bf5c6b29fa
This PR implements the LES server RPC API. Methods for server capacity, client balance and client priority management are provided.
280 lines
9.2 KiB
Go
280 lines
9.2 KiB
Go
// Copyright 2016 The go-ethereum Authors
|
|
// This file is part of the go-ethereum library.
|
|
//
|
|
// The go-ethereum library is free software: you can redistribute it and/or modify
|
|
// it under the terms of the GNU Lesser General Public License as published by
|
|
// the Free Software Foundation, either version 3 of the License, or
|
|
// (at your option) any later version.
|
|
//
|
|
// The go-ethereum library is distributed in the hope that it will be useful,
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU Lesser General Public License for more details.
|
|
//
|
|
// You should have received a copy of the GNU Lesser General Public License
|
|
// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
package les
|
|
|
|
import (
|
|
"crypto/ecdsa"
|
|
"time"
|
|
|
|
"github.com/ethereum/go-ethereum/accounts/abi/bind"
|
|
"github.com/ethereum/go-ethereum/common/mclock"
|
|
"github.com/ethereum/go-ethereum/core"
|
|
"github.com/ethereum/go-ethereum/eth"
|
|
"github.com/ethereum/go-ethereum/les/flowcontrol"
|
|
"github.com/ethereum/go-ethereum/light"
|
|
"github.com/ethereum/go-ethereum/log"
|
|
"github.com/ethereum/go-ethereum/p2p"
|
|
"github.com/ethereum/go-ethereum/p2p/discv5"
|
|
"github.com/ethereum/go-ethereum/p2p/enode"
|
|
"github.com/ethereum/go-ethereum/p2p/enr"
|
|
"github.com/ethereum/go-ethereum/params"
|
|
"github.com/ethereum/go-ethereum/rpc"
|
|
)
|
|
|
|
type LesServer struct {
|
|
lesCommons
|
|
|
|
archiveMode bool // Flag whether the ethereum node runs in archive mode.
|
|
handler *serverHandler
|
|
lesTopics []discv5.Topic
|
|
privateKey *ecdsa.PrivateKey
|
|
|
|
// Flow control and capacity management
|
|
fcManager *flowcontrol.ClientManager
|
|
costTracker *costTracker
|
|
defParams flowcontrol.ServerParams
|
|
servingQueue *servingQueue
|
|
clientPool *clientPool
|
|
|
|
minCapacity, maxCapacity, freeCapacity uint64
|
|
threadsIdle int // Request serving threads count when system is idle.
|
|
threadsBusy int // Request serving threads count when system is busy(block insertion).
|
|
}
|
|
|
|
func NewLesServer(e *eth.Ethereum, config *eth.Config) (*LesServer, error) {
|
|
// Collect les protocol version information supported by local node.
|
|
lesTopics := make([]discv5.Topic, len(AdvertiseProtocolVersions))
|
|
for i, pv := range AdvertiseProtocolVersions {
|
|
lesTopics[i] = lesTopic(e.BlockChain().Genesis().Hash(), pv)
|
|
}
|
|
// Calculate the number of threads used to service the light client
|
|
// requests based on the user-specified value.
|
|
threads := config.LightServ * 4 / 100
|
|
if threads < 4 {
|
|
threads = 4
|
|
}
|
|
srv := &LesServer{
|
|
lesCommons: lesCommons{
|
|
genesis: e.BlockChain().Genesis().Hash(),
|
|
config: config,
|
|
chainConfig: e.BlockChain().Config(),
|
|
iConfig: light.DefaultServerIndexerConfig,
|
|
chainDb: e.ChainDb(),
|
|
peers: newPeerSet(),
|
|
chainReader: e.BlockChain(),
|
|
chtIndexer: light.NewChtIndexer(e.ChainDb(), nil, params.CHTFrequency, params.HelperTrieProcessConfirmations),
|
|
bloomTrieIndexer: light.NewBloomTrieIndexer(e.ChainDb(), nil, params.BloomBitsBlocks, params.BloomTrieFrequency),
|
|
closeCh: make(chan struct{}),
|
|
},
|
|
archiveMode: e.ArchiveMode(),
|
|
lesTopics: lesTopics,
|
|
fcManager: flowcontrol.NewClientManager(nil, &mclock.System{}),
|
|
servingQueue: newServingQueue(int64(time.Millisecond*10), float64(config.LightServ)/100),
|
|
threadsBusy: config.LightServ/100 + 1,
|
|
threadsIdle: threads,
|
|
}
|
|
srv.handler = newServerHandler(srv, e.BlockChain(), e.ChainDb(), e.TxPool(), e.Synced)
|
|
srv.costTracker, srv.minCapacity = newCostTracker(e.ChainDb(), config)
|
|
srv.freeCapacity = srv.minCapacity
|
|
|
|
// Set up checkpoint oracle.
