancellation")
+	}
+}
+
+func TestWaitWakesAllWaiters(t *testing.T) {
+	s := New()
+	s.Set(false)
+
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	const waiters = 10
+	var wg sync.WaitGroup
+	results := make(chan bool, waiters)
+	for i := 0; i < waiters; i++ {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			waited, err := s.Wait(ctx)
+			if err != nil {
+				results <- false
+				return
+			}
+			results <- waited
+		}()
+	}
+
+	time.Sleep(100 * time.Millisecond)
+	s.Set(true)
+	wg.Wait()
+
+	close(results)
+	count := 0
+	for waited := range results {
+		assert.True(t, waited, "every waiter should report that it waited")
+		count++
+	}
+	assert.Equal(t, waiters, count, "all waiters should have returned")
+}
+
+func TestNilStateReadsAreNoops(t *testing.T) {
+	var s *State
+
+	assert.True(t, s.IsOnline(), "nil State should report online")
+
+	waited, err := s.Wait(context.Background())
+	require.NoError(t, err)
+	assert.False(t, waited, "nil State's Wait should not block")
+}
+
+func TestConcurrentSetAndWait(t *testing.T) {
+	s := New()
+
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	var wg sync.WaitGroup
+	for i := 0; i < 4; i++ {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			for j := 0; j < 100; j++ {
+				s.Set(j%2 == 0)
+				s.IsOnline()
+			}
+		}()
+	}
+	for i := 0; i < 4; i++ {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			for j := 0; j < 100; j++ {
+				if _, err := s.Wait(ctx); err != nil {
+					return
+				}
+			}
+		}()
+	}
+
+	wg.Wait()
+}

client/netsweep/netsweep.go

@@ -0,0 +1,171 @@
+// Package netsweep cuts network-bound activity when the OS switches networks:
+// a sweep closes the registered connections and aborts the in-flight dials, so
+// their owners redial immediately instead of waiting for the old sockets to
+// time out.
+//
+// A nil *Sweeper disables everything: all methods are nil-safe no-ops.
+package netsweep
+
+import (
+	"context"
+	"errors"
+	"net"
+	"sync"
+
+	log "github.com/sirupsen/logrus"
+)
+
+// ErrSwept reports that a dial finished after a network change swept its
+// registration. The connection is already closed; the caller must treat it
+// as a failed dial and redial on the new network.
+var ErrSwept = errors.New("netsweep: connection swept by network change")
+
+// sweptConn deregisters itself from the sweeper when closed.
+type sweptConn struct {
+	net.Conn
+	sweeper *Sweeper
+	id      uint64
+}
+
+func (c *sweptConn) Close() error {
+	c.sweeper.deregister(c.id)
+	return c.Conn.Close()
+}
+
+// Sweeper registers live connections and in-flight dials so Sweep can cut
+// everything that started before the network changed.
+type Sweeper struct {
+	mu     sync.Mutex
+	conns  map[uint64]net.Conn
+	dials  map[uint64]*Dial
+	nextID uint64
+}
+
+// New creates an empty sweeper.
+func New() *Sweeper {
+	return &Sweeper{
+		conns: make(map[uint64]net.Conn),
+		dials: make(map[uint64]*Dial),
+	}
+}
+
+// Dial tracks one dial from start to connection registration. It hands the
+// dialed connection to the sweeper atomically, so a sweep can never fall
+// between the dial finishing and the connection being registered.
+type Dial struct {
+	sweeper *Sweeper
+	ctx     context.Context
+	cancel  context.CancelFunc
+	id      uint64
+	done    bool // set by Sweep, WrapConn or Release; guarded by sweeper.mu
+}
+
+// StartDial registers an in-flight dial. Dial with Ctx, hand the result to
+// WrapConn, and Release the dial when the attempt is over, typically deferred.
+func (s *Sweeper) StartDial(ctx context.Context) *Dial {
+	if s == nil {
+		return &Dial{ctx: ctx}
+	}
+
+	ctx, cancel := context.WithCancel(ctx)
+	d := &Dial{sweeper: s, ctx: ctx, cancel: cancel}
+
+	s.mu.Lock()
+	d.id = s.nextID
+	s.nextID++
+	s.dials[d.id] = d
+	s.mu.Unlock()
+
+	return d
+}
+
+// Ctx returns the dial's context. Sweep cancels it, so a dial started on the
+// old network aborts instead of waiting out its handshake timeout.
+func (d *Dial) Ctx() context.Context {
+	return d.ctx
+}
+
+// WrapConn hands conn over to the sweeper. If a sweep ran since StartDial,
+// the connection belongs to the old network: it is closed and ErrSwept is
+// returned. Otherwise conn is registered against the next sweep and returned
+// wrapped, deregistering itself on Close. Call it once, before Release.
+func (d *Dial) WrapConn(conn net.Conn) (net.Conn, error) {
+	s := d.sweeper
+	if s == nil {
+		return conn, nil
+	}
+
+	s.mu.Lock()
+	if d.done {
+		s.mu.Unlock()
+		if err := conn.Close(); err != nil {
+			log.Debugf("swept dial close error: %v", err)
+		}
+		return nil, ErrSwept
+	}
+	d.done = true
+	delete(s.dials, d.id)
+	id := s.nextID
+	s.nextID++
+	s.conns[id] = conn
+	s.mu.Unlock()
+
+	return &sweptConn{Conn: conn, sweeper: s, id: id}, nil
+}
+
+// Release ends the dial's registration and cancels its context. It is
+// idempotent and safe after WrapConn, so callers can defer it.
+func (d *Dial) Release() {
+	s := d.sweeper
+	if s == nil {
+		return
+	}
+
+	s.mu.Lock()
+	d.done = true
+	delete(s.dials, d.id)
+	s.mu.Unlock()
+
+	d.cancel()
+}
+
+// Sweep closes every registered connection, aborts every in-flight dial, and
+// returns how many connections it closed. A dial whose connection was not
+// yet handed to WrapConn is marked, so the late WrapConn closes it instead
+// of registering it.
+func (s *Sweeper) Sweep() int {
+	if s == nil {
+		return 0
+	}
+
+	s.mu.Lock()
+	conns := s.conns
+	dials := s.dials
+	s.conns = make(map[uint64]net.Conn)
+	s.dials = make(map[uint64]*Dial)
+	for _, d := range dials {
+		d.done = true
+	}
+	s.mu.Unlock()
+
+	if len(dials) > 0 {
+		log.Debugf("aborting %d in-flight dials", len(dials))
+		for _, d := range dials {
+			d.cancel()
+		}
+	}
+
+	for _, conn := range conns {
+		log.Debugf("sweeping connection %s -> %s", conn.LocalAddr(), conn.RemoteAddr())
+		if err := conn.Close(); err != nil {
+			log.Debugf("swept connection close error: %v", err)
+		}
+	}
+	return len(conns)
+}
+
+func (s *Sweeper) deregister(id uint64) {
+	s.mu.Lock()
+	delete(s.conns, id)
+	s.mu.Unlock()
+}

