[client] Sweep connections on network loss via a shared netevents manager

Losing the last network only flipped the availability state: the dead
management, signal and relay sockets stayed silently connected until their
own timeouts, so the client kept reporting Connected with no network at all.

Introduce client/netevents with a Manager that ties the availability state,
the connection sweeper and the status recorder together, and move the
netstate and netsweep packages under it (netsweep renamed to sweep).
SetNetworkAvailable(false) now also sweeps the registered connections so
their owners redial and the listener reaches the NoNetwork state.

The Android and iOS bindings own a Manager instance and inject it through
the constructors; consumers hold the concrete *Manager whose nil zero value
reports always-online and never sweeps, with interfaces kept only as
parameter contracts. The relay guard settle wait moved into the Manager as
WaitSettled, removing the netevents import from the relay package.
This commit is contained in:
Zoltán Papp
2026-08-19 17:15:57 +02:00
parent 77791b5858
commit bd99d89bac
26 changed files with 382 additions and 253 deletions
+17 -18
View File
@@ -26,8 +26,7 @@ import (
"github.com/netbirdio/netbird/client/internal/routemanager"
"github.com/netbirdio/netbird/client/internal/stdnet"
"github.com/netbirdio/netbird/client/net"
"github.com/netbirdio/netbird/client/netstate"
"github.com/netbirdio/netbird/client/netsweep"
"github.com/netbirdio/netbird/client/netevents"
"github.com/netbirdio/netbird/client/system"
"github.com/netbirdio/netbird/formatter"
"github.com/netbirdio/netbird/route"
@@ -82,13 +81,10 @@ 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
// netMgr outlives engine restarts: it mirrors the OS connectivity, not
// the engine lifecycle. Run and RunWithoutLogin inject its state and
// sweeper into each new ConnectClient.
netMgr *netevents.Manager
stateMu sync.RWMutex
connectClient *internal.ConnectClient
@@ -153,16 +149,16 @@ func NewClient(androidSDKVersion int, deviceName string, uiVersion string, tunAd
net.SetAndroidProtectSocketFn(tunAdapter.ProtectSocket)
system.SetIFaceDiscover(iFaceDiscover)
recorder := peer.NewRecorder("")
return &Client{
deviceName: deviceName,
uiVersion: uiVersion,
tunAdapter: tunAdapter,
iFaceDiscover: iFaceDiscover,
recorder: peer.NewRecorder(""),
recorder: recorder,
ctxCancelLock: &sync.Mutex{},
networkChangeListener: networkChangeListener,
netState: netstate.New(),
sweeper: netsweep.New(),
netMgr: netevents.NewManager(recorder),
}
}
@@ -203,8 +199,9 @@ 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,
internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
internal.WithNetEvents(c.netMgr))
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
@@ -246,7 +243,7 @@ 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,
internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
internal.WithNetEvents(c.netMgr))
c.setState(cfg, cacheDir, cfgFile, connectClient)
return connectClient.RunOnAndroid(c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, slices.Clone(dns.items), dnsReadyListener, stateFile, cacheDir)
}
@@ -298,9 +295,12 @@ func (c *Client) GetTunSettings() (*TunSettings, error) {
// 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.
// Losing the last network also sweeps the registered connections: nothing can
// redial while offline, so the stale sockets would otherwise stay silently
// "connected" until their own timeouts and the client would keep reporting
// Connected with no network at all.
func (c *Client) SetNetworkAvailable(available bool) {
c.netState.Set(available)
c.recorder.SetNetworkAvailable(available)
c.netMgr.SetNetworkAvailable(available)
}
// NotifyNetworkChange marks the management, signal and relay connections
@@ -308,8 +308,7 @@ func (c *Client) SetNetworkAvailable(available bool) {
// whatever has not redialed on the new network by then. The engine and the
// TUN device stay untouched.
func (c *Client) NotifyNetworkChange() {
c.sweeper.MarkNetworkChange()
log.Infof("network change: connections marked stale")
c.netMgr.NotifyNetworkChange()
}
// DebugBundle generates a debug bundle, uploads it, and returns the upload key.