|
|
oracle := config.CheckpointOracle
|
|
if oracle == nil {
|
|
oracle = params.CheckpointOracles[e.BlockChain().Genesis().Hash()]
|
|
}
|
|
srv.oracle = newCheckpointOracle(oracle, srv.localCheckpoint)
|
|
|
|
// Initialize server capacity management fields.
|
|
srv.defParams = flowcontrol.ServerParams{
|
|
BufLimit: srv.freeCapacity * bufLimitRatio,
|
|
MinRecharge: srv.freeCapacity,
|
|
}
|
|
// LES flow control tries to more or less guarantee the possibility for the
|
|
// clients to send a certain amount of requests at any time and get a quick
|
|
// response. Most of the clients want this guarantee but don't actually need
|
|
// to send requests most of the time. Our goal is to serve as many clients as
|
|
// possible while the actually used server capacity does not exceed the limits
|
|
totalRecharge := srv.costTracker.totalRecharge()
|
|
srv.maxCapacity = srv.freeCapacity * uint64(srv.config.LightPeers)
|
|
if totalRecharge > srv.maxCapacity {
|
|
srv.maxCapacity = totalRecharge
|
|
}
|
|
srv.fcManager.SetCapacityLimits(srv.freeCapacity, srv.maxCapacity, srv.freeCapacity*2)
|
|
srv.clientPool = newClientPool(srv.chainDb, srv.freeCapacity, mclock.System{}, func(id enode.ID) { go srv.peers.Unregister(peerIdToString(id)) })
|
|
srv.clientPool.setDefaultFactors(priceFactors{0, 1, 1}, priceFactors{0, 1, 1})
|
|
|
|
checkpoint := srv.latestLocalCheckpoint()
|
|
if !checkpoint.Empty() {
|
|
log.Info("Loaded latest checkpoint", "section", checkpoint.SectionIndex, "head", checkpoint.SectionHead,
|
|
"chtroot", checkpoint.CHTRoot, "bloomroot", checkpoint.BloomRoot)
|
|
}
|
|
srv.chtIndexer.Start(e.BlockChain())
|
|
return srv, nil
|
|
}
|
|
|
|
func (s *LesServer) APIs() []rpc.API {
|
|
return []rpc.API{
|
|
{
|
|
Namespace: "les",
|
|
Version: "1.0",
|
|
Service: NewPrivateLightAPI(&s.lesCommons),
|
|
Public: false,
|
|
},
|
|
{
|
|
Namespace: "les",
|
|
Version: "1.0",
|
|
Service: NewPrivateLightServerAPI(s),
|
|
Public: false,
|
|
},
|
|
{
|
|
Namespace: "debug",
|
|
Version: "1.0",
|
|
Service: NewPrivateDebugAPI(s),
|
|
Public: false,
|
|
},
|
|
}
|
|
}
|
|
|
|
func (s *LesServer) Protocols() []p2p.Protocol {
|
|
ps := s.makeProtocols(ServerProtocolVersions, s.handler.runPeer, func(id enode.ID) interface{} {
|
|
if p := s.peers.Peer(peerIdToString(id)); p != nil {
|
|
return p.Info()
|
|
}
|
|
return nil
|
|
})
|
|
// Add "les" ENR entries.
|
|
for i := range ps {
|
|
ps[i].Attributes = []enr.Entry{&lesEntry{}}
|
|
}
|
|
return ps
|
|
}
|
|
|
|
// Start starts the LES server
|
|
func (s *LesServer) Start(srvr *p2p.Server) {
|
|
s.privateKey = srvr.PrivateKey
|
|
s.handler.start()
|
|
|
|
s.wg.Add(1)
|
|
go s.capacityManagement()
|
|
|
|
if srvr.DiscV5 != nil {
|
|
for _, topic := range s.lesTopics {
|
|
topic := topic
|
|
go func() {
|
|
logger := log.New("topic", topic)
|
|
logger.Info("Starting topic registration")
|
|
defer logger.Info("Terminated topic registration")
|
|
|
|
srvr.DiscV5.RegisterTopic(topic, s.closeCh)
|
|
}()
|
|
}
|
|
}
|
|
}
|
|
|
|
// Stop stops the LES service
|
|
func (s *LesServer) Stop() {
|
|
close(s.closeCh)
|
|
|
|
// Disconnect existing sessions.