client/netsweep/netsweep_test.go

@@ -0,0 +1,168 @@
+package netsweep
+
+import (
+	"context"
+	"net"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestSweepClosesRegisteredConns(t *testing.T) {
+	sweeper := New()
+
+	c1 := wrap(t, sweeper, connPair(t))
+	c2 := wrap(t, sweeper, connPair(t))
+
+	assert.Equal(t, 2, sweeper.Sweep(), "both live connections should be closed")
+
+	// The wrappers must report closed now.
+	buf := make([]byte, 1)
+	_, err := c1.Read(buf)
+	assert.Error(t, err, "first connection should be unusable after the sweep")
+	_, err = c2.Read(buf)
+	assert.Error(t, err, "second connection should be unusable after the sweep")
+
+	assert.Equal(t, 0, sweeper.Sweep(), "second sweep should find nothing")
+}
+
+func TestCloseDeregisters(t *testing.T) {
+	sweeper := New()
+
+	conn := wrap(t, sweeper, connPair(t))
+	require.NoError(t, conn.Close())
+
+	assert.Equal(t, 0, sweeper.Sweep(), "closed connection must leave the registry")
+}
+
+func TestCloseIsIdempotent(t *testing.T) {
+	sweeper := New()
+
+	conn := wrap(t, sweeper, connPair(t))
+	require.NoError(t, conn.Close())
+	assert.Error(t, conn.Close(), "double close surfaces the underlying error but must not panic")
+}
+
+func TestSweepOnlyAffectsOlderConns(t *testing.T) {
+	sweeper := New()
+
+	_ = wrap(t, sweeper, connPair(t))
+	assert.Equal(t, 1, sweeper.Sweep())
+
+	// A connection dialed after the sweep must survive until the next one.
+	_ = wrap(t, sweeper, connPair(t))
+	assert.Equal(t, 1, sweeper.Sweep(), "post-sweep connection belongs to the next sweep")
+}
+
+func TestSweepAbortsInFlightDials(t *testing.T) {
+	sweeper := New()
+
+	dial := sweeper.StartDial(context.Background())
+	defer dial.Release()
+
+	sweeper.Sweep()
+
+	assert.ErrorIs(t, dial.Ctx().Err(), context.Canceled, "sweep must cancel the in-flight dial context")
+}
+
+func TestReleasedDialIsNotAborted(t *testing.T) {
+	sweeper := New()
+
+	// Simulate a dial that finished before the sweep.
+	released := sweeper.StartDial(context.Background())
+	released.Release()
+
+	// A dial still in flight during the sweep.
+	pending := sweeper.StartDial(context.Background())
+	defer pending.Release()
+
+	sweeper.Sweep()
+	assert.ErrorIs(t, pending.Ctx().Err(), context.Canceled, "pending dial must be aborted")
+}
+
+func TestSweepBetweenDialAndHandoffClosesConn(t *testing.T) {
+	sweeper := New()
+
+	dial := sweeper.StartDial(context.Background())
+	defer dial.Release()
+
+	// The dial succeeds on the old network, then the sweep lands before the
+	// connection is handed over.
+	conn := connPair(t)
+	assert.Equal(t, 0, sweeper.Sweep(), "the connection is not registered yet")
+
+	wrapped, err := dial.WrapConn(conn)
+	require.ErrorIs(t, err, ErrSwept)
+	require.Nil(t, wrapped)
+
+	buf := make([]byte, 1)
+	_, err = conn.Read(buf)
+	assert.Error(t, err, "the old-network connection must be closed, not leaked")
+
+	assert.Equal(t, 0, sweeper.Sweep(), "nothing may leak into the next sweep")
+}
+
+func TestNilSweeperIsNoop(t *testing.T) {
+	var sweeper *Sweeper
+
+	conn := connPair(t)
+	dial := sweeper.StartDial(context.Background())
+	defer dial.Release()
+
+	wrapped, err := dial.WrapConn(conn)
+	require.NoError(t, err)
+	assert.Equal(t, conn, wrapped, "nil sweeper must return the conn unchanged")
+	assert.NoError(t, dial.Ctx().Err(), "nil sweeper must not cancel the dial context")
+	assert.Equal(t, 0, sweeper.Sweep(), "nil sweeper closes nothing")
+}
+
+// wrap registers conn with the sweeper through a completed dial.
+func wrap(t *testing.T, sweeper *Sweeper, conn net.Conn) net.Conn {
+	t.Helper()
+
+	dial := sweeper.StartDial(context.Background())
+	defer dial.Release()
+
+	wrapped, err := dial.WrapConn(conn)
+	require.NoError(t, err)
+	return wrapped
+}
+
+// connPair dials a loopback TCP connection and keeps the accepted peer open
+// until the test ends: a peer that closed early would make the connection
+// unreadable on its own, so a read error after the sweep would prove nothing.
+func connPair(t *testing.T) net.Conn {
+	t.Helper()
+
+	l, err := net.Listen("tcp", "127.0.0.1:0")
+	require.NoError(t, err)
+	t.Cleanup(func() {
+		if err := l.Close(); err != nil {
+			t.Logf("listener close error: %v", err)
+		}
+	})
+
+	accepted := make(chan net.Conn, 1)
+	go func() {
+		conn, err := l.Accept()
+		if err != nil {
+			close(accepted)
+			return
+		}
+		accepted <- conn
+	}()
+
+	conn, err := net.Dial("tcp", l.Addr().String())
+	require.NoError(t, err)
+
+	peer, ok := <-accepted
+	require.True(t, ok, "listener must accept the dialed connection")
+	t.Cleanup(func() {
+		if err := peer.Close(); err != nil {
+			t.Logf("peer close error: %v", err)
+		}
+	})
+
+	return conn
+}