+7 -2
View File
@@ -16,9 +16,14 @@ import (
"google.golang.org/grpc"
nbnet "github.com/netbirdio/netbird/client/net"
"github.com/netbirdio/netbird/client/netsweep"
"github.com/netbirdio/netbird/client/netevents/sweep"
)
// Sweeper registers in-flight dials for the network change sweep.
type Sweeper interface {
StartDial(ctx context.Context) *sweep.Dial
}
func WithCustomDialer(_ bool, _ string) grpc.DialOption {
return grpc.WithContextDialer(dialContext)
}
@@ -26,7 +31,7 @@ func WithCustomDialer(_ bool, _ string) grpc.DialOption {
// 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 {
func WithSweeper(sweeper Sweeper) grpc.DialOption {
return grpc.WithContextDialer(func(ctx context.Context, addr string) (net.Conn, error) {
dial := sweeper.StartDial(ctx)
defer dial.Release()
+9 -2
View File
@@ -1,12 +1,19 @@
package grpc
import (
"context"
"google.golang.org/grpc"
"github.com/netbirdio/netbird/client/netsweep"
"github.com/netbirdio/netbird/client/netevents/sweep"
"github.com/netbirdio/netbird/util/wsproxy/client"
)
// Sweeper registers in-flight dials for the network change sweep.
type Sweeper interface {
StartDial(ctx context.Context) *sweep.Dial
}
// WithCustomDialer returns a gRPC dial option that uses WebSocket transport for WASM/JS environments.
// The component parameter specifies the WebSocket proxy component path (e.g., "/management", "/signal").
func WithCustomDialer(tlsEnabled bool, component string) grpc.DialOption {
@@ -14,6 +21,6 @@ func WithCustomDialer(tlsEnabled bool, component string) grpc.DialOption {
}
// WithSweeper is a no-op on WASM/JS: there is no network change signal.
func WithSweeper(_ *netsweep.Sweeper) grpc.DialOption {
func WithSweeper(_ Sweeper) grpc.DialOption {
return grpc.EmptyDialOption{}
}
+12 -5
View File
@@ -6,16 +6,19 @@ import (
"time"
"github.com/cenkalti/backoff/v4"
"github.com/netbirdio/netbird/client/netstate"
)
// ChangeWatcher exposes OS network availability transitions.
type ChangeWatcher interface {
Changed() <-chan struct{}
}
// Retry mirrors backoff.Retry, but the sleep between attempts also wakes on
// OS network availability transitions: an operation cut down by a network
// change retries the moment the network settles instead of sleeping through
// the recovery. A nil netState never fires, leaving plain backoff.Retry
// the recovery. A nil watcher never fires, leaving plain backoff.Retry
// behavior.
func Retry(ctx context.Context, operation backoff.Operation, bo backoff.BackOff, netState *netstate.State) error {
func Retry(ctx context.Context, operation backoff.Operation, bo backoff.BackOff, watcher ChangeWatcher) error {
bo.Reset()
for {
err := operation()
@@ -36,10 +39,14 @@ func Retry(ctx context.Context, operation backoff.Operation, bo backoff.BackOff,
return err
}
var changed <-chan struct{}
if watcher != nil {
changed = watcher.Changed()
}
timer := time.NewTimer(next)
select {
case <-timer.C:
case <-netState.Changed():
case <-changed:
timer.Stop()
case <-ctx.Done():
timer.Stop()
+1 -1
View File
@@ -10,7 +10,7 @@ import (
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"github.com/netbirdio/netbird/client/netstate"
"github.com/netbirdio/netbird/client/netevents/netstate"
)
func TestRetryWakesOnNetworkChange(t *testing.T) {
+14 -26
View File
@@ -38,8 +38,7 @@ 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"
"github.com/netbirdio/netbird/client/netevents"
cProto "github.com/netbirdio/netbird/client/proto"
"github.com/netbirdio/netbird/client/ssh"
sshconfig "github.com/netbirdio/netbird/client/ssh/config"
@@ -73,28 +72,17 @@ type ConnectClient struct {
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
// netEvents gates every reconnection loop on OS-reported network
// availability and sweeps connections on network change.