|
|
// This also closes the gate for any new registrations on the peer set.
|
|
// sessions which are already established but not added to pm.peers yet
|
|
// will exit when they try to register.
|
|
s.peers.Close()
|
|
|
|
s.fcManager.Stop()
|
|
s.costTracker.stop()
|
|
s.handler.stop()
|
|
s.clientPool.stop() // client pool should be closed after handler.
|
|
s.servingQueue.stop()
|
|
|
|
// Note, bloom trie indexer is closed by parent bloombits indexer.
|
|
s.chtIndexer.Close()
|
|
s.wg.Wait()
|
|
log.Info("Les server stopped")
|
|
}
|
|
|
|
func (s *LesServer) SetBloomBitsIndexer(bloomIndexer *core.ChainIndexer) {
|
|
bloomIndexer.AddChildIndexer(s.bloomTrieIndexer)
|
|
}
|
|
|
|
// SetClient sets the rpc client and starts running checkpoint contract if it is not yet watched.
|
|
func (s *LesServer) SetContractBackend(backend bind.ContractBackend) {
|
|
if s.oracle == nil {
|
|
return
|
|
}
|
|
s.oracle.start(backend)
|
|
}
|
|
|
|
// capacityManagement starts an event handler loop that updates the recharge curve of
|
|
// the client manager and adjusts the client pool's size according to the total
|
|
// capacity updates coming from the client manager
|
|
func (s *LesServer) capacityManagement() {
|
|
defer s.wg.Done()
|
|
|
|
processCh := make(chan bool, 100)
|
|
sub := s.handler.blockchain.SubscribeBlockProcessingEvent(processCh)
|
|
defer sub.Unsubscribe()
|
|
|
|
totalRechargeCh := make(chan uint64, 100)
|
|
totalRecharge := s.costTracker.subscribeTotalRecharge(totalRechargeCh)
|
|
|
|
totalCapacityCh := make(chan uint64, 100)
|
|
totalCapacity := s.fcManager.SubscribeTotalCapacity(totalCapacityCh)
|
|
s.clientPool.setLimits(s.config.LightPeers, totalCapacity)
|
|
|
|
var (
|
|
busy bool
|
|
freePeers uint64
|
|
blockProcess mclock.AbsTime
|
|
)
|
|
updateRecharge := func() {
|
|
if busy {
|
|
s.servingQueue.setThreads(s.threadsBusy)
|
|
s.fcManager.SetRechargeCurve(flowcontrol.PieceWiseLinear{{0, 0}, {totalRecharge, totalRecharge}})
|
|
} else {
|
|
s.servingQueue.setThreads(s.threadsIdle)
|
|
s.fcManager.SetRechargeCurve(flowcontrol.PieceWiseLinear{{0, 0}, {totalRecharge / 10, totalRecharge}, {totalRecharge, totalRecharge}})
|
|
}
|
|
}
|
|
updateRecharge()
|
|
|
|
for {
|
|
select {
|
|
case busy = <-processCh:
|
|
if busy {
|
|
blockProcess = mclock.Now()
|
|
} else {
|
|
blockProcessingTimer.Update(time.Duration(mclock.Now() - blockProcess))
|
|
}
|
|
updateRecharge()
|
|
case totalRecharge = <-totalRechargeCh:
|
|
totalRechargeGauge.Update(int64(totalRecharge))
|
|
updateRecharge()
|
|
case totalCapacity = <-totalCapacityCh:
|
|
totalCapacityGauge.Update(int64(totalCapacity))
|
|
newFreePeers := totalCapacity / s.freeCapacity
|
|
if newFreePeers < freePeers && newFreePeers < uint64(s.config.LightPeers) {
|
|
log.Warn("Reduced free peer connections", "from", freePeers, "to", newFreePeers)
|
|
}
|
|
freePeers = newFreePeers
|
|
s.clientPool.setLimits(s.config.LightPeers, totalCapacity)
|
|
case <-s.closeCh:
|
|
return
|
|
}
|
|
}
|
|
}
|