client/server/mdm.go

@@ -21,6 +21,10 @@ import (
 // a no-op echo, never as a conflict with the policy.
 const preSharedKeyRedactedSentinel = "**********"
 
+// loadMDMPolicy is the indirection used by server handlers to read the
+// active MDM policy. Tests override this to inject a fake policy.
+var loadMDMPolicy = mdm.LoadPolicy
+
 // conflictCheck is a value-aware comparison between a single field in
 // the incoming request and the corresponding MDM-enforced value. It
 // runs only when the field was actually set in the request (presence

client/server/server.go

@@ -132,15 +132,6 @@ type Server struct {
 	// stopped by the rootCtx cancellation.
 	mdmTicker *mdm.Ticker
 
-	// mdmLoader is the daemon-owned source of the active MDM policy.
-	// Constructed once during Server.Start (with a nil PolicyFetcher on
-	// desktop — the build-tagged Loader.loadPlatform reads the OS
-	// registry / plist directly) and injected into every consumer:
-	// mdmTicker for its periodic reload, the SetConfig / Login MDM
-	// gates for conflict detection, and every Config produced via
-	// getConfig() so its apply() picks up the same overlay.
-	mdmLoader *mdm.Loader
-
 	updateManager *updater.Manager
 
 	jwtCache *jwtCache
@@ -222,14 +213,8 @@ func (s *Server) Start() error {
 	// Runs re-resolves Config (re-running profilemanager.Config.apply which
 	// applies the freshly-read MDM policy as the last layer) and brings
 	// the engine back with the new values.
-	if s.mdmLoader == nil {
-		// Desktop builds pass a nil PolicyFetcher: the Loader's
-		// build-tagged loadPlatform reads the OS source directly
-		// (registry on Windows, plist on macOS, no-op elsewhere).
-		s.mdmLoader = mdm.NewLoader(nil)
-	}
 	if s.mdmTicker == nil {
-		s.mdmTicker = mdm.NewTicker(mdm.DefaultReloadInterval, s.mdmLoader)
+		s.mdmTicker = mdm.NewTicker(mdm.DefaultReloadInterval)
 		go s.mdmTicker.Run(s.rootCtx, s.onMDMPolicyChange)
 	}
 
@@ -474,7 +459,7 @@ func (s *Server) SetConfig(callerCtx context.Context, msg *proto.SetConfigReques
 	// by the active MDM policy. The error carries an MDMManagedFields-
 	// Violation detail listing the offending key names. Non-conflicting
 	// fields in the same request are not applied either.
-	policy := s.mdmLoader.Load()
+	policy := loadMDMPolicy()
 	if err := rejectMDMManagedFieldConflicts(mdmManagedFieldConflicts(msg, policy)); err != nil {
 		return nil, err
 	}
@@ -607,7 +592,7 @@ func (s *Server) Login(callerCtx context.Context, msg *proto.LoginRequest) (*pro
 		if s.checkUpdateSettingsDisabled() {
 			return nil, gstatus.Errorf(codes.Unavailable, errUpdateSettingsDisabled)
 		}
-		policy := s.mdmLoader.Load()
+		policy := loadMDMPolicy()
 		if err := rejectMDMManagedFieldConflicts(loginRequestMDMConflicts(msg, policy)); err != nil {
 			return nil, err
 		}
@@ -1400,12 +1385,6 @@ func (s *Server) getConfig(activeProf *profilemanager.ActiveProfileState) (*prof
 		return nil, false, fmt.Errorf("failed to get config: %w", err)
 	}
 
