Merge pull request #3584 from JackyWYX/stream_download

[Stream] Add downloader module
pull/3587/head
Rongjian Lan 4 years ago committed by GitHub
commit 7e2302d179
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
  1. 38
      hmy/downloader/adapter.go
  2. 323
      hmy/downloader/adapter_test.go
  3. 152
      hmy/downloader/beaconhelper.go
  4. 68
      hmy/downloader/const.go
  5. 250
      hmy/downloader/downloader.go
  6. 97
      hmy/downloader/downloader_test.go
  7. 74
      hmy/downloader/downloaders.go
  8. 220
      hmy/downloader/inserthelper.go
  9. 19
      hmy/downloader/inserthelper_test.go
  10. 513
      hmy/downloader/longrange.go
  11. 336
      hmy/downloader/longrange_test.go
  12. 448
      hmy/downloader/shortrange.go
  13. 438
      hmy/downloader/shortrange_test.go
  14. 292
      hmy/downloader/types.go
  15. 261
      hmy/downloader/types_test.go

@ -0,0 +1,38 @@
package downloader
import (
"context"
"github.com/ethereum/go-ethereum/common"
"github.com/ethereum/go-ethereum/event"
"github.com/harmony-one/harmony/consensus/engine"
"github.com/harmony-one/harmony/core/types"
"github.com/harmony-one/harmony/p2p/stream/common/streammanager"
syncproto "github.com/harmony-one/harmony/p2p/stream/protocols/sync"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
)
type syncProtocol interface {
GetCurrentBlockNumber(ctx context.Context, opts ...syncproto.Option) (uint64, sttypes.StreamID, error)
GetBlocksByNumber(ctx context.Context, bns []uint64, opts ...syncproto.Option) ([]*types.Block, sttypes.StreamID, error)
GetBlockHashes(ctx context.Context, bns []uint64, opts ...syncproto.Option) ([]common.Hash, sttypes.StreamID, error)
GetBlocksByHashes(ctx context.Context, hs []common.Hash, opts ...syncproto.Option) ([]*types.Block, sttypes.StreamID, error)
RemoveStream(stID sttypes.StreamID) // If a stream delivers invalid data, remove the stream
SubscribeAddStreamEvent(ch chan<- streammanager.EvtStreamAdded) event.Subscription
NumStreams() int
}
type blockChain interface {
engine.ChainReader
Engine() engine.Engine
InsertChain(chain types.Blocks, verifyHeaders bool) (int, error)
WriteCommitSig(blockNum uint64, lastCommits []byte) error
}
// insertHelper is the interface help to verify and insert a block.
type insertHelper interface {
verifyAndInsertBlocks(blocks types.Blocks) (int, error)
verifyAndInsertBlock(block *types.Block) error
}

@ -0,0 +1,323 @@
package downloader
import (
"context"
"fmt"
"math/big"
"sync"
"github.com/harmony-one/harmony/consensus/engine"
staking "github.com/harmony-one/harmony/staking/types"
"github.com/harmony-one/harmony/block"
"github.com/ethereum/go-ethereum/common"
"github.com/ethereum/go-ethereum/event"
"github.com/harmony-one/harmony/core/types"
"github.com/harmony-one/harmony/internal/params"
"github.com/harmony-one/harmony/p2p/stream/common/streammanager"
syncproto "github.com/harmony-one/harmony/p2p/stream/protocols/sync"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
"github.com/harmony-one/harmony/shard"
)
type testBlockChain struct {
curBN uint64
insertErrHook func(bn uint64) error
lock sync.Mutex
}
func newTestBlockChain(curBN uint64, insertErrHook func(bn uint64) error) *testBlockChain {
return &testBlockChain{
curBN: curBN,
insertErrHook: insertErrHook,
}
}
func (bc *testBlockChain) CurrentBlock() *types.Block {
bc.lock.Lock()
defer bc.lock.Unlock()
return makeTestBlock(bc.curBN)
}
func (bc *testBlockChain) CurrentHeader() *block.Header {
bc.lock.Lock()
defer bc.lock.Unlock()
return makeTestBlock(bc.curBN).Header()
}
func (bc *testBlockChain) currentBlockNumber() uint64 {
bc.lock.Lock()
defer bc.lock.Unlock()
return bc.curBN
}
func (bc *testBlockChain) InsertChain(chain types.Blocks, verifyHeaders bool) (int, error) {
bc.lock.Lock()
defer bc.lock.Unlock()
for i, block := range chain {
if bc.insertErrHook != nil {
if err := bc.insertErrHook(block.NumberU64()); err != nil {
return i, err
}
}
if block.NumberU64() <= bc.curBN {
continue
}
if block.NumberU64() != bc.curBN+1 {
return i, fmt.Errorf("not expected block number: %v / %v", block.NumberU64(), bc.curBN+1)
}
bc.curBN++
}
return len(chain), nil
}
func (bc *testBlockChain) changeBlockNumber(val uint64) {
bc.lock.Lock()
defer bc.lock.Unlock()
bc.curBN = val
}
func (bc *testBlockChain) ShardID() uint32 { return 0 }
func (bc *testBlockChain) ReadShardState(epoch *big.Int) (*shard.State, error) { return nil, nil }
func (bc *testBlockChain) Config() *params.ChainConfig { return nil }
func (bc *testBlockChain) WriteCommitSig(blockNum uint64, lastCommits []byte) error { return nil }
func (bc *testBlockChain) GetHeader(hash common.Hash, number uint64) *block.Header { return nil }
func (bc *testBlockChain) GetHeaderByNumber(number uint64) *block.Header { return nil }
func (bc *testBlockChain) GetHeaderByHash(hash common.Hash) *block.Header { return nil }
func (bc *testBlockChain) GetBlock(hash common.Hash, number uint64) *types.Block { return nil }
func (bc *testBlockChain) ReadValidatorList() ([]common.Address, error) { return nil, nil }
func (bc *testBlockChain) ReadCommitSig(blockNum uint64) ([]byte, error) { return nil, nil }
func (bc *testBlockChain) ReadBlockRewardAccumulator(uint64) (*big.Int, error) { return nil, nil }
func (bc *testBlockChain) ValidatorCandidates() []common.Address { return nil }
func (bc *testBlockChain) Engine() engine.Engine { return nil }
func (bc *testBlockChain) ReadValidatorInformation(addr common.Address) (*staking.ValidatorWrapper, error) {
return nil, nil
}
func (bc *testBlockChain) ReadValidatorSnapshot(addr common.Address) (*staking.ValidatorSnapshot, error) {
return nil, nil
}
func (bc *testBlockChain) ReadValidatorSnapshotAtEpoch(epoch *big.Int, addr common.Address) (*staking.ValidatorSnapshot, error) {
return nil, nil
}
func (bc *testBlockChain) ReadValidatorStats(addr common.Address) (*staking.ValidatorStats, error) {
return nil, nil
}
func (bc *testBlockChain) SuperCommitteeForNextEpoch(beacon engine.ChainReader, header *block.Header, isVerify bool) (*shard.State, error) {
return nil, nil
}
type testInsertHelper struct {
bc *testBlockChain
}
func (ch *testInsertHelper) verifyAndInsertBlock(block *types.Block) error {
_, err := ch.bc.InsertChain(types.Blocks{block}, true)
return err
}
func (ch *testInsertHelper) verifyAndInsertBlocks(blocks types.Blocks) (int, error) {
return ch.bc.InsertChain(blocks, true)