netEvents *netevents.Manager
}
// 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 }
// WithNetEvents injects the OS network event handling.
func WithNetEvents(events *netevents.Manager) ConnectClientOption {
return func(c *ConnectClient) { c.netEvents = events }
}
func NewConnectClient(
@@ -305,7 +293,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
}
// suspend connection attempts while the OS reports no usable network
if waited, err := c.netState.Wait(c.ctx); err != nil {
if waited, err := c.netEvents.Wait(c.ctx); err != nil {
return nil
} else if waited {
backOff.Reset()
@@ -323,7 +311,7 @@ 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,
mgm.WithNetworkState(c.netState), mgm.WithSweeper(c.sweeper))
mgm.WithNetEvents(c.netEvents))
if err != nil {
// On daemon shutdown / Down() the parent context is cancelled
// and the dial fails with "context canceled". Wrapping that
@@ -398,7 +386,7 @@ 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, c.netState, c.sweeper)
signalClient, err := connectToSignal(engineCtx, loginResp.GetNetbirdConfig(), myPrivateKey, c.netEvents)
if err != nil {
log.Error(err)
return wrapErr(err)
@@ -435,7 +423,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
}
relayManager := relayClient.NewManager(engineCtx, relayURLs, myPrivateKey.PublicKey().String(), engineConfig.MTU,
relayClient.WithNetworkState(c.netState), relayClient.WithSweeper(c.sweeper))
relayClient.WithNetEvents(c.netEvents))
c.statusRecorder.SetRelayMgr(relayManager)
if len(relayURLs) > 0 {
if token != nil {
@@ -463,7 +451,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
UpdateManager: c.updateManager,
ClientMetrics: c.clientMetrics,
MetricsCtx: c.ctx,
NetState: c.netState,
NetState: c.netEvents,
}, mobileDependency)
engine.SetSyncResponsePersistence(c.persistSyncResponse)
c.engine = engine
@@ -723,7 +711,7 @@ 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, netState *netstate.State, sweeper *netsweep.Sweeper) (*signal.GrpcClient, error) {
func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourPrivateKey wgtypes.Key, netEvents *netevents.Manager) (*signal.GrpcClient, error) {
var sigTLSEnabled bool
if wtConfig.Signal.Protocol == mgmProto.HostConfig_HTTPS {
sigTLSEnabled = true
@@ -732,7 +720,7 @@ func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourP
}
signalClient, err := signal.NewClient(ctx, wtConfig.Signal.Uri, ourPrivateKey, sigTLSEnabled,
signal.WithNetworkState(netState), signal.WithSweeper(sweeper))
signal.WithNetEvents(netEvents))
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)
+3 -3
View File
@@ -59,7 +59,7 @@ 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"
"github.com/netbirdio/netbird/client/netevents"
cProto "github.com/netbirdio/netbird/client/proto"
"github.com/netbirdio/netbird/client/system"
nbdns "github.com/netbirdio/netbird/dns"
@@ -184,7 +184,7 @@ type EngineServices struct {
MetricsCtx context.Context
// NetState gates the reconnection loops on OS-reported network
// availability; nil disables gating.
NetState *netstate.State
NetState *netevents.Manager
}
// Engine is a mechanism responsible for reacting on Signal and Management stream events and managing connections to the remote peers.
@@ -210,7 +210,7 @@ type Engine struct {
// netState gates the peer reconnection guards on OS-reported network
// availability; nil disables gating.
netState *netstate.State
netState *netevents.Manager
// STUNs is a list of STUN servers used by ICE
STUNs []*stun.URI
+2 -2
View File
@@ -26,7 +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/client/netevents"
"github.com/netbirdio/netbird/route"
relayClient "github.com/netbirdio/netbird/shared/relay/client"
)
@@ -97,7 +97,7 @@ type ConnConfig struct {
// NetworkState gates the reconnection guard on OS-reported network
// availability; nil disables gating.
