Files
netbird/client/internal/peer/guard/guard.go
Zoltán Papp 53f2518af1 [client] Resume peer reconnection when the network comes back
Ticks taken while the OS reported no network were skipped, but they still
advanced the exponential backoff, so a peer could be tens of seconds from its
next attempt by the time connectivity returned. Recovery then depended on a
signal or relay event, which never arrives when both stayed up across the
outage, e.g. a short airplane mode toggle over Wi-Fi. A peer parked in ICE
hourly mode stayed there for the same reason.

React to the offline-to-online transition directly: re-arm the reconnect
ticker and reset the ICE retry state, so the peer retries at once. Add
netstate.Changed for callers that own a select loop and cannot block in Wait.
2026-08-12 09:43:50 +02:00

206 lines
6.5 KiB
Go

package guard
import (
"context"
"time"
"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.
type ConnStatus int
const (
// ConnStatusDisconnected means neither ICE nor Relay is connected.
ConnStatusDisconnected ConnStatus = iota
// ConnStatusPartiallyConnected means Relay is connected but ICE is not.
ConnStatusPartiallyConnected
// ConnStatusConnected means all required connections are established.
ConnStatusConnected
)
type connStatusFunc func() ConnStatus
// Guard is responsible for the reconnection logic.
// It will trigger to send an offer to the peer then has connection issues.
// Watch these events:
// - Relay client reconnected to home server
// - Signal server connection state changed
// - ICE connection disconnected
// - Relayed connection disconnected
// - ICE candidate changes
type Guard struct {
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{}
}
// 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),
}
}
func (g *Guard) Start(ctx context.Context, eventCallback func()) {
g.log.Infof("starting guard for reconnection with MaxInterval: %s", g.timeout)
g.reconnectLoopWithRetry(ctx, eventCallback)
}
func (g *Guard) SetRelayedConnDisconnected() {
select {
case g.relayedConnDisconnected <- struct{}{}:
default:
}
}
func (g *Guard) SetICEConnDisconnected() {
select {
case g.iCEConnDisconnected <- struct{}{}:
default:
}
}
// reconnectLoopWithRetry periodically checks the connection status and sends offers to re-establish connectivity.
//
// Behavior depends on the connection state reported by isConnectedOnAllWay:
// - Connected: no action, the peer is fully reachable.
// - Disconnected (neither ICE nor Relay): retries aggressively with exponential backoff (800ms doubling
// up to timeout), never gives up. This ensures rapid recovery when the peer has no connectivity at all.
// - PartiallyConnected (Relay up, ICE not): retries up to 3 times with exponential backoff, then switches
// to one attempt per hour. This limits signaling traffic when relay already provides connectivity.
//
// External events (relay/ICE disconnect, signal/relay reconnect, candidate changes) reset the retry
// counter and backoff ticker, giving ICE a fresh chance after network conditions change.
func (g *Guard) reconnectLoopWithRetry(ctx context.Context, callback func()) {
srReconnectedChan := g.srWatcher.NewListener()
defer g.srWatcher.RemoveListener(srReconnectedChan)
ticker := g.initialTicker(ctx)
defer func() {
// If backoff.Ticker.send is blocked, context.Done will not close the Ticker goroutine.
// We have to explicitly call Stop, even if we use backoff.WithContext.
ticker.Stop()
}()
tickerChannel := ticker.C
iceState := &iceRetryState{log: g.log}
defer iceState.reset()
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() {
continue
}
switch g.isConnectedOnAllWay() {
case ConnStatusConnected:
// all good, nothing to do
case ConnStatusDisconnected:
callback()
case ConnStatusPartiallyConnected:
if iceState.shouldRetry() {
callback()
} else {
iceState.enterHourlyMode()
ticker.Stop()
tickerChannel = iceState.hourlyC()
}
}
case <-g.relayedConnDisconnected:
g.log.Debugf("Relay connection changed, reset reconnection ticker")
ticker.Stop()
ticker = g.newReconnectTicker(ctx)
tickerChannel = ticker.C
iceState.reset()
case <-g.iCEConnDisconnected:
g.log.Debugf("ICE connection changed, reset reconnection ticker")
ticker.Stop()
ticker = g.newReconnectTicker(ctx)
tickerChannel = ticker.C
iceState.reset()
case <-srReconnectedChan:
g.log.Debugf("has network changes, reset reconnection ticker")
ticker.Stop()
ticker = g.newReconnectTicker(ctx)
tickerChannel = ticker.C
iceState.reset()
case <-netChanged:
// Re-arm for the next transition before acting on this one.
netChanged = g.netState.Changed()
if !g.netState.IsOnline() {
continue
}
// Ticks skipped while offline drove the backoff towards its
// maximum without ever attempting, and left the ICE budget
// frozen — possibly in hourly mode. Recover on our own so the
// peer does not depend on a signal or relay event that never
// comes when both stayed up across the outage.
g.log.Debugf("network is back, reset reconnection ticker")
ticker.Stop()
ticker = g.newReconnectTicker(ctx)
tickerChannel = ticker.C
iceState.reset()
case <-ctx.Done():
g.log.Debugf("context is done, stop reconnect loop")
return
}
}
}
// initialTicker give chance to the peer to establish the initial connection.
func (g *Guard) initialTicker(ctx context.Context) *backoff.Ticker {
bo := backoff.WithContext(&backoff.ExponentialBackOff{
InitialInterval: 3 * time.Second,
RandomizationFactor: 0.1,
Multiplier: 2,
MaxInterval: g.timeout,
Stop: backoff.Stop,
Clock: backoff.SystemClock,
}, ctx)
return backoff.NewTicker(bo)
}
func (g *Guard) newReconnectTicker(ctx context.Context) *backoff.Ticker {
bo := backoff.WithContext(&backoff.ExponentialBackOff{
InitialInterval: 800 * time.Millisecond,
RandomizationFactor: 0.1,
Multiplier: 2,
MaxInterval: g.timeout,
Stop: backoff.Stop,
Clock: backoff.SystemClock,
}, ctx)
ticker := backoff.NewTicker(bo)
<-ticker.C // consume the initial tick what is happening right after the ticker has been created
return ticker
}