}
const (
initStreamNum = 32
minStreamNum = 16
)
type testSyncProtocol struct {
streamIDs []sttypes.StreamID
remoteChain *testBlockChain
requestErrHook func(uint64) error
curIndex int
numStreams int
lock sync.Mutex
}
func newTestSyncProtocol(targetBN uint64, numStreams int, requestErrHook func(uint64) error) *testSyncProtocol {
return &testSyncProtocol{
streamIDs: makeStreamIDs(numStreams),
remoteChain: newTestBlockChain(targetBN, nil),
requestErrHook: requestErrHook,
curIndex: 0,
numStreams: numStreams,
}
}
func (sp *testSyncProtocol) GetCurrentBlockNumber(ctx context.Context, opts ...syncproto.Option) (uint64, sttypes.StreamID, error) {
sp.lock.Lock()
defer sp.lock.Unlock()
bn := sp.remoteChain.currentBlockNumber()
return bn, sp.nextStreamID(), nil
}
func (sp *testSyncProtocol) GetBlocksByNumber(ctx context.Context, bns []uint64, opts ...syncproto.Option) ([]*types.Block, sttypes.StreamID, error) {
sp.lock.Lock()
defer sp.lock.Unlock()
res := make([]*types.Block, 0, len(bns))
for _, bn := range bns {
if sp.requestErrHook != nil {
if err := sp.requestErrHook(bn); err != nil {
return nil, sp.nextStreamID(), err
}
}
if bn > sp.remoteChain.currentBlockNumber() {
res = append(res, nil)
} else {
res = append(res, makeTestBlock(bn))
}
}
return res, sp.nextStreamID(), nil
}
func (sp *testSyncProtocol) GetBlockHashes(ctx context.Context, bns []uint64, opts ...syncproto.Option) ([]common.Hash, sttypes.StreamID, error) {
sp.lock.Lock()
defer sp.lock.Unlock()
res := make([]common.Hash, 0, len(bns))
for _, bn := range bns {
if sp.requestErrHook != nil {
if err := sp.requestErrHook(bn); err != nil {
return nil, sp.nextStreamID(), err
}
}
if bn > sp.remoteChain.currentBlockNumber() {
res = append(res, emptyHash)
} else {
res = append(res, makeTestBlockHash(bn))
}
}
return res, sp.nextStreamID(), nil
}
func (sp *testSyncProtocol) GetBlocksByHashes(ctx context.Context, hs []common.Hash, opts ...syncproto.Option) ([]*types.Block, sttypes.StreamID, error) {
sp.lock.Lock()
defer sp.lock.Unlock()
res := make([]*types.Block, 0, len(hs))
for _, h := range hs {
bn := testHashToNumber(h)
if sp.requestErrHook != nil {
if err := sp.requestErrHook(bn); err != nil {
return nil, sp.nextStreamID(), err
}
}
if bn > sp.remoteChain.currentBlockNumber() {
res = append(res, nil)
} else {
res = append(res, makeTestBlock(bn))
}
}
return res, sp.nextStreamID(), nil
}
func (sp *testSyncProtocol) RemoveStream(target sttypes.StreamID) {
sp.lock.Lock()
defer sp.lock.Unlock()
for i, stid := range sp.streamIDs {
if stid == target {
if i == len(sp.streamIDs)-1 {
sp.streamIDs = sp.streamIDs[:i]
} else {
sp.streamIDs = append(sp.streamIDs[:i], sp.streamIDs[i+1:]...)
}
// mock discovery
if len(sp.streamIDs) < minStreamNum {
sp.streamIDs = append(sp.streamIDs, makeStreamID(sp.numStreams))
sp.numStreams++
}
}
}
}
func (sp *testSyncProtocol) NumStreams() int {
sp.lock.Lock()
defer sp.lock.Unlock()
return len(sp.streamIDs)
}
func (sp *testSyncProtocol) SubscribeAddStreamEvent(ch chan<- streammanager.EvtStreamAdded) event.Subscription {
var evtFeed event.Feed
go func() {
sp.lock.Lock()
num := len(sp.streamIDs)
sp.lock.Unlock()
for i := 0; i != num; i++ {
evtFeed.Send(streammanager.EvtStreamAdded{Stream: nil})
}
}()
return evtFeed.Subscribe(ch)
}
// TODO: add with whitelist stuff
func (sp *testSyncProtocol) nextStreamID() sttypes.StreamID {
if sp.curIndex >= len(sp.streamIDs) {
sp.curIndex = 0
}
index := sp.curIndex
sp.curIndex++
if sp.curIndex >= len(sp.streamIDs) {
sp.curIndex = 0
}
return sp.streamIDs[index]
}
func (sp *testSyncProtocol) changeBlockNumber(val uint64) {
sp.remoteChain.changeBlockNumber(val)
}
func makeStreamIDs(size int) []sttypes.StreamID {
res := make([]sttypes.StreamID, 0, size)
for i := 0; i != size; i++ {
res = append(res, makeStreamID(i))
}
return res
}
func makeStreamID(index int) sttypes.StreamID {
return sttypes.StreamID(fmt.Sprintf("test stream %v", index))
}
var (
hashNumberMap = map[common.Hash]uint64{}
computed uint64
hashNumberLock sync.Mutex
)
func testHashToNumber(h common.Hash) uint64 {
hashNumberLock.Lock()
defer hashNumberLock.Unlock()
if h == emptyHash {
panic("not allowed")
}
if bn, ok := hashNumberMap[h]; ok {
return bn
}
for ; ; computed++ {
ch := makeTestBlockHash(computed)
hashNumberMap[ch] = computed
if ch == h {
return computed
}
}
}
func testNumberToHashes(nums []uint64) []common.Hash {
hashes := make([]common.Hash, 0, len(nums))
for _, num := range nums {
hashes = append(hashes, makeTestBlockHash(num))
}
return hashes
}

@ -0,0 +1,152 @@
package downloader
import (
"time"
"github.com/harmony-one/harmony/core/types"
"github.com/harmony-one/harmony/internal/utils"
"github.com/rs/zerolog"
)
// lastMileCache keeps the last 50 number blocks in memory cache
const lastMileCap = 50
type (
// beaconHelper is the helper for the beacon downloader. The beaconHelper is only started
// when node is running on side chain, listening to beacon client pub-sub message and
// insert the latest blocks to the beacon chain.
beaconHelper struct {
bc blockChain
ih insertHelper
blockC <-chan *types.Block
// TODO: refactor this hook to consensus module. We'd better put it in
// consensus module under a subscription.
insertHook func()
lastMileCache *blocksByNumber
insertC chan insertTask
closeC chan struct{}
logger zerolog.Logger
}
insertTask struct {
doneC chan struct{}
}
)
func newBeaconHelper(bc blockChain, ih insertHelper, blockC <-chan *types.Block, insertHook func()) *beaconHelper {
return &beaconHelper{
bc: bc,
ih: ih,
blockC: blockC,
insertHook: insertHook,
lastMileCache: newBlocksByNumber(lastMileCap),
insertC: make(chan insertTask, 1),
closeC: make(chan struct{}),
logger: utils.Logger().With().
Str("module", "downloader").
Str("sub-module", "beacon helper").
Logger(),
}
}
func (bh *beaconHelper) start() {
go bh.loop()
}
func (bh *beaconHelper) close() {
close(bh.closeC)
}
func (bh *beaconHelper) loop() {
for {
select {
case <-time.Tick(10 * time.Second):
bh.insertAsync()
case b, ok := <-bh.blockC:
if !ok {
return // blockC closed. Node exited
}
if b == nil {
continue
}
bh.lastMileCache.push(b)
bh.insertAsync()
case it := <-bh.insertC:
inserted, bn, err := bh.insertLastMileBlocks()
if err != nil {
bh.logger.Warn().Err(err).Msg("insert last mile blocks error")
continue
}
bh.logger.Info().Int("inserted", inserted).
Uint64("end height", bn).
Uint32("shard", bh.bc.ShardID()).