-	// Apply the daemon-owned MDM policy on top of the just-resolved
-	// Config. profilemanager's apply() initialises the policy to
-	// empty — the Loader lives outside Config, so this overlay step
-	// is driven externally here.
-	config.ApplyMDMPolicy(s.mdmLoader.Load())
-
 	return config, configExisted, nil
 }
 
@@ -1455,9 +1434,6 @@ func (s *Server) logoutFromProfile(ctx context.Context, profile *profilemanager.
 	if err != nil {
 		return fmt.Errorf("profile '%s' not found", profile.ID)
 	}
-	// Honour any MDM-enforced ManagementURL when issuing the logout
-	// RPC: the user-stored value may have been overridden by policy.
-	config.ApplyMDMPolicy(s.mdmLoader.Load())
 
 	return s.sendLogoutRequestWithConfig(ctx, config)
 }
@@ -2068,11 +2044,6 @@ func (s *Server) GetConfig(ctx context.Context, req *proto.GetConfigRequest) (*p
 		log.Errorf("failed to get active profile config: %v", err)
 		return nil, fmt.Errorf("failed to get active profile config: %w", err)
 	}
-	// Overlay the active MDM policy so the response's MDMManagedFields
-	// list reflects what the GUI / CLI must render as read-only.
-	// profilemanager.GetConfig itself returns a Config without the
-	// overlay (Loader lives outside profilemanager).
-	cfg.ApplyMDMPolicy(s.mdmLoader.Load())
 	managementURL := cfg.ManagementURL
 	adminURL := cfg.AdminURL
 

client/server/setconfig_mdm_test.go

@@ -16,40 +16,14 @@ import (
 	"github.com/netbirdio/netbird/client/proto"
 )
 
-// fakeMDMFetcher implements mdm.PolicyFetcher returning a pre-set
-// policy map. Tests build one per Server instance to inject a
-// scripted MDM overlay via a Loader rather than via package-level state.
-type fakeMDMFetcher struct{ values map[string]any }
-
-func (f *fakeMDMFetcher) Fetch() map[string]any { return f.values }
-
-// withMDMPolicy installs an mdm.Loader on the given Server whose
-// loadPlatform returns the supplied Policy's underlying values. Use
-// after setupServerWithProfile to inject the scripted policy the
-// SetConfig / Login MDM gates will observe.
-func withMDMPolicy(t *testing.T, s *Server, policy *mdm.Policy) {
+// withMDMPolicy temporarily overrides the server-package loadMDMPolicy hook
+// so SetConfig observes the supplied Policy. Restores the original loader
+// at test cleanup.
+func withMDMPolicy(t *testing.T, policy *mdm.Policy) {
 	t.Helper()
-	values := map[string]any{}
-	if policy != nil {
-		for _, k := range policy.ManagedKeys() {
-			if v, ok := policy.GetString(k); ok {
-				values[k] = v
-				continue
-			}
-			if v, ok := policy.GetBool(k); ok {
-				values[k] = v
-				continue
-			}
-			if v, ok := policy.GetInt(k); ok {
-				values[k] = v
-				continue
-			}
-			if v, ok := policy.GetStringSlice(k); ok {
-				values[k] = v
-			}
-		}
-	}
-	s.mdmLoader = mdm.NewLoader(&fakeMDMFetcher{values: values})
+	prev := loadMDMPolicy
+	loadMDMPolicy = func() *mdm.Policy { return policy }
+	t.Cleanup(func() { loadMDMPolicy = prev })
 }
 
 // setupServerWithProfile mirrors the boilerplate of TestSetConfig_AllFieldsSaved:
@@ -119,11 +93,12 @@ func extractViolation(t *testing.T, err error) *proto.MDMManagedFieldsViolation
 }
 
 func TestSetConfig_MDMReject_SingleField(t *testing.T) {
-	s, ctx, profName, username, _ := setupServerWithProfile(t)
-	withMDMPolicy(t, s, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL: "https://mdm.example.com:443",
 	}))
 
+	s, ctx, profName, username, _ := setupServerWithProfile(t)
+
 	_, err := s.SetConfig(ctx, &proto.SetConfigRequest{
 		ProfileName:   profName,
 		Username:      username,
@@ -135,13 +110,14 @@ func TestSetConfig_MDMReject_SingleField(t *testing.T) {
 }
 
 func TestSetConfig_MDMReject_MultipleFields(t *testing.T) {
-	s, ctx, profName, username, _ := setupServerWithProfile(t)
-	withMDMPolicy(t, s, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL:    "https://mdm.example.com:443",
 		mdm.KeyBlockInbound:     true,
 		mdm.KeyRosenpassEnabled: true,
 	}))
 
