pickfirst: Implement Happy Eyeballs

balancer/pickfirst/internal/internal.go

@@ -18,7 +18,18 @@
 // Package internal contains code internal to the pickfirst package.
 package internal
 
-import "math/rand"
+import (
+	rand "math/rand/v2"
+	"time"
+)
 
-// RandShuffle pseudo-randomizes the order of addresses.
-var RandShuffle = rand.Shuffle
+var (
+	// RandShuffle pseudo-randomizes the order of addresses.
+	RandShuffle = rand.Shuffle
+	// TimeAfterFunc allows mocking the timer for testing connection delay
+	// related functionality.
+	TimeAfterFunc = func(d time.Duration, f func()) func() {
+		timer := time.AfterFunc(d, f)
+		return func() { timer.Stop() }
+	}
+)

balancer/pickfirst/pickfirstleaf/pickfirstleaf.go

@@ -31,6 +31,7 @@
 	"fmt"
 	"net"
 	"sync"
+	"time"
 
 	"google.golang.org/grpc/balancer"
 	"google.golang.org/grpc/balancer/pickfirst/internal"
@@ -59,8 +60,13 @@
 	Name = "pick_first_leaf"
 )
 
-// TODO: change to pick-first when this becomes the default pick_first policy.
-const logPrefix = "[pick-first-leaf-lb %p] "
+const (
+	// TODO: change to pick-first when this becomes the default pick_first policy.
+	logPrefix = "[pick-first-leaf-lb %p] "
+	// connectionDelayInterval is the time to wait for during the happy eyeballs
+	// pass before starting the next connection attempt.
+	connectionDelayInterval = 250 * time.Millisecond
+)
 
 type ipAddrFamily int
 
@@ -76,11 +82,12 @@
 
 func (pickfirstBuilder) Build(cc balancer.ClientConn, _ balancer.BuildOptions) balancer.Balancer {
 	b := &pickfirstBalancer{
-		cc:          cc,
-		addressList: addressList{},
-		subConns:    resolver.NewAddressMap(),
-		state:       connectivity.Connecting,
-		mu:          sync.Mutex{},
+		cc:                    cc,
+		addressList:           addressList{},
+		subConns:              resolver.NewAddressMap(),
+		state:                 connectivity.Connecting,
+		mu:                    sync.Mutex{},
+		cancelConnectionTimer: func() {},
 	}
 	b.logger = internalgrpclog.NewPrefixLogger(logger, fmt.Sprintf(logPrefix, b))
 	return b
@@ -115,8 +122,9 @@
 	subConn balancer.SubConn
 	addr    resolver.Address
 
-	state   connectivity.State
-	lastErr error
+	state            connectivity.State
+	lastErr          error
+	connectionFailed bool
 }
 
 func (b *pickfirstBalancer) newSCData(addr resolver.Address) (*scData, error) {
@@ -148,10 +156,11 @@
 	mu    sync.Mutex
 	state connectivity.State
 	// scData for active subonns mapped by address.
-	subConns    *resolver.AddressMap
-	addressList addressList
-	firstPass   bool
-	numTF       int
+	subConns              *resolver.AddressMap
+	addressList           addressList
+	firstPass             bool
+	numTF                 int
+	cancelConnectionTimer func()
 }
 
 // ResolverError is called by the ClientConn when the name resolver produces
@@ -186,6 +195,7 @@
 func (b *pickfirstBalancer) UpdateClientConnState(state balancer.ClientConnState) error {
 	b.mu.Lock()
 	defer b.mu.Unlock()
+	b.cancelConnectionTimer()
 	if len(state.ResolverState.Addresses) == 0 && len(state.ResolverState.Endpoints) == 0 {
 		// Cleanup state pertaining to the previous resolver state.
 		// Treat an empty address list like an error by calling b.ResolverError.
@@ -239,12 +249,8 @@
 	// Not de-duplicating would result in attempting to connect to the same
 	// SubConn multiple times in the same pass. We don't want this.
 	newAddrs = deDupAddresses(newAddrs)
-
 	newAddrs = interleaveAddresses(newAddrs)
 