Msg("insert last mile blocks")
close(it.doneC)
case <-bh.closeC:
return
}
}
}
// insertSync triggers the insert last mile without blocking
func (bh *beaconHelper) insertAsync() {
select {
case bh.insertC <- insertTask{
doneC: make(chan struct{}),
}:
default:
}
}
// insertSync triggers the insert last mile while blocking
func (bh *beaconHelper) insertSync() {
task := insertTask{
doneC: make(chan struct{}),
}
bh.insertC <- task
<-task.doneC
}
func (bh *beaconHelper) insertLastMileBlocks() (inserted int, bn uint64, err error) {
bn = bh.bc.CurrentBlock().NumberU64() + 1
for {
b := bh.getNextBlock(bn)
if b == nil {
bn--
return
}
if err = bh.ih.verifyAndInsertBlock(b); err != nil {
bn--
return
}
bh.logger.Info().Uint64("number", b.NumberU64()).Msg("Inserted block from beacon pub-sub")
if bh.insertHook != nil {
bh.insertHook()
}
inserted++
bn++
}
}
func (bh *beaconHelper) getNextBlock(expBN uint64) *types.Block {
for bh.lastMileCache.len() > 0 {
b := bh.lastMileCache.pop()
if b == nil {
return nil
}
if b.NumberU64() < expBN {
continue
}
if b.NumberU64() > expBN {
bh.lastMileCache.push(b)
return nil
}
return b
}
return nil
}

@ -0,0 +1,68 @@
package downloader
import (
"github.com/harmony-one/harmony/core/types"
nodeconfig "github.com/harmony-one/harmony/internal/configs/node"
)
const (
numBlocksByNumPerRequest int = 10 // number of blocks for each request
blocksPerInsert int = 50 // number of blocks for each insert batch
numBlockHashesPerRequest int = 20 // number of get block hashes for short range sync
numBlocksByHashesUpperCap int = 10 // number of get blocks by hashes upper cap
numBlocksByHashesLowerCap int = 3 // number of get blocks by hashes lower cap
lastMileThres int = 10
// soft cap of size in resultQueue. When the queue size is larger than this limit,
// no more request will be assigned to workers to wait for InsertChain to finish.
softQueueCap int = 100
defaultConcurrency = 16
)
type (
// Config is the downloader config
Config struct {
// parameters
Network nodeconfig.NetworkType
Concurrency int // Number of concurrent sync requests
MinStreams int // Minimum number of streams to do sync
InitStreams int // Number of streams requirement for initial bootstrap
// stream manager config
SmSoftLowCap int
SmHardLowCap int
SmHiCap int
SmDiscBatch int
// config for beacon config
BHConfig *BeaconHelperConfig
}
// BeaconHelperConfig is the extra config used for beaconHelper which uses
// pub-sub block message to do sync.
BeaconHelperConfig struct {
BlockC <-chan *types.Block
InsertHook func()
}
)
func (c *Config) fixValues() {
if c.Concurrency == 0 {
c.Concurrency = defaultConcurrency
}
if c.Concurrency > c.MinStreams {
c.MinStreams = c.Concurrency
}
if c.MinStreams > c.InitStreams {
c.InitStreams = c.MinStreams
}
if c.MinStreams > c.SmSoftLowCap {
c.SmSoftLowCap = c.MinStreams
}
if c.MinStreams > c.SmHardLowCap {
c.SmHardLowCap = c.MinStreams
}
}

@ -0,0 +1,250 @@
package downloader
import (
"context"
"time"
"github.com/ethereum/go-ethereum/event"
"github.com/harmony-one/harmony/core"
nodeconfig "github.com/harmony-one/harmony/internal/configs/node"
"github.com/harmony-one/harmony/internal/utils"
"github.com/harmony-one/harmony/p2p"
"github.com/harmony-one/harmony/p2p/stream/common/streammanager"
"github.com/harmony-one/harmony/p2p/stream/protocols/sync"
"github.com/rs/zerolog"
)
type (
// Downloader is responsible for sync task of one shard
Downloader struct {
bc blockChain
ih insertHelper
syncProtocol syncProtocol
bh *beaconHelper
downloadC chan struct{}
closeC chan struct{}
ctx context.Context
cancel func()
evtDownloadFinished event.Feed // channel for each download task finished
evtDownloadStarted event.Feed // channel for each download has started
status status
config Config
logger zerolog.Logger
}
)
// NewDownloader creates a new downloader
func NewDownloader(host p2p.Host, bc *core.BlockChain, config Config) *Downloader {
config.fixValues()
ih := newInsertHelper(bc)
sp := sync.NewProtocol(sync.Config{
Chain: bc,
Host: host.GetP2PHost(),
Discovery: host.GetDiscovery(),
ShardID: nodeconfig.ShardID(bc.ShardID()),
Network: config.Network,
SmSoftLowCap: config.SmSoftLowCap,
SmHardLowCap: config.SmHardLowCap,
SmHiCap: config.SmHiCap,
DiscBatch: config.SmDiscBatch,
})
host.AddStreamProtocol(sp)
var bh *beaconHelper
if config.BHConfig != nil && bc.ShardID() == 0 {
bh = newBeaconHelper(bc, ih, config.BHConfig.BlockC, config.BHConfig.InsertHook)
}
ctx, cancel := context.WithCancel(context.Background())
return &Downloader{
bc: bc,
ih: ih,
syncProtocol: sp,
bh: bh,
downloadC: make(chan struct{}),
closeC: make(chan struct{}),
ctx: ctx,
cancel: cancel,
status: newStatus(),
config: config,
logger: utils.Logger().With().Str("module", "downloader").Logger(),
}
}
// Start start the downloader
func (d *Downloader) Start() {
go d.run()
if d.bh != nil {
d.bh.start()
}
}
// Close close the downloader
func (d *Downloader) Close() {
close(d.closeC)
d.cancel()
if d.bh != nil {
d.bh.close()
}
}
// DownloadAsync triggers the download async. If there is already a download task that is
// in progress, return ErrDownloadInProgress.
func (d *Downloader) DownloadAsync() {
select {
case d.downloadC <- struct{}{}:
case <-time.After(100 * time.Millisecond):
}
}
// NumPeers returns the number of peers connected of a specific shard.
func (d *Downloader) NumPeers() int {
return d.syncProtocol.NumStreams()
}
// IsSyncing return the current sync status
func (d *Downloader) SyncStatus() (bool, uint64) {
syncing, target := d.status.get()
if !syncing {
target = d.bc.CurrentBlock().NumberU64()
}
return syncing, target
}
// SubscribeDownloadStarted subscribe download started
func (d *Downloader) SubscribeDownloadStarted(ch chan struct{}) event.Subscription {
return d.evtDownloadStarted.Subscribe(ch)
}
// SubscribeDownloadFinishedEvent subscribe the download finished
func (d *Downloader) SubscribeDownloadFinished(ch chan struct{}) event.Subscription {
return d.evtDownloadFinished.Subscribe(ch)
}
func (d *Downloader) run() {
d.waitForBootFinish()
d.loop()
}
// waitForBootFinish wait for stream manager to finish the initial discovery and have
// enough peers to start downloader
func (d *Downloader) waitForBootFinish() {
evtCh := make(chan streammanager.EvtStreamAdded, 1)
sub := d.syncProtocol.SubscribeAddStreamEvent(evtCh)
defer sub.Unsubscribe()
checkCh := make(chan struct{}, 1)
trigger := func() {
select {
case checkCh <- struct{}{}:
default:
}
}
trigger()
t := time.NewTicker(10 * time.Second)
for {
d.logger.Info().Msg("waiting for initial bootstrap discovery")
select {
case <-t.C:
trigger()
case <-evtCh:
trigger()
case <-checkCh:
if d.syncProtocol.NumStreams() >= d.config.InitStreams {
return
}
case <-d.closeC:
return
}
}
}
func (d *Downloader) loop() {
ticker := time.NewTicker(10 * time.Second)
initSync := true
trigger := func() {
select {
case d.downloadC <- struct{}{}:
case <-time.After(100 * time.Millisecond):
}
}
go trigger()
for {
select {
case <-ticker.C:
go trigger()
case <-d.downloadC:
addedBN, err := d.doDownload(initSync)
if err != nil {
// If error happens, sleep 5 seconds and retry
d.logger.Warn().Err(err).Bool("bootstrap", initSync).Msg("failed to download")
go func() {
time.Sleep(5 * time.Second)
trigger()
}()
continue
}
d.logger.Info().Int("block added", addedBN).