+	s, ctx, profName, username, _ := setupServerWithProfile(t)
+
 	blockInbound := false
 	rosenpassEnabled := false
 	_, err := s.SetConfig(ctx, &proto.SetConfigRequest{
@@ -165,11 +141,12 @@ func TestSetConfig_MDMReject_AllOrNothing(t *testing.T) {
 	// enforced field AND a non-enforced field (RosenpassEnabled).
 	// The whole request must be rejected — non-conflicting fields are not
 	// applied either.
-	s, ctx, profName, username, cfgPath := setupServerWithProfile(t)
-	withMDMPolicy(t, s, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL: "https://mdm.example.com:443",
 	}))
 
+	s, ctx, profName, username, cfgPath := setupServerWithProfile(t)
+
 	rosenpassEnabled := true
 	_, err := s.SetConfig(ctx, &proto.SetConfigRequest{
 		ProfileName:      profName,
@@ -191,11 +168,12 @@ func TestSetConfig_MDMReject_AllOrNothing(t *testing.T) {
 func TestSetConfig_MDMAllow_NonManagedFields(t *testing.T) {
 	// MDM enforces ManagementURL but the user only writes RosenpassEnabled.
 	// Request must succeed.
-	s, ctx, profName, username, _ := setupServerWithProfile(t)
-	withMDMPolicy(t, s, mdm.NewPolicy(map[string]any{
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 		mdm.KeyManagementURL: "https://mdm.example.com:443",
 	}))
 
+	s, ctx, profName, username, _ := setupServerWithProfile(t)
+
 	rosenpassEnabled := true
 	resp, err := s.SetConfig(ctx, &proto.SetConfigRequest{
 		ProfileName:      profName,
@@ -224,11 +202,12 @@ func TestSetConfig_MDMAllow_ManagementURLPortNormalized(t *testing.T) {
 
 	for _, tc := range tests {
 		t.Run(tc.name, func(t *testing.T) {
-			s, ctx, profName, username, _ := setupServerWithProfile(t)
-			withMDMPolicy(t, s, mdm.NewPolicy(map[string]any{
+			withMDMPolicy(t, mdm.NewPolicy(map[string]any{
 				mdm.KeyManagementURL: tc.mdmURL,
 			}))
 
+			s, ctx, profName, username, _ := setupServerWithProfile(t)
+
 			rosenpassEnabled := true
 			resp, err := s.SetConfig(ctx, &proto.SetConfigRequest{
 				ProfileName:      profName,
@@ -245,8 +224,9 @@ func TestSetConfig_MDMAllow_ManagementURLPortNormalized(t *testing.T) {
 
 func TestSetConfig_MDMEmpty_NoEnforcement(t *testing.T) {
 	// No MDM policy active: any field can be written.
+	withMDMPolicy(t, mdm.NewPolicy(nil))
+
 	s, ctx, profName, username, _ := setupServerWithProfile(t)
-	withMDMPolicy(t, s, mdm.NewPolicy(nil))
 
 	resp, err := s.SetConfig(ctx, &proto.SetConfigRequest{
 		ProfileName:   profName,

client/ui/autostart_default.go

@@ -72,7 +72,7 @@ func netbirdFootprintExists() bool {
 // retrying autostart entry writes on every launch. A user's later disable in
 // Settings is never overridden: the marker guarantees at-most-once, ever.
 func applyAutostartDefault(ctx context.Context, autostart *services.Autostart, prefs *preferences.Store, prefsFileExisted bool) {
-	mdmDisabled := autostartDisabledByMDM(mdm.NewLoader(nil).Load())
+	mdmDisabled := autostartDisabledByMDM(mdm.LoadPolicy())
 
 	if mdmDisabled {
 		if enabled, err := autostart.IsEnabled(ctx); err != nil {

shared/management/client/grpc.go

@@ -21,6 +21,8 @@ import (
 	"google.golang.org/grpc/connectivity"
 
 	nbgrpc "github.com/netbirdio/netbird/client/grpc"
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/encryption"
 	"github.com/netbirdio/netbird/shared/management/domain"
@@ -62,6 +64,13 @@ type GrpcClient struct {
 	connStateCallbackLock sync.RWMutex
 	serverURL             string
 
+	// netState gates the stream retry loop on OS-reported network
+	// availability; nil (the default) disables gating.
+	netState *netstate.State
+
+	// sweeper cuts the transport connections on network change; nil disables it.
+	sweeper *netsweep.Sweeper
+
 	// syncStreamErr holds the last Sync stream error, or nil while the stream
 	// is established and healthy. GetServerKey succeeds even when the peer
 	// cannot sync (e.g. the server returns "settings not found"), so the
@@ -111,16 +120,43 @@ func MaxRecvMsgSize() int {
 	return size
 }
 