NetworkState *netstate.State
NetworkState *netevents.Manager
}
type Conn struct {
+13 -6
View File
@@ -6,8 +6,6 @@ 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.
@@ -24,6 +22,12 @@ const (
type connStatusFunc func() ConnStatus
// NetworkWatcher is the availability view the guard gates reconnects on.
type NetworkWatcher interface {
IsOnline() bool
Changed() <-chan struct{}
}
// Guard is responsible for the reconnection logic.
// It will trigger to send an offer to the peer then has connection issues.
// Watch these events:
@@ -39,14 +43,14 @@ type Guard struct {
srWatcher *SRWatcher
// netState gates reconnect attempts on OS-reported network availability;
// nil disables gating.
netState *netstate.State
netState NetworkWatcher
relayedConnDisconnected chan struct{}
iCEConnDisconnected chan struct{}
}
// 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 {
func NewGuard(log *log.Entry, isConnectedFn connStatusFunc, timeout time.Duration, srWatcher *SRWatcher, netState NetworkWatcher) *Guard {
return &Guard{
log: log,
isConnectedOnAllWay: isConnectedFn,
@@ -104,14 +108,17 @@ func (g *Guard) reconnectLoopWithRetry(ctx context.Context, callback func()) {
iceState := &iceRetryState{log: g.log}
defer iceState.reset()
netChanged := g.netState.Changed()
var netChanged <-chan struct{}
if g.netState != nil {
netChanged = g.netState.Changed()
}
for {
select {
case <-tickerChannel:
// skip attempts while the OS reports no usable network; the
// netChanged case below resumes the loop once it returns
if !g.netState.IsOnline() {
if g.netState != nil && !g.netState.IsOnline() {
continue
}
switch g.isConnectedOnAllWay() {
@@ -9,7 +9,7 @@ import (
log "github.com/sirupsen/logrus"
"github.com/netbirdio/netbird/client/internal/peer/ice"
"github.com/netbirdio/netbird/client/netstate"
"github.com/netbirdio/netbird/client/netevents/netstate"
)
// newTestGuardWithNetState builds a guard with a realistic MaxInterval: the
+14 -17
View File
@@ -22,8 +22,7 @@ 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/netstate"
"github.com/netbirdio/netbird/client/netsweep"
"github.com/netbirdio/netbird/client/netevents"
"github.com/netbirdio/netbird/client/system"
"github.com/netbirdio/netbird/formatter"
"github.com/netbirdio/netbird/route"
@@ -84,12 +83,10 @@ 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
// netMgr outlives engine restarts: it mirrors the OS connectivity, not
// the engine lifecycle. Run injects its state and sweeper into each new
// ConnectClient.
netMgr *netevents.Manager
// preloadedConfig holds config loaded from JSON (used on tvOS where file writes are blocked)
preloadedConfig *profilemanager.Config
@@ -100,6 +97,7 @@ type Client struct {
// NewClient instantiate a new Client
func NewClient(cfgFile, stateFile, cacheDir, logFilePath, deviceName string, osVersion string, osName string, networkChangeListener NetworkChangeListener, dnsManager DnsManager) *Client {
recorder := peer.NewRecorder("")
return &Client{
cfgFile: cfgFile,
stateFile: stateFile,
@@ -108,12 +106,11 @@ func NewClient(cfgFile, stateFile, cacheDir, logFilePath, deviceName string, osV
deviceName: deviceName,
osName: osName,
osVersion: osVersion,
recorder: peer.NewRecorder(""),
recorder: recorder,
ctxCancelLock: &sync.Mutex{},
networkChangeListener: networkChangeListener,
dnsManager: dnsManager,
netState: netstate.New(),
sweeper: netsweep.New(),
netMgr: netevents.NewManager(recorder),
}
}
@@ -190,7 +187,7 @@ func (c *Client) Run(fd int32, interfaceName string, envList *EnvList) error {
cfg.WgIface = interfaceName
connectClient := internal.NewConnectClient(ctx, cfg, c.recorder,
internal.WithNetworkState(c.netState), internal.WithSweeper(c.sweeper))
internal.WithNetEvents(c.netMgr))
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
@@ -203,10 +200,11 @@ func (c *Client) Run(fd int32, interfaceName string, envList *EnvList) error {
// (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.