Uint64("current height", d.bc.CurrentBlock().NumberU64()).
Bool("initSync", initSync).
Uint32("shard", d.bc.ShardID()).
Msg("sync finished")
if addedBN != 0 {
// If block number has been changed, trigger another sync
// and try to add last mile from pub-sub (blocking)
go trigger()
if d.bh != nil {
d.bh.insertSync()
}
}
initSync = false
case <-d.closeC:
return
}
}
}
func (d *Downloader) doDownload(initSync bool) (n int, err error) {
if initSync {
d.logger.Info().Uint64("current number", d.bc.CurrentBlock().NumberU64()).
Uint32("shard ID", d.bc.ShardID()).Msg("start long range sync")
n, err = d.doLongRangeSync()
} else {
d.logger.Info().Uint64("current number", d.bc.CurrentBlock().NumberU64()).
Uint32("shard ID", d.bc.ShardID()).Msg("start short range sync")
n, err = d.doShortRangeSync()
}
if err != nil {
return
}
return
}
func (d *Downloader) startSyncing() {
d.status.startSyncing()
d.evtDownloadStarted.Send(struct{}{})
}
func (d *Downloader) finishSyncing() {
d.status.finishSyncing()
d.evtDownloadFinished.Send(struct{}{})
}

@ -0,0 +1,97 @@
package downloader
import (
"context"
"fmt"
"testing"
"time"
)
func TestDownloader_Integration(t *testing.T) {
sp := newTestSyncProtocol(1000, 48, nil)
bc := newTestBlockChain(0, nil)
ctx, cancel := context.WithCancel(context.Background())
c := Config{}
c.fixValues() // use default config values
d := &Downloader{
bc: bc,
ih: &testInsertHelper{bc},
syncProtocol: sp,
downloadC: make(chan struct{}),
closeC: make(chan struct{}),
ctx: ctx,
cancel: cancel,
config: c,
}
// subscribe download event
finishedCh := make(chan struct{}, 1)
finishedSub := d.SubscribeDownloadFinished(finishedCh)
startedCh := make(chan struct{}, 1)
startedSub := d.SubscribeDownloadStarted(startedCh)
defer finishedSub.Unsubscribe()
defer startedSub.Unsubscribe()
// Start the downloader
d.Start()
defer d.Close()
// During bootstrap, trigger two download task: one long range, one short range.
// The second one will not trigger start / finish events.
if err := checkReceiveChanMulTimes(startedCh, 1, 10*time.Second); err != nil {
t.Fatal(err)
}
if err := checkReceiveChanMulTimes(finishedCh, 1, 10*time.Second); err != nil {
t.Fatal(err)
}
if curBN := d.bc.CurrentBlock().NumberU64(); curBN != 1000 {
t.Fatal("blockchain not synced to the latest")
}
// Increase the remote block number, and trigger one download task manually
sp.changeBlockNumber(1010)
d.DownloadAsync()
// We shall do short range test twice
if err := checkReceiveChanMulTimes(startedCh, 1, 10*time.Second); err != nil {
t.Fatal(err)
}
if err := checkReceiveChanMulTimes(finishedCh, 1, 10*time.Second); err != nil {
t.Fatal(err)
}
if curBN := d.bc.CurrentBlock().NumberU64(); curBN != 1010 {
t.Fatal("blockchain not synced to the latest")
}
// Remote block number unchanged, and trigger one download task manually
d.DownloadAsync()
if err := checkReceiveChanMulTimes(startedCh, 0, 10*time.Second); err != nil {
t.Fatal(err)
}
if err := checkReceiveChanMulTimes(finishedCh, 0, 10*time.Second); err != nil {
t.Fatal(err)
}
// At last, check number of streams, should be exactly the same as the initial number
if sp.numStreams != 48 {
t.Errorf("unexpected number of streams at the end: %v / %v", sp.numStreams, 48)
}
}
func checkReceiveChanMulTimes(ch chan struct{}, times int, timeout time.Duration) error {
t := time.Tick(timeout)
for i := 0; i != times; i++ {
select {
case <-ch:
case <-t:
return fmt.Errorf("timed out %v", timeout)
}
}
select {
case <-ch:
return fmt.Errorf("received an extra event")
case <-time.After(100 * time.Millisecond):
}
return nil
}

@ -0,0 +1,74 @@
package downloader
import (
"github.com/harmony-one/harmony/core"
"github.com/harmony-one/harmony/p2p"
)
// Downloaders is the set of downloaders
type Downloaders struct {
ds map[uint32]*Downloader
}
// NewDownloaders creates Downloaders for sync of multiple blockchains
func NewDownloaders(host p2p.Host, bcs []*core.BlockChain, config Config) *Downloaders {
ds := make(map[uint32]*Downloader)
for _, bc := range bcs {
if bc == nil {
continue
}
if _, ok := ds[bc.ShardID()]; ok {
continue
}
ds[bc.ShardID()] = NewDownloader(host, bc, config)
}
return &Downloaders{ds}
}
// Start start the downloaders
func (ds *Downloaders) Start() {
for _, d := range ds.ds {
d.Start()
}
}
// Close close the downloaders
func (ds *Downloaders) Close() {
for _, d := range ds.ds {
d.Close()
}
}
// DownloadAsync triggers a download
func (ds *Downloaders) DownloadAsync(shardID uint32) {
d, ok := ds.ds[shardID]
if !ok && d != nil {
d.DownloadAsync()
}
}
// GetShardDownloader get the downloader with the given shard ID
func (ds *Downloaders) GetShardDownloader(shardID uint32) *Downloader {
return ds.ds[shardID]
}
// NumPeers returns the connected peers for each shard
func (ds *Downloaders) NumPeers() map[uint32]int {
res := make(map[uint32]int)
for sid, d := range ds.ds {
res[sid] = d.NumPeers()
}
return res
}
// SyncStatus returns whether the given shard is doing syncing task and the target block
// number.
func (ds *Downloaders) SyncStatus(shardID uint32) (bool, uint64) {
d, ok := ds.ds[shardID]
if !ok {
return false, 0
}
return d.SyncStatus()
}

@ -0,0 +1,220 @@
package downloader
import (
"fmt"
"hash"
"math/big"
"sync"
"github.com/ethereum/go-ethereum/common"
lru "github.com/hashicorp/golang-lru"
"github.com/pkg/errors"
"golang.org/x/crypto/sha3"
bls_core "github.com/harmony-one/bls/ffi/go/bls"
"github.com/harmony-one/harmony/consensus/quorum"
"github.com/harmony-one/harmony/consensus/signature"
"github.com/harmony-one/harmony/core/types"
bls_cosi "github.com/harmony-one/harmony/crypto/bls"
"github.com/harmony-one/harmony/internal/chain"
"github.com/harmony-one/harmony/multibls"
"github.com/harmony-one/harmony/shard"
)
// sigVerifyError is the error type of failing verify the signature of the current block.
// Since this is a sanity field and is not included the block hash, it needs extra verification.
