mirror of
https://github.com/netbirdio/netbird.git
synced 2026-10-02 03:29:07 +02:00
[client] Suspend reconnection loops while the OS reports no network
On mobile the client kept dialing management, signal, relay and peer connections while the device had no usable network at all (airplane mode), burning battery for attempts that cannot succeed. Stopping the engine is not an option: tearing it down destroys the TUN device, and traffic can leak outside the tunnel until it is rebuilt. Add client/netstate, a small gate the platform feeds from its own connectivity callbacks. Every reconnection loop waits on it instead of retrying blindly, and resets its backoff when the network returns so recovery is immediate. The state is injected through functional options and consumers hold a *State that may be nil, so every platform that does not report availability behaves exactly as before. The relay quick-reconnect rechecks availability after its 1.5s wait: the disconnect that triggers it is usually the first symptom of the network going away, so the flag typically arrives while it sleeps. Report the suspension to the UI as well. peer.Listener grows OnStateChanged with a typed ClientState, re-exported across the gomobile boundary as integer constants, and the notifier maps Connecting to a new NoNetwork state while the OS reports no network, so mobile clients can show "no network available" instead of a misleading "connecting". Finally, exit the client retry loop cleanly when its context is cancelled. backoff.WithContext surfaces the bare context error, which callers could not distinguish from a real failure — on Android that turned an engine restart into an unrecoverable error.
This commit is contained in:
@@ -19,6 +19,7 @@ import (
|
||||
"google.golang.org/grpc/status"
|
||||
|
||||
nbgrpc "github.com/netbirdio/netbird/client/grpc"
|
||||
"github.com/netbirdio/netbird/client/netstate"
|
||||
"github.com/netbirdio/netbird/encryption"
|
||||
"github.com/netbirdio/netbird/shared/management/client"
|
||||
"github.com/netbirdio/netbird/shared/signal/proto"
|
||||
@@ -65,6 +66,10 @@ 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
|
||||
|
||||
onReconnectedListenerFn func()
|
||||
|
||||
decryptionWorker *Worker
|
||||
@@ -89,7 +94,16 @@ type GrpcClient struct {
|
||||
}
|
||||
|
||||
// NewClient creates a new Signal client
|
||||
func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled bool) (*GrpcClient, error) {
|
||||
// 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 }
|
||||
}
|
||||
|
||||
func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled bool, opts ...ClientOption) (*GrpcClient, error) {
|
||||
var conn *grpc.ClientConn
|
||||
|
||||
operation := func() error {
|
||||
@@ -109,7 +123,7 @@ func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled boo
|
||||
|
||||
log.Debugf("connected to Signal Service: %v", conn.Target())
|
||||
|
||||
return &GrpcClient{
|
||||
c := &GrpcClient{
|
||||
realClient: proto.NewSignalExchangeClient(conn),
|
||||
ctx: ctx,
|
||||
signalConn: conn,
|
||||
@@ -117,7 +131,11 @@ func NewClient(ctx context.Context, addr string, key wgtypes.Key, tlsEnabled boo
|
||||
mux: sync.Mutex{},
|
||||
status: StreamDisconnected,
|
||||
connStateCallbackLock: sync.RWMutex{},
|
||||
}, nil
|
||||
}
|
||||
for _, opt := range opts {
|
||||
opt(c)
|
||||
}
|
||||
return c, nil
|
||||
}
|
||||
|
||||
func (c *GrpcClient) StreamConnected() bool {
|
||||
@@ -168,6 +186,15 @@ 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()
|
||||
|
||||
|
||||
Reference in New Issue
Block a user