+// ClientOption configures optional GrpcClient behavior.
+type ClientOption func(*GrpcClient)
+
+// WithNetworkState injects the OS network availability state that gates the
+// stream retry loop; without it gating is disabled.
+func WithNetworkState(netState *netstate.State) ClientOption {
+	return func(c *GrpcClient) { c.netState = netState }
+}
+
+// WithSweeper injects the network change sweeper.
+func WithSweeper(sweeper *netsweep.Sweeper) ClientOption {
+	return func(c *GrpcClient) { c.sweeper = sweeper }
+}
+
 // NewClient creates a new client to Management service
-func NewClient(ctx context.Context, addr string, ourPrivateKey wgtypes.Key, tlsEnabled bool) (*GrpcClient, error) {
-	var conn *grpc.ClientConn
+func NewClient(ctx context.Context, addr string, ourPrivateKey wgtypes.Key, tlsEnabled bool, opts ...ClientOption) (*GrpcClient, error) {
+	// Options apply before dialing: the sweeper must wrap the first connection too.
+	c := &GrpcClient{
+		key:                   ourPrivateKey,
+		ctx:                   ctx,
+		connStateCallbackLock: sync.RWMutex{},
+		serverURL:             addr,
+	}
+	for _, opt := range opts {
+		opt(c)
+	}
 
 	var extraOpts []grpc.DialOption
 	if maxSize := MaxRecvMsgSize(); maxSize > 0 {
 		extraOpts = append(extraOpts, grpc.WithDefaultCallOptions(grpc.MaxCallRecvMsgSize(maxSize)))
 		log.Infof("management gRPC max receive message size set to %d bytes", maxSize)
 	}
+	if c.sweeper != nil {
+		extraOpts = append(extraOpts, nbgrpc.WithSweeper(c.sweeper))
+	}
 
+	var conn *grpc.ClientConn
 	operation := func() error {
 		var err error
 		conn, err = nbgrpc.CreateConnection(ctx, addr, tlsEnabled, wsproxy.ManagementComponent, extraOpts...)
@@ -136,16 +172,9 @@ func NewClient(ctx context.Context, addr string, ourPrivateKey wgtypes.Key, tlsE
 		return nil, err
 	}
 
-	realClient := proto.NewManagementServiceClient(conn)
-
-	return &GrpcClient{
-		key:                   ourPrivateKey,
-		realClient:            realClient,
-		ctx:                   ctx,
-		conn:                  conn,
-		connStateCallbackLock: sync.RWMutex{},
-		serverURL:             addr,
-	}, nil
+	c.conn = conn
+	c.realClient = proto.NewManagementServiceClient(conn)
+	return c, nil
 }
 
 // GetServerURL returns the management server URL
@@ -208,6 +237,16 @@ func (c *GrpcClient) withMgmtStream(
 ) error {
 	backOff := defaultBackoff(ctx)
 	operation := func() error {
+		// suspend reconnect attempts while the OS reports no usable network.
+		// Wait only errors on a cancelled context, which means shutdown, so
+		// stop the loop without reporting a failure.
+		if waited, err := c.netState.Wait(ctx); err != nil {
+			log.Debugf("management connection context has been canceled while offline, this usually indicates shutdown")
+			return nil
+		} else if waited {
+			backOff.Reset()
+		}
+
 		log.Debugf("management connection state %v", c.conn.GetState())
 		connState := c.conn.GetState()
 

shared/relay/client/client.go

@@ -14,6 +14,7 @@ import (
 
 	log "github.com/sirupsen/logrus"
 
+	"github.com/netbirdio/netbird/client/netsweep"
 	auth "github.com/netbirdio/netbird/shared/relay/auth/hmac"
 	"github.com/netbirdio/netbird/shared/relay/client/dialer"
 	netErr "github.com/netbirdio/netbird/shared/relay/client/dialer/net"
@@ -184,6 +185,10 @@ type Client struct {
 	// datagram-sized transport is avoided on subsequent connects. Shared via
 	// the manager.
 	transportFallback *transportFallback
+
+	// sweeper cuts the relay connection on network change; the read loop
+	// reports the disconnect and the guard reconnects. Shared via the manager.
+	sweeper *netsweep.Sweeper
 	// datagramFallbackTriggered guards a single fallback per connection so a
 	// burst of oversized datagrams triggers one reconnect, not many.
 	datagramFallbackTriggered atomic.Bool
@@ -393,6 +398,12 @@ func (c *Client) Close() error {
 }
 
 func (c *Client) connect(ctx context.Context) (*RelayAddr, error) {
+	// A sweep cancels this context, so a dial started on the old network
+	// aborts instead of waiting out its handshake timeout.
+	dial := c.sweeper.StartDial(ctx)
+	defer dial.Release()
+	ctx = dial.Ctx()
+
 	mode := transportModeFromEnv()
 	dialers := c.getDialers(mode)
 
@@ -417,12 +428,19 @@ func (c *Client) connect(ctx context.Context) (*RelayAddr, error) {
 			return nil, fmt.Errorf("dial via FQDN: %w", err)
 		}
 	}
-	c.relayConn = conn
-	c.datagramFallbackTriggered.Store(false)
+	// Read the transport off the concrete connection: the sweeper's wrapper
+	// embeds net.Conn only, so it does not promote Protocol().
 	if tc, ok := conn.(transportConn); ok {
 		c.transport = tc.Protocol()
 	}
 
+	conn, err := dial.WrapConn(conn)
+	if err != nil {
+		return nil, fmt.Errorf("register connection: %w", err)
+	}
+	c.relayConn = conn
+	c.datagramFallbackTriggered.Store(false)
+
 	instanceURL, err := c.handShake(ctx)
 	if err != nil {
 		cErr := conn.Close()

shared/relay/client/guard.go

@@ -7,6 +7,8 @@ import (
 
 	"github.com/cenkalti/backoff/v4"
 	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/netstate"
 )
 
 const defaultMaxBackoffInterval = 60 * time.Second
@@ -22,14 +24,19 @@ type Guard struct {
 	// attempts.
 	maxBackoffInterval time.Duration
 