// The error types is used to differentiate the error of signature verification VS insert error.
type sigVerifyError struct {
err error
}
func (err *sigVerifyError) Error() string {
return fmt.Sprintf("failed verify signature: %v", err.err.Error())
}
// insertHelperImpl helps to verify and insert blocks, along with some caching mechanism.
type insertHelperImpl struct {
bc blockChain
deciderCache *lru.Cache // Epoch -> quorum.Decider
shardStateCache *lru.Cache // Epoch -> *shard.State
verifiedSigCache *lru.Cache // verifiedSigKey -> struct{}{}
}
func newInsertHelper(bc blockChain) insertHelper {
deciderCache, _ := lru.New(5)
shardStateCache, _ := lru.New(5)
sigCache, _ := lru.New(20)
return &insertHelperImpl{
bc: bc,
deciderCache: deciderCache,
shardStateCache: shardStateCache,
verifiedSigCache: sigCache,
}
}
func (ch *insertHelperImpl) verifyAndInsertBlocks(blocks types.Blocks) (int, error) {
for i, block := range blocks {
if err := ch.verifyAndInsertBlock(block); err != nil {
return i, err
}
}
return len(blocks), nil
}
func (ch *insertHelperImpl) verifyAndInsertBlock(block *types.Block) error {
// verify the commit sig of current block
if err := ch.verifyBlockSignature(block); err != nil {
return &sigVerifyError{err}
}
ch.markBlockSigVerified(block, block.GetCurrentCommitSig())
// verify header. Skip verify the previous seal if we have already verified
verifySeal := !ch.isBlockLastSigVerified(block)
if err := ch.bc.Engine().VerifyHeader(ch.bc, block.Header(), verifySeal); err != nil {
return err
}
// Insert chain.
if _, err := ch.bc.InsertChain(types.Blocks{block}, false); err != nil {
return err
}
// Write commit sig data
return ch.bc.WriteCommitSig(block.NumberU64(), block.GetCurrentCommitSig())
}
func (ch *insertHelperImpl) verifyBlockSignature(block *types.Block) error {
// TODO: This is the duplicate logic to the implementation of verifySeal and consensus.
// Better refactor to the blockchain or engine structure
decider, err := ch.readDeciderByEpoch(block.Epoch())
if err != nil {
return err
}
sig, mask, err := decodeCommitSig(block.GetCurrentCommitSig(), decider.Participants())
if err != nil {
return err
}
if !decider.IsQuorumAchievedByMask(mask) {
return errors.New("quorum not achieved")
}
commitSigBytes := signature.ConstructCommitPayload(ch.bc, block.Epoch(), block.Hash(),
block.NumberU64(), block.Header().ViewID().Uint64())
if !sig.VerifyHash(mask.AggregatePublic, commitSigBytes) {
return errors.New("aggregate signature failed verification")
}
return nil
}
func (ch *insertHelperImpl) writeBlockSignature(block *types.Block) error {
return ch.bc.WriteCommitSig(block.NumberU64(), block.GetCurrentCommitSig())
}
func (ch *insertHelperImpl) getDeciderByEpoch(epoch *big.Int) (quorum.Decider, error) {
epochUint := epoch.Uint64()
if decider, ok := ch.deciderCache.Get(epochUint); ok && decider != nil {
return decider.(quorum.Decider), nil
}
decider, err := ch.getDeciderByEpoch(epoch)
if err != nil {
return nil, errors.Wrapf(err, "unable to read quorum of epoch %v", epoch.Uint64())
}
ch.deciderCache.Add(epochUint, decider)
return decider, nil
}
func (ch *insertHelperImpl) readDeciderByEpoch(epoch *big.Int) (quorum.Decider, error) {
isStaking := ch.bc.Config().IsStaking(epoch)
decider := ch.getNewDecider(isStaking)
ss, err := ch.getShardState(epoch)
if err != nil {
return nil, err
}
subComm, err := ss.FindCommitteeByID(ch.shardID())
if err != nil {
return nil, err
}
pubKeys, err := subComm.BLSPublicKeys()
if err != nil {
return nil, err
}
decider.UpdateParticipants(pubKeys)
if _, err := decider.SetVoters(subComm, epoch); err != nil {
return nil, err
}
return decider, nil
}
func (ch *insertHelperImpl) getNewDecider(isStaking bool) quorum.Decider {
if isStaking {
return quorum.NewDecider(quorum.SuperMajorityVote, ch.bc.ShardID())
} else {
return quorum.NewDecider(quorum.SuperMajorityStake, ch.bc.ShardID())
}
}
func (ch *insertHelperImpl) getShardState(epoch *big.Int) (*shard.State, error) {
if ss, ok := ch.shardStateCache.Get(epoch.Uint64()); ok && ss != nil {
return ss.(*shard.State), nil
}
ss, err := ch.bc.ReadShardState(epoch)
if err != nil {
return nil, err
}
ch.shardStateCache.Add(epoch.Uint64(), ss)
return ss, nil
}
func (ch *insertHelperImpl) markBlockSigVerified(block *types.Block, sigAndBitmap []byte) {
key := newVerifiedSigKey(block.Hash(), sigAndBitmap)
ch.verifiedSigCache.Add(key, struct{}{})
}
func (ch *insertHelperImpl) isBlockLastSigVerified(block *types.Block) bool {
lastSig := block.Header().LastCommitSignature()
lastBM := block.Header().LastCommitBitmap()
lastSigAndBM := append(lastSig[:], lastBM...)
key := newVerifiedSigKey(block.Hash(), lastSigAndBM)
_, ok := ch.verifiedSigCache.Get(key)
return ok
}
func (ch *insertHelperImpl) shardID() uint32 {
return ch.bc.ShardID()
}
func decodeCommitSig(commitBytes []byte, publicKeys multibls.PublicKeys) (*bls_core.Sign, *bls_cosi.Mask, error) {
if len(commitBytes) < bls_cosi.BLSSignatureSizeInBytes {
return nil, nil, fmt.Errorf("unexpected signature bytes size: %v / %v", len(commitBytes),
bls_cosi.BLSSignatureSizeInBytes)
}
return chain.ReadSignatureBitmapByPublicKeys(commitBytes, publicKeys)
}
type verifiedSigKey struct {
blockHash common.Hash
sbHash common.Hash // hash of block signature + bitmap
}
var hasherPool = sync.Pool{
New: func() interface{} {
return sha3.New256()
},
}
func newVerifiedSigKey(blockHash common.Hash, sigAndBitmap []byte) verifiedSigKey {
hasher := hasherPool.Get().(hash.Hash)
defer func() {
hasher.Reset()
hasherPool.Put(hasher)
}()
var sbHash common.Hash
hasher.Write(sigAndBitmap)
hasher.Sum(sbHash[0:])
return verifiedSigKey{
blockHash: blockHash,
sbHash: sbHash,
}
}

@ -0,0 +1,19 @@
package downloader
import (
"testing"
"github.com/ethereum/go-ethereum/common"
)
func BenchmarkNewVerifiedSigKey(b *testing.B) {
var bh common.Hash
commitSig := make([]byte, 100)
for i := 0; i != len(commitSig); i++ {
commitSig[i] = 0xf
}
for i := 0; i != b.N; i++ {
newVerifiedSigKey(bh, commitSig)
}
}

@ -0,0 +1,513 @@
package downloader
import (
"context"
"fmt"
"sync"
"time"
"github.com/harmony-one/harmony/core/types"
syncproto "github.com/harmony-one/harmony/p2p/stream/protocols/sync"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
"github.com/pkg/errors"
"github.com/rs/zerolog"
)
// doLongRangeSync does the long range sync.
// One LongRangeSync consists of several iterations.
// For each iteration, estimate the current block number, then fetch block & insert to blockchain
func (d *Downloader) doLongRangeSync() (int, error) {
var totalInserted int
d.startSyncing()
defer d.finishSyncing()
for {
ctx, cancel := context.WithCancel(d.ctx)
iter := &lrSyncIter{
bc: d.bc,
p: d.syncProtocol,
ih: d.ih,
d: d,
ctx: ctx,
config: d.config,
logger: d.logger.With().Str("mode", "long range").Logger(),
}
if err := iter.doLongRangeSync(); err != nil {
cancel()
return totalInserted + iter.inserted, err
}
cancel()
totalInserted += iter.inserted
if iter.inserted < lastMileThres {
return totalInserted, nil
}
}
}
// lrSyncIter run a single iteration of a full long range sync.