-	// Since we have a new set of addresses, we are again at first pass.
-	b.firstPass = true
-
 	// If the previous ready SubConn exists in new address list,
 	// keep this connection and don't create new SubConns.
 	prevAddr := b.addressList.currentAddress()
@@ -269,11 +275,11 @@
 			ConnectivityState: connectivity.Connecting,
 			Picker:            &picker{err: balancer.ErrNoSubConnAvailable},
 		})
-		b.requestConnectionLocked()
+		b.startFirstPassLocked()
 	} else if b.state == connectivity.TransientFailure {
 		// If we're in TRANSIENT_FAILURE, we stay in TRANSIENT_FAILURE until
 		// we're READY. See A62.
-		b.requestConnectionLocked()
+		b.startFirstPassLocked()
 	}
 	return nil
 }
@@ -288,6 +294,7 @@
 	b.mu.Lock()
 	defer b.mu.Unlock()
 	b.closeSubConnsLocked()
+	b.cancelConnectionTimer()
 	b.state = connectivity.Shutdown
 }
 
@@ -297,12 +304,21 @@
 func (b *pickfirstBalancer) ExitIdle() {
 	b.mu.Lock()
 	defer b.mu.Unlock()
-	if b.state == connectivity.Idle && b.addressList.currentAddress() == b.addressList.first() {
-		b.firstPass = true
-		b.requestConnectionLocked()
+	if b.state == connectivity.Idle {
+		b.startFirstPassLocked()
 	}
 }
 
+func (b *pickfirstBalancer) startFirstPassLocked() {
+	b.firstPass = true
+	b.numTF = 0
+	// Reset the connection attempt record for existing SubConns.
+	for _, sd := range b.subConns.Values() {
+		sd.(*scData).connectionFailed = false
+	}
+	b.requestConnectionLocked()
+}
+
 func (b *pickfirstBalancer) closeSubConnsLocked() {
 	for _, sd := range b.subConns.Values() {
 		sd.(*scData).subConn.Shutdown()
@@ -413,6 +429,7 @@
 // shutdownRemainingLocked shuts down remaining subConns. Called when a subConn
 // becomes ready, which means that all other subConn must be shutdown.
 func (b *pickfirstBalancer) shutdownRemainingLocked(selected *scData) {
+	b.cancelConnectionTimer()
 	for _, v := range b.subConns.Values() {
 		sd := v.(*scData)
 		if sd.subConn != selected.subConn {
@@ -456,30 +473,73 @@
 		switch scd.state {
 		case connectivity.Idle:
 			scd.subConn.Connect()
+			b.scheduleNextConnectionLocked()
+			return
 		case connectivity.TransientFailure:
-			// Try the next address.
+			// The SubConn is being re-used and failed during a previous pass
+			// over the addressList. It has not completed backoff yet.
+			// Mark it as having failed and try the next address.
+			scd.connectionFailed = true
 			lastErr = scd.lastErr
 			continue
 		case connectivity.Ready:
 			// Should never happen.
 			b.logger.Errorf("Requesting a connection even though we have a READY SubConn")
+			return
 		case connectivity.Shutdown:
 			// Should never happen.
 			b.logger.Errorf("SubConn with state SHUTDOWN present in SubConns map")
+			return
 		case connectivity.Connecting:
-			// Wait for the SubConn to report success or failure.
+			// Wait for the connection attempt to complete or the timer to fire
+			// before attempting the next address.
+			b.scheduleNextConnectionLocked()
+			return
 		}
-		return
 	}
+
 	// All the remaining addresses in the list are in TRANSIENT_FAILURE, end the
-	// first pass.
-	b.endFirstPassLocked(lastErr)
+	// first pass if possible.
+	b.endFirstPassIfPossibleLocked(lastErr)
+}
+
+func (b *pickfirstBalancer) scheduleNextConnectionLocked() {
+	b.cancelConnectionTimer()
+	if !b.addressList.hasNext() {
+		return
+	}
+	curAddr := b.addressList.currentAddress()
+	cancelled := false // Access to this is protected by the balancer's mutex.
+	closeFn := internal.TimeAfterFunc(connectionDelayInterval, func() {
+		b.mu.Lock()
+		defer b.mu.Unlock()
+		// If the scheduled task is cancelled while acquiring the mutex, return.
+		if cancelled {
+			return
+		}
+		if b.logger.V(2) {
+			b.logger.Infof("Happy Eyeballs timer expired while waiting for connection to %q.", curAddr.Addr)
+		}
+		if b.addressList.increment() {
+			b.requestConnectionLocked()
+		}
+	})
+	// Access to the cancellation callback held by the balancer is guarded by
+	// the balancer's mutex, so it's safe to set the boolean from the callback.
+	b.cancelConnectionTimer = sync.OnceFunc(func() {
+		cancelled = true
+		closeFn()
+	})
 }
 