// immediately with a fresh backoff. Losing the last network also sweeps the
// registered connections, so the client does not keep reporting Connected
// over stale sockets with no network at all.
func (c *Client) SetNetworkAvailable(available bool) {
c.netState.Set(available)
c.recorder.SetNetworkAvailable(available)
c.netMgr.SetNetworkAvailable(available)
}
// NotifyNetworkChange marks the management, signal and relay connections
@@ -214,8 +212,7 @@ func (c *Client) SetNetworkAvailable(available bool) {
// whatever has not redialed on the new network by then. The engine and the
// TUN device stay untouched.
func (c *Client) NotifyNetworkChange() {
c.sweeper.MarkNetworkChange()
log.Infof("network change: connections marked stale")
c.netMgr.NotifyNetworkChange()
}
// Stop the internal client and free the resources
+158
View File
@@ -0,0 +1,158 @@
// Package netevents owns the OS network event handling shared by the mobile
// bindings: availability changes park or wake the reconnection loops and drive
// the NoNetwork listener state, and both losing the last network and switching
// networks sweep the stale connections so their owners redial immediately.
package netevents
import (
"context"
"time"
"github.com/cenkalti/backoff/v4"
log "github.com/sirupsen/logrus"
"github.com/netbirdio/netbird/client/netevents/netstate"
"github.com/netbirdio/netbird/client/netevents/sweep"
)
// Recorder receives the availability changes for listener state reporting.
type Recorder interface {
SetNetworkAvailable(available bool)
}
// Manager ties the network availability state, the connection sweeper and the
// status recorder together; it outlives engine restarts. A nil *Manager is
// the valid no-events value: the read methods report always-online and never
// sweep.
type Manager struct {
netState *netstate.State
sweeper *sweep.Sweeper
recorder Recorder
}
// NewManager creates a Manager reporting into recorder, starting online.
func NewManager(recorder Recorder) *Manager {
return &Manager{
netState: netstate.New(),
sweeper: sweep.New(),
recorder: recorder,
}
}
// SetNetworkAvailable records OS-reported network availability. While
// unavailable, the reconnection loops suspend their attempts and the
// connection listener reports NoNetwork instead of Connecting; when
// availability returns, the loops resume immediately with a fresh backoff.
// Losing the last network also sweeps the registered connections: nothing can
// redial while offline, so the stale sockets would otherwise stay silently
// "connected" until their own timeouts and the client would keep reporting
// Connected with no network at all.
func (m *Manager) SetNetworkAvailable(available bool) {
if !available && m.netState.IsOnline() {
m.sweeper.MarkNetworkChange()
}
m.netState.Set(available)
m.recorder.SetNetworkAvailable(available)
}
// NotifyNetworkChange marks the management, signal and relay connections
// stale after the OS switched networks and schedules a sweep that cuts
// whatever has not redialed on the new network by then. The engine and the
// TUN device stay untouched.
func (m *Manager) NotifyNetworkChange() {
m.sweeper.MarkNetworkChange()
log.Infof("network change: connections marked stale")
}
// IsOnline reports whether the OS reports at least one usable network.
func (m *Manager) IsOnline() bool {
if m == nil {
return true
}
return m.netState.IsOnline()
}
// Changed returns a channel closed on the next availability transition.
func (m *Manager) Changed() <-chan struct{} {
if m == nil {
return nil
}
return m.netState.Changed()
}
// Wait blocks while the network is offline; see netstate.State.Wait.