// First get a rough estimate of the current block height, and then sync to this
// block number
type lrSyncIter struct {
bc blockChain
p syncProtocol
ih insertHelper
d *Downloader
gbm *getBlocksManager // initialized when finished get block number
inserted int
config Config
ctx context.Context
logger zerolog.Logger
}
func (lsi *lrSyncIter) doLongRangeSync() error {
if err := lsi.checkPrerequisites(); err != nil {
return err
}
bn, err := lsi.estimateCurrentNumber()
if err != nil {
return err
}
lsi.logger.Info().Uint64("target number", bn).Msg("estimated remote current number")
lsi.d.status.setTargetBN(bn)
return lsi.fetchAndInsertBlocks(bn)
}
func (lsi *lrSyncIter) checkPrerequisites() error {
return lsi.checkHaveEnoughStreams()
}
// estimateCurrentNumber roughly estimate the current block number.
// The block number does not need to be exact, but just a temporary target of the iteration
func (lsi *lrSyncIter) estimateCurrentNumber() (uint64, error) {
var (
cnResults = make(map[sttypes.StreamID]uint64)
lock sync.Mutex
wg sync.WaitGroup
)
wg.Add(lsi.config.Concurrency)
for i := 0; i != lsi.config.Concurrency; i++ {
go func() {
defer wg.Done()
bn, stid, err := lsi.doGetCurrentNumberRequest()
if err != nil {
lsi.logger.Err(err).Str("streamID", string(stid)).
Msg("getCurrentNumber request failed. Removing stream")
if !errors.Is(err, context.Canceled) {
lsi.p.RemoveStream(stid)
}
return
}
lock.Lock()
cnResults[stid] = bn
lock.Unlock()
}()
}
wg.Wait()
if len(cnResults) == 0 {
select {
case <-lsi.ctx.Done():
return 0, lsi.ctx.Err()
default:
}
return 0, errors.New("zero block number response from remote nodes")
}
bn := computeBNMaxVote(cnResults)
return bn, nil
}
func (lsi *lrSyncIter) doGetCurrentNumberRequest() (uint64, sttypes.StreamID, error) {
ctx, cancel := context.WithTimeout(lsi.ctx, 10*time.Second)
defer cancel()
bn, stid, err := lsi.p.GetCurrentBlockNumber(ctx, syncproto.WithHighPriority())
if err != nil {
return 0, stid, err
}
return bn, stid, nil
}
// fetchAndInsertBlocks use the pipeline pattern to boost the performance of inserting blocks.
// TODO: For resharding, use the pipeline to do fast sync (epoch loop, header loop, body loop)
func (lsi *lrSyncIter) fetchAndInsertBlocks(targetBN uint64) error {
gbm := newGetBlocksManager(lsi.bc, targetBN, lsi.logger)
lsi.gbm = gbm
// Setup workers to fetch blocks from remote node
for i := 0; i != lsi.config.Concurrency; i++ {
worker := &getBlocksWorker{
gbm: gbm,
protocol: lsi.p,
ctx: lsi.ctx,
}
go worker.workLoop()
}
// insert the blocks to chain. Return when the target block number is reached.
lsi.insertChainLoop(targetBN)
select {
case <-lsi.ctx.Done():
return lsi.ctx.Err()
default:
}
return nil
}
func (lsi *lrSyncIter) insertChainLoop(targetBN uint64) {
var (
gbm = lsi.gbm
t = time.NewTicker(100 * time.Millisecond)
resultC = make(chan struct{}, 1)
)
trigger := func() {
select {
case resultC <- struct{}{}:
default:
}
}
for {
select {
case <-lsi.ctx.Done():
return
case <-t.C:
// Redundancy, periodically check whether there is blocks that can be processed
trigger()
case <-gbm.resultC:
// New block arrive in resultQueue
trigger()
case <-resultC:
blockResults := gbm.PullContinuousBlocks(blocksPerInsert)
if len(blockResults) > 0 {
lsi.processBlocks(blockResults, targetBN)
// more blocks is expected being able to be pulled from queue
trigger()
}
if lsi.bc.CurrentBlock().NumberU64() >= targetBN {
return
}
}
}
}
func (lsi *lrSyncIter) processBlocks(results []*blockResult, targetBN uint64) {
blocks := blockResultsToBlocks(results)
for i, block := range blocks {
if err := lsi.ih.verifyAndInsertBlock(block); err != nil {
lsi.logger.Warn().Err(err).Uint64("target block", targetBN).
Uint64("block number", block.NumberU64()).
Msg("insert blocks failed in long range")
lsi.p.RemoveStream(results[i].stid)
lsi.gbm.HandleInsertError(results, i)
return
}
lsi.inserted++
}
lsi.gbm.HandleInsertResult(results)
}
func (lsi *lrSyncIter) checkHaveEnoughStreams() error {
numStreams := lsi.p.NumStreams()
if numStreams < lsi.config.MinStreams {
return fmt.Errorf("number of streams smaller than minimum: %v < %v",
numStreams, lsi.config.MinStreams)
}
return nil
}
// getBlocksWorker does the request job
type getBlocksWorker struct {
gbm *getBlocksManager
protocol syncProtocol
ctx context.Context
}
func (w *getBlocksWorker) workLoop() {
for {
select {
case <-w.ctx.Done():
return
default:
}
batch := w.gbm.GetNextBatch()
if len(batch) == 0 {
select {
case <-w.ctx.Done():
return
case <-time.After(100 * time.Millisecond):
continue
}
}
blocks, stid, err := w.doBatch(batch)
if err != nil {
if !errors.Is(err, context.Canceled) {
w.protocol.RemoveStream(stid)
}
err = errors.Wrap(err, "request error")
w.gbm.HandleRequestError(batch, err, stid)
} else {
w.gbm.HandleRequestResult(batch, blocks, stid)
}
}
}
func (w *getBlocksWorker) doBatch(bns []uint64) ([]*types.Block, sttypes.StreamID, error) {
ctx, cancel := context.WithTimeout(w.ctx, 10*time.Second)
defer cancel()
blocks, stid, err := w.protocol.GetBlocksByNumber(ctx, bns)
if err != nil {
return nil, stid, err
}
if err := validateGetBlocksResult(bns, blocks); err != nil {
return nil, stid, err
}
return blocks, stid, nil
}
// getBlocksManager is the helper structure for get blocks request management
type getBlocksManager struct {
chain blockChain
targetBN uint64
requesting map[uint64]struct{} // block numbers that have been assigned to workers but not received
processing map[uint64]struct{} // block numbers received requests but not inserted
retries *prioritizedNumbers // requests where error happens
rq *resultQueue // result queue wait to be inserted into blockchain
resultC chan struct{}
logger zerolog.Logger
lock sync.Mutex
}
func newGetBlocksManager(chain blockChain, targetBN uint64, logger zerolog.Logger) *getBlocksManager {
return &getBlocksManager{
chain: chain,
targetBN: targetBN,
requesting: make(map[uint64]struct{}),
processing: make(map[uint64]struct{}),
retries: newPrioritizedNumbers(),
rq: newResultQueue(),
resultC: make(chan struct{}, 1),
logger: logger,
}
}
// GetNextBatch get the next block numbers batch
func (gbm *getBlocksManager) GetNextBatch() []uint64 {
gbm.lock.Lock()
defer gbm.lock.Unlock()
cap := numBlocksByNumPerRequest
bns := gbm.getBatchFromRetries(cap)
cap -= len(bns)
gbm.addBatchToRequesting(bns)
if gbm.availableForMoreTasks() {
addBNs := gbm.getBatchFromUnprocessed(cap)
gbm.addBatchToRequesting(addBNs)
bns = append(bns, addBNs...)