 func (b *pickfirstBalancer) updateSubConnState(sd *scData, newState balancer.SubConnState) {
 	b.mu.Lock()
 	defer b.mu.Unlock()
 	oldState := sd.state
+	// Record a connection attempt when exiting CONNECTING.
+	if newState.ConnectivityState == connectivity.TransientFailure {
+		sd.connectionFailed = true
+	}
 	sd.state = newState.ConnectivityState
 	// Previously relevant SubConns can still callback with state updates.
 	// To prevent pickers from returning these obsolete SubConns, this logic
@@ -545,17 +605,20 @@
 			sd.lastErr = newState.ConnectionError
 			// Since we're re-using common SubConns while handling resolver
 			// updates, we could receive an out of turn TRANSIENT_FAILURE from
-			// a pass over the previous address list. We ignore such updates.
-
-			if curAddr := b.addressList.currentAddress(); !equalAddressIgnoringBalAttributes(&curAddr, &sd.addr) {
-				return
-			}
-			if b.addressList.increment() {
-				b.requestConnectionLocked()
-				return
+			// a pass over the previous address list. Happy Eyeballs will also
+			// cause out of order updates to arrive.
+
+			if curAddr := b.addressList.currentAddress(); equalAddressIgnoringBalAttributes(&curAddr, &sd.addr) {
+				b.cancelConnectionTimer()
+				if b.addressList.increment() {
+					b.requestConnectionLocked()
+					return
+				}
 			}
-			// End of the first pass.
-			b.endFirstPassLocked(newState.ConnectionError)
+
+			// End the first pass if we've seen a TRANSIENT_FAILURE from all
+			// SubConns once.
+			b.endFirstPassIfPossibleLocked(newState.ConnectionError)
 		}
 		return
 	}
@@ -580,9 +643,17 @@
 	}
 }
 
-func (b *pickfirstBalancer) endFirstPassLocked(lastErr error) {
+func (b *pickfirstBalancer) endFirstPassIfPossibleLocked(lastErr error) {
+	if b.addressList.isValid() || b.subConns.Len() < b.addressList.size() {
+		return
+	}
+	for _, v := range b.subConns.Values() {
+		sd := v.(*scData)
+		if !sd.connectionFailed {
+			return
+		}
+	}
 	b.firstPass = false
-	b.numTF = 0
 	b.state = connectivity.TransientFailure
 
 	b.cc.UpdateState(balancer.State{
@@ -654,15 +725,6 @@
 	return al.addresses[al.idx]
 }
 
-// first returns the first address in the list. If the list is empty, it returns
-// an empty address instead.
-func (al *addressList) first() resolver.Address {
-	if len(al.addresses) == 0 {
-		return resolver.Address{}
-	}
-	return al.addresses[0]
-}
-
 func (al *addressList) reset() {
 	al.idx = 0
 }
@@ -685,6 +747,16 @@
 	return false
 }
 
+// hasNext returns whether incrementing the addressList will result in moving
+// past the end of the list. If the list has already moved past the end, it
+// returns false.
+func (al *addressList) hasNext() bool {
+	if !al.isValid() {
+		return false
+	}
+	return al.idx+1 < len(al.addresses)
+}
+
 // equalAddressIgnoringBalAttributes returns true is a and b are considered
 // equal. This is different from the Equal method on the resolver.Address type
 // which considers all fields to determine equality. Here, we only consider