+	// netState gates reconnect attempts on OS-reported network availability;
+	// nil disables gating.
+	netState *netstate.State
+
 	// lastErr is the error from the most recent failed reconnect attempt,
 	// surfaced as the home relay status while disconnected.
 	lastErr atomic.Pointer[error]
 }
 
 // NewGuard creates a new guard for the relay client. A non-positive
-// maxBackoffInterval falls back to defaultMaxBackoffInterval.
-func NewGuard(sp *ServerPicker, maxBackoffInterval time.Duration) *Guard {
+// maxBackoffInterval falls back to defaultMaxBackoffInterval. A nil netState
+// disables network availability gating.
+func NewGuard(sp *ServerPicker, maxBackoffInterval time.Duration, netState *netstate.State) *Guard {
 	if maxBackoffInterval <= 0 {
 		maxBackoffInterval = defaultMaxBackoffInterval
 	}
@@ -38,6 +45,7 @@ func NewGuard(sp *ServerPicker, maxBackoffInterval time.Duration) *Guard {
 		OnReconnected:      make(chan struct{}, 1),
 		serverPicker:       sp,
 		maxBackoffInterval: maxBackoffInterval,
+		netState:           netState,
 	}
 	return g
 }
@@ -70,11 +78,21 @@ func (g *Guard) StartReconnectTrys(ctx context.Context, relayClient *Client) {
 
 	// start a ticker to pick a new server
 	ticker := g.exponentTicker(ctx)
-	defer ticker.Stop()
+	defer func() {
+		ticker.Stop()
+	}()
 
 	for {
 		select {
 		case <-ticker.C:
+			// suspend reconnect attempts while the OS reports no usable network
+			if waited, err := g.netState.Wait(ctx); err != nil {
+				return
+			} else if waited {
+				ticker.Stop()
+				ticker = g.exponentTicker(ctx)
+				continue
+			}
 			if err := g.retry(ctx); err != nil {
 				log.Errorf("failed to pick new Relay server: %s", err)
 				g.setLastError(err)
@@ -104,6 +122,13 @@ func (g *Guard) tryToQuickReconnect(parentCtx context.Context, rc *Client) bool
 		return false
 	}
 
+	// Re-check after the wait: the disconnect that triggered this reconnect
+	// is often the first symptom of the network going away, so the
+	// availability flag typically arrives while we sleep here.
+	if !g.netState.IsOnline() {
+		return false
+	}
+
 	log.Infof("try to reconnect to Relay server: %s", rc.connectionURL)
 
 	if err := rc.Connect(parentCtx); err != nil {

shared/relay/client/manager.go

@@ -12,6 +12,8 @@ import (
 
 	log "github.com/sirupsen/logrus"
 
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	relayAuth "github.com/netbirdio/netbird/shared/relay/auth/hmac"
 )
 
@@ -65,6 +67,17 @@ func WithMaxBackoffInterval(d time.Duration) ManagerOption {
 	return func(m *Manager) { m.maxBackoffInterval = d }
 }
 
+// WithNetworkState injects the OS network availability state that gates the
+// reconnect guard; without it reconnect attempts are not gated.
+func WithNetworkState(netState *netstate.State) ManagerOption {
+	return func(m *Manager) { m.netState = netState }
+}
+
+// WithSweeper injects the network change sweeper.
+func WithSweeper(sweeper *netsweep.Sweeper) ManagerOption {
+	return func(m *Manager) { m.sweeper = sweeper }
+}
+
 // Manager is a manager for the relay client instances. It establishes one persistent connection to the given relay URL
 // and automatically reconnect to them in case disconnection.
 // The manager also manage temporary relay connection. If a client wants to communicate with a client on a
@@ -92,6 +105,8 @@ type Manager struct {
 
 	mtu                uint16
 	maxBackoffInterval time.Duration
+	netState           *netstate.State
+	sweeper            *netsweep.Sweeper
 
 	cleanupInterval      time.Duration
 	keepUnusedServerTime time.Duration
@@ -128,8 +143,9 @@ func NewManager(ctx context.Context, serverURLs []string, peerID string, mtu uin
 	for _, opt := range opts {
 		opt(m)
 	}
+	m.serverPicker.Sweeper = m.sweeper
 	m.serverPicker.ServerURLs.Store(serverURLs)
-	m.reconnectGuard = NewGuard(m.serverPicker, m.maxBackoffInterval)
+	m.reconnectGuard = NewGuard(m.serverPicker, m.maxBackoffInterval, m.netState)
 	return m
 }
 
@@ -354,6 +370,7 @@ func (m *Manager) openConnVia(ctx context.Context, serverAddress, peerKey string
 
 	relayClient := NewClientWithServerIP(serverAddress, serverIP, m.tokenStore, m.peerID, m.mtu)
 	relayClient.SetTransportFallback(m.transportFallback)
+	relayClient.sweeper = m.sweeper
 	err := relayClient.Connect(m.ctx)
 	if err != nil {
 		rt.Lock()

shared/relay/client/picker.go

@@ -9,6 +9,7 @@ import (
 
 	log "github.com/sirupsen/logrus"
 