}
return bns
}
// HandleRequestError handles the error result
func (gbm *getBlocksManager) HandleRequestError(bns []uint64, err error, stid sttypes.StreamID) {
gbm.lock.Lock()
defer gbm.lock.Unlock()
gbm.logger.Warn().Err(err).Str("stream", string(stid)).Msg("get blocks error")
// add requested block numbers to retries
for _, bn := range bns {
delete(gbm.requesting, bn)
gbm.retries.push(bn)
}
// remove results from result queue by the stream and add back to retries
removed := gbm.rq.removeResultsByStreamID(stid)
for _, bn := range removed {
delete(gbm.processing, bn)
gbm.retries.push(bn)
}
}
// HandleRequestResult handles get blocks result
func (gbm *getBlocksManager) HandleRequestResult(bns []uint64, blocks []*types.Block, stid sttypes.StreamID) {
gbm.lock.Lock()
defer gbm.lock.Unlock()
for i, bn := range bns {
delete(gbm.requesting, bn)
if blocks[i] == nil {
gbm.retries.push(bn)
} else {
gbm.processing[bn] = struct{}{}
}
}
gbm.rq.addBlockResults(blocks, stid)
select {
case gbm.resultC <- struct{}{}:
default:
}
}
// HandleInsertResult handle the insert result
func (gbm *getBlocksManager) HandleInsertResult(inserted []*blockResult) {
gbm.lock.Lock()
defer gbm.lock.Unlock()
for _, block := range inserted {
delete(gbm.processing, block.getBlockNumber())
}
}
// HandleInsertError handles the error during InsertChain
func (gbm *getBlocksManager) HandleInsertError(results []*blockResult, n int) {
gbm.lock.Lock()
defer gbm.lock.Unlock()
var (
inserted []*blockResult
errResult *blockResult
abandoned []*blockResult
)
inserted = results[:n]
errResult = results[n]
if n != len(results) {
abandoned = results[n+1:]
}
for _, res := range inserted {
delete(gbm.processing, res.getBlockNumber())
}
for _, res := range abandoned {
gbm.rq.addBlockResults([]*types.Block{res.block}, res.stid)
}
delete(gbm.processing, errResult.getBlockNumber())
gbm.retries.push(errResult.getBlockNumber())
removed := gbm.rq.removeResultsByStreamID(errResult.stid)
for _, bn := range removed {
delete(gbm.processing, bn)
gbm.retries.push(bn)
}
}
// PullContinuousBlocks pull continuous blocks from request queue
func (gbm *getBlocksManager) PullContinuousBlocks(cap int) []*blockResult {
gbm.lock.Lock()
defer gbm.lock.Unlock()
expHeight := gbm.chain.CurrentBlock().NumberU64() + 1
results, stales := gbm.rq.popBlockResults(expHeight, cap)
// For stale blocks, we remove them from processing
for _, bn := range stales {
delete(gbm.processing, bn)
}
return results
}
func (gbm *getBlocksManager) getBatchFromRetries(cap int) []uint64 {
var (
requestBNs []uint64
curHeight = gbm.chain.CurrentBlock().NumberU64()
)
for cnt := 0; cnt < cap; cnt++ {
bn := gbm.retries.pop()
if bn == 0 {
break // no more retries
}
if bn <= curHeight {
continue
}
requestBNs = append(requestBNs, bn)
}
return requestBNs
}
func (gbm *getBlocksManager) getBatchFromUnprocessed(cap int) []uint64 {
var (
requestBNs []uint64
curHeight = gbm.chain.CurrentBlock().NumberU64()
)
bn := curHeight + 1
// TODO: this algorithm can be potentially optimized.
for cnt := 0; cnt < cap && bn <= gbm.targetBN; cnt++ {
for bn <= gbm.targetBN {
_, ok1 := gbm.requesting[bn]
_, ok2 := gbm.processing[bn]
if !ok1 && !ok2 {
requestBNs = append(requestBNs, bn)
bn++
break
}
bn++
}
}
return requestBNs
}
func (gbm *getBlocksManager) availableForMoreTasks() bool {
return gbm.rq.results.Len() < softQueueCap
}
func (gbm *getBlocksManager) addBatchToRequesting(bns []uint64) {
for _, bn := range bns {
gbm.requesting[bn] = struct{}{}
}
}
func validateGetBlocksResult(requested []uint64, result []*types.Block) error {
if len(result) != len(requested) {
return fmt.Errorf("unexpected number of blocks delivered: %v / %v", len(result), len(requested))
}
for i, block := range result {
if block != nil && block.NumberU64() != requested[i] {
return fmt.Errorf("block with unexpected number delivered: %v / %v", block.NumberU64(), requested[i])
}
}
return nil
}
func computeBNMaxVote(votes map[sttypes.StreamID]uint64) uint64 {
var (
nm = make(map[uint64]int)
res uint64
maxCnt int
)
for _, bn := range votes {
_, ok := nm[bn]
if !ok {
nm[bn] = 0
}
nm[bn]++
cnt := nm[bn]
if cnt > maxCnt || (cnt == maxCnt && bn > res) {
res = bn
maxCnt = cnt
}
}
return res
}

@ -0,0 +1,336 @@
package downloader
import (
"context"
"fmt"
"math/rand"
"sync"
"testing"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
"github.com/pkg/errors"
)
func TestDownloader_doLongRangeSync(t *testing.T) {
targetBN := uint64(1000)
bc := newTestBlockChain(1, nil)
d := &Downloader{
bc: bc,
ih: &testInsertHelper{bc},
syncProtocol: newTestSyncProtocol(targetBN, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
ctx: context.Background(),
}
synced, err := d.doLongRangeSync()
if err != nil {
t.Error(err)
}
if synced == 0 {
t.Errorf("synced false")
}
if curNum := d.bc.CurrentBlock().NumberU64(); curNum != targetBN {
t.Errorf("block number not expected: %v / %v", curNum, targetBN)
}
}
func TestLrSyncIter_EstimateCurrentNumber(t *testing.T) {
lsi := &lrSyncIter{
p: newTestSyncProtocol(100, 32, nil),
ctx: context.Background(),
config: Config{
Concurrency: 16,
MinStreams: 10,
},
}
bn, err := lsi.estimateCurrentNumber()
if err != nil {
t.Error(err)
}
if bn != 100 {
t.Errorf("unexpected block number: %v / %v", bn, 100)
}
}
func TestGetBlocksManager_GetNextBatch(t *testing.T) {
tests := []struct {
gbm *getBlocksManager
expBNs []uint64
}{
{
gbm: makeGetBlocksManager(
10, 100, []uint64{9, 11, 12, 13},
[]uint64{14, 15, 16}, []uint64{}, 0,
),
expBNs: []uint64{17, 18, 19, 20, 21, 22, 23, 24, 25, 26},
},
{
gbm: makeGetBlocksManager(
10, 100, []uint64{9, 13, 14, 15, 16},
[]uint64{}, []uint64{10, 11, 12}, 0,
),
expBNs: []uint64{11, 12, 17, 18, 19, 20, 21, 22, 23, 24},
},
{
gbm: makeGetBlocksManager(
10, 100, []uint64{9, 13, 14, 15, 16},
[]uint64{}, []uint64{10, 11, 12}, 120,
),
expBNs: []uint64{11, 12},
},
{
gbm: makeGetBlocksManager(
10, 100, []uint64{9, 13, 14, 15, 16},
[]uint64{}, []uint64{}, 120,
),
expBNs: []uint64{},
},
{
gbm: makeGetBlocksManager(