func (m *Manager) Wait(ctx context.Context) (bool, error) {
if m == nil {
return false, nil
}
return m.netState.Wait(ctx)
}
// WaitSettled waits until an online verdict holds for a full settleWindow, or
// while offline until the budget runs out. Returns false when ctx is
// cancelled. The settle window exists because a disconnect often precedes the
// OS offline flag by a few milliseconds, so a fresh online verdict cannot be
// trusted immediately. A nil Manager has no events to watch: it degrades to a
// fixed budget-long sleep.
func (m *Manager) WaitSettled(ctx context.Context, budget, settleWindow time.Duration) bool {
if m == nil {
select {
case <-time.After(budget):
return true
case <-ctx.Done():
return false
}
}
budgetTimer := time.NewTimer(budget)
defer budgetTimer.Stop()
settle := time.NewTimer(settleWindow)
defer settle.Stop()
for {
// Channel first, flag second: a flip in between still fires the channel.
changedCh := m.netState.Changed()
if m.netState.IsOnline() {
select {
case <-settle.C:
return true
case <-changedCh:
case <-ctx.Done():
return false
}
} else {
select {
case <-budgetTimer.C:
return true
case <-changedCh:
case <-ctx.Done():
return false
}
}
if !settle.Stop() {
select {
case <-settle.C:
default:
}
}
settle.Reset(settleWindow)
}
}
// StartDial registers an in-flight dial with the sweeper; see sweep.Sweeper.StartDial.
func (m *Manager) StartDial(ctx context.Context) *sweep.Dial {
if m == nil {
return (*sweep.Sweeper)(nil).StartDial(ctx)
}
return m.sweeper.StartDial(ctx)
}
// QuickRetryBackoff wraps bo for a quick retry after a network change; see
// sweep.Sweeper.QuickRetryBackoff.
func (m *Manager) QuickRetryBackoff(ctx context.Context, bo backoff.BackOff) backoff.BackOff {
if m == nil {
return bo
}
return m.sweeper.QuickRetryBackoff(ctx, bo, m.netState)
}
+34
View File
@@ -0,0 +1,34 @@
package netevents
import (
"context"
"testing"
"time"
"github.com/stretchr/testify/assert"
)
type recorderStub struct{}
func (recorderStub) SetNetworkAvailable(bool) {}
func TestWaitSettledAfterOutage(t *testing.T) {
const budget = 1500 * time.Millisecond
const settleWindow = 200 * time.Millisecond
const outage = 2 * settleWindow
m := NewManager(recorderStub{})
m.SetNetworkAvailable(false)
start := time.Now()
go func() {
time.Sleep(outage)
m.SetNetworkAvailable(true)
}()
ok := m.WaitSettled(context.Background(), budget, settleWindow)
elapsed := time.Since(start)
assert.True(t, ok, "recovered network must let the caller proceed")
assert.GreaterOrEqual(t, elapsed, outage+settleWindow, "an online verdict must hold a full settle window before it is trusted")
}
@@ -1,11 +1,11 @@
package netsweep
package sweep
import (
"time"
"github.com/cenkalti/backoff/v4"
"github.com/netbirdio/netbird/client/netstate"
"github.com/netbirdio/netbird/client/netevents/netstate"
)
const quickRetryDelay = 200 * time.Millisecond
@@ -1,4 +1,4 @@
package netsweep
package sweep
import (
"context"
@@ -1,10 +1,10 @@
// Package netsweep cuts network-bound activity when the OS switches networks:
// Package sweep 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
package sweep
import (
"context"
@@ -16,7 +16,7 @@ import (
"github.com/cenkalti/backoff/v4"
log "github.com/sirupsen/logrus"
"github.com/netbirdio/netbird/client/netstate"
"github.com/netbirdio/netbird/client/netevents/netstate"
)
// DefaultSweepDelay absorbs network flapping while the OS settles on a
@@ -34,7 +34,7 @@ type Config struct {
// 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")
var ErrSwept = errors.New("sweep: connection swept by network change")
// sweepID identifies one registration in a sweeper. Connections and dials
// draw from the same counter, so an id is unique across both registries.
@@ -1,4 +1,4 @@
package netsweep
package sweep
import (
"context"