+	"github.com/netbirdio/netbird/client/netsweep"
 	auth "github.com/netbirdio/netbird/shared/relay/auth/hmac"
 )
 
@@ -30,6 +31,7 @@ type ServerPicker struct {
 	MTU               uint16
 	ConnectionTimeout time.Duration
 	TransportFallback *transportFallback
+	Sweeper           *netsweep.Sweeper
 }
 
 func (sp *ServerPicker) PickServer(parentCtx context.Context) (*Client, error) {
@@ -73,6 +75,7 @@ func (sp *ServerPicker) startConnection(ctx context.Context, resultChan chan con
 	log.Infof("try to connecting to relay server: %s", url)
 	relayClient := NewClient(url, sp.TokenStore, sp.PeerID, sp.MTU)
 	relayClient.SetTransportFallback(sp.TransportFallback)
+	relayClient.sweeper = sp.Sweeper
 	err := relayClient.Connect(ctx)
 	resultChan <- connResult{
 		RelayClient: relayClient,

shared/signal/client/grpc.go

@@ -19,6 +19,8 @@ import (
 	"google.golang.org/grpc/status"
 
 	nbgrpc "github.com/netbirdio/netbird/client/grpc"
+	"github.com/netbirdio/netbird/client/netstate"
+	"github.com/netbirdio/netbird/client/netsweep"
 	"github.com/netbirdio/netbird/encryption"
 	"github.com/netbirdio/netbird/shared/management/client"
 	"github.com/netbirdio/netbird/shared/signal/proto"
@@ -65,6 +67,13 @@ type GrpcClient struct {
 	connStateCallback     ConnStateNotifier
 	connStateCallbackLock sync.RWMutex
 
+	// netState gates the Receive retry loop on OS-reported network
+	// availability; nil (the default) disables gating.
+	netState *netstate.State
+
+	// sweeper cuts the transport connections on network change; nil disables it.
+	sweeper *netsweep.Sweeper
+
 	onReconnectedListenerFn func()
 
 	decryptionWorker       *Worker
@@ -88,13 +97,43 @@ type GrpcClient struct {
 	watchdogWg            sync.WaitGroup
 }
 
-// NewClient creates a new Signal client
-func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled bool) (*GrpcClient, error) {
-	var conn *grpc.ClientConn
+// ClientOption configures optional GrpcClient behavior.
+type ClientOption func(*GrpcClient)
 
+// WithNetworkState injects the OS network availability state that gates the
+// Receive retry loop; without it gating is disabled.
+func WithNetworkState(netState *netstate.State) ClientOption {
+	return func(c *GrpcClient) { c.netState = netState }
+}
+
+// WithSweeper injects the network change sweeper.
+func WithSweeper(sweeper *netsweep.Sweeper) ClientOption {
+	return func(c *GrpcClient) { c.sweeper = sweeper }
+}
+
+// NewClient creates a new Signal client
+func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled bool, opts ...ClientOption) (*GrpcClient, error) {
+	// Options apply before dialing: the sweeper must wrap the first connection too.
+	c := &GrpcClient{
+		ctx:                   ctx,
+		key:                   key,
+		mux:                   sync.Mutex{},
+		status:                StreamDisconnected,
+		connStateCallbackLock: sync.RWMutex{},
+	}
+	for _, opt := range opts {
+		opt(c)
+	}
+
+	var extraOpts []grpc.DialOption
+	if c.sweeper != nil {
+		extraOpts = append(extraOpts, nbgrpc.WithSweeper(c.sweeper))
+	}
+
+	var conn *grpc.ClientConn
 	operation := func() error {
 		var err error
-		conn, err = nbgrpc.CreateConnection(ctx, addr, tlsEnabled, wsproxy.SignalComponent)
+		conn, err = nbgrpc.CreateConnection(ctx, addr, tlsEnabled, wsproxy.SignalComponent, extraOpts...)
 		if err != nil {
 			return fmt.Errorf("create connection: %w", err)
 		}
@@ -109,15 +148,9 @@ func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled boo
 
 	log.Debugf("connected to Signal Service: %v", conn.Target())
 
-	return &GrpcClient{
-		realClient:            proto.NewSignalExchangeClient(conn),
-		ctx:                   ctx,
-		signalConn:            conn,
-		key:                   key,
-		mux:                   sync.Mutex{},
-		status:                StreamDisconnected,
-		connStateCallbackLock: sync.RWMutex{},
-	}, nil
+	c.signalConn = conn
+	c.realClient = proto.NewSignalExchangeClient(conn)
+	return c, nil
 }
 
 func (c *GrpcClient) StreamConnected() bool {
@@ -168,6 +201,15 @@ func (c *GrpcClient) Receive(ctx context.Context, msgHandler func(msg *proto.Mes
 	var backOff = defaultBackoff(ctx)
 
 	operation := func() error {
+		// suspend reconnect attempts while the OS reports no usable network.
+		// Wait only errors on a cancelled context, which means shutdown, so
+		// stop the loop without reporting a failure.
+		if waited, err := c.netState.Wait(ctx); err != nil {
+			log.Debugf("signal connection context has been canceled while offline, this usually indicates shutdown")
+			return nil
+		} else if waited {
+			backOff.Reset()
+		}
 
 		c.notifyStreamDisconnected()