10, 20, []uint64{9, 13, 14, 15, 16},
[]uint64{}, []uint64{}, 0,
),
expBNs: []uint64{11, 12, 17, 18, 19, 20},
},
{
gbm: makeGetBlocksManager(
10, 100, []uint64{9, 13, 14, 15, 16},
[]uint64{}, []uint64{}, 0,
),
expBNs: []uint64{11, 12, 17, 18, 19, 20, 21, 22, 23, 24},
},
}
for i, test := range tests {
if i < 4 {
continue
}
batch := test.gbm.GetNextBatch()
if len(test.expBNs) != len(batch) {
t.Errorf("Test %v: unexpected size [%v] / [%v]", i, batch, test.expBNs)
}
for i := range test.expBNs {
if test.expBNs[i] != batch[i] {
t.Errorf("Test %v: [%v] / [%v]", i, batch, test.expBNs)
}
}
}
}
func TestLrSyncIter_FetchAndInsertBlocks(t *testing.T) {
targetBN := uint64(1000)
chain := newTestBlockChain(0, nil)
protocol := newTestSyncProtocol(targetBN, 32, nil)
ctx, _ := context.WithCancel(context.Background())
lsi := &lrSyncIter{
bc: chain,
ih: &testInsertHelper{chain},
d: &Downloader{},
p: protocol,
gbm: nil,
config: Config{
Concurrency: 100,
},
ctx: ctx,
}
lsi.fetchAndInsertBlocks(targetBN)
if err := fetchAndInsertBlocksResultCheck(lsi, targetBN, initStreamNum); err != nil {
t.Error(err)
}
}
// When FetchAndInsertBlocks, one request has an error
func TestLrSyncIter_FetchAndInsertBlocks_ErrRequest(t *testing.T) {
targetBN := uint64(1000)
var once sync.Once
errHook := func(bn uint64) error {
var err error
once.Do(func() {
err = errors.New("test error expected")
})
return err
}
chain := newTestBlockChain(0, nil)
protocol := newTestSyncProtocol(targetBN, 32, errHook)
ctx, _ := context.WithCancel(context.Background())
lsi := &lrSyncIter{
bc: chain,
ih: &testInsertHelper{chain},
d: &Downloader{},
p: protocol,
gbm: nil,
config: Config{
Concurrency: 100,
},
ctx: ctx,
}
lsi.fetchAndInsertBlocks(targetBN)
if err := fetchAndInsertBlocksResultCheck(lsi, targetBN, initStreamNum-1); err != nil {
t.Error(err)
}
}
// When FetchAndInsertBlocks, one insertion has an error
func TestLrSyncIter_FetchAndInsertBlocks_ErrInsert(t *testing.T) {
targetBN := uint64(1000)
var once sync.Once
errHook := func(bn uint64) error {
var err error
once.Do(func() {
err = errors.New("test error expected")
})
return err
}
chain := newTestBlockChain(0, errHook)
protocol := newTestSyncProtocol(targetBN, 32, nil)
ctx, _ := context.WithCancel(context.Background())
lsi := &lrSyncIter{
bc: chain,
ih: &testInsertHelper{chain},
d: &Downloader{},
p: protocol,
gbm: nil,
config: Config{
Concurrency: 100,
},
ctx: ctx,
}
lsi.fetchAndInsertBlocks(targetBN)
if err := fetchAndInsertBlocksResultCheck(lsi, targetBN, initStreamNum-1); err != nil {
t.Error(err)
}
}
// When FetchAndInsertBlocks, randomly error happens
func TestLrSyncIter_FetchAndInsertBlocks_RandomErr(t *testing.T) {
targetBN := uint64(10000)
rand.Seed(0)
errHook := func(bn uint64) error {
// 10% error happens
if rand.Intn(10)%10 == 0 {
return errors.New("error expected")
}
return nil
}
chain := newTestBlockChain(0, errHook)
protocol := newTestSyncProtocol(targetBN, 32, errHook)
ctx, _ := context.WithCancel(context.Background())
lsi := &lrSyncIter{
bc: chain,
ih: &testInsertHelper{chain},
d: &Downloader{},
p: protocol,
gbm: nil,
config: Config{
Concurrency: 100,
},
ctx: ctx,
}
lsi.fetchAndInsertBlocks(targetBN)
if err := fetchAndInsertBlocksResultCheck(lsi, targetBN, minStreamNum); err != nil {
t.Error(err)
}
}
func fetchAndInsertBlocksResultCheck(lsi *lrSyncIter, targetBN uint64, expNumStreams int) error {
if bn := lsi.bc.CurrentBlock().NumberU64(); bn != targetBN {
return fmt.Errorf("did not reached targetBN: %v / %v", bn, targetBN)
}
lsi.gbm.lock.Lock()
defer lsi.gbm.lock.Unlock()
if len(lsi.gbm.processing) != 0 {
return fmt.Errorf("not empty processing: %v", lsi.gbm.processing)
}
if len(lsi.gbm.requesting) != 0 {
return fmt.Errorf("not empty requesting: %v", lsi.gbm.requesting)
}
if lsi.gbm.retries.length() != 0 {
return fmt.Errorf("not empty retries: %v", lsi.gbm.retries)
}
if lsi.gbm.rq.length() != 0 {
return fmt.Errorf("not empty result queue: %v", lsi.gbm.rq.results)
}
tsp := lsi.p.(*testSyncProtocol)
if len(tsp.streamIDs) != expNumStreams {
return fmt.Errorf("num streams not expected: %v / %v", len(tsp.streamIDs), expNumStreams)
}
return nil
}
func TestComputeBNMaxVote(t *testing.T) {
tests := []struct {
votes map[sttypes.StreamID]uint64
exp uint64
}{
{
votes: map[sttypes.StreamID]uint64{
makeStreamID(0): 10,
makeStreamID(1): 10,
makeStreamID(2): 20,
},
exp: 10,
},
{
votes: map[sttypes.StreamID]uint64{
makeStreamID(0): 10,
makeStreamID(1): 20,
},
exp: 20,
},
{
votes: map[sttypes.StreamID]uint64{
makeStreamID(0): 20,
makeStreamID(1): 10,
makeStreamID(2): 20,
},
exp: 20,
},
}
for i, test := range tests {
res := computeBNMaxVote(test.votes)
if res != test.exp {
t.Errorf("Test %v: unexpected bn %v / %v", i, res, test.exp)
}
}
}
func makeGetBlocksManager(curBN, targetBN uint64, requesting, processing, retries []uint64, sizeRQ int) *getBlocksManager {
chain := newTestBlockChain(curBN, nil)
requestingM := make(map[uint64]struct{})
for _, bn := range requesting {
requestingM[bn] = struct{}{}
}
processingM := make(map[uint64]struct{})
for _, bn := range processing {
processingM[bn] = struct{}{}
}
retriesPN := newPrioritizedNumbers()
for _, retry := range retries {
retriesPN.push(retry)
}
rq := newResultQueue()
for i := uint64(0); i != uint64(sizeRQ); i++ {
rq.addBlockResults(makeTestBlocks([]uint64{i + curBN}), "")
}
return &getBlocksManager{
chain: chain,
targetBN: targetBN,
requesting: requestingM,
processing: processingM,
retries: retriesPN,
rq: rq,
resultC: make(chan struct{}, 1),
}
}

@ -0,0 +1,448 @@
package downloader
import (
"context"
"fmt"
"math"
"sync"
"time"
"github.com/ethereum/go-ethereum/common"
"github.com/harmony-one/harmony/core/types"
syncProto "github.com/harmony-one/harmony/p2p/stream/protocols/sync"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
"github.com/pkg/errors"
"github.com/rs/zerolog"
)
// doShortRangeSync does the short range sync.
// Compared with long range sync, short range sync is more focused on syncing to the latest block.
// It consist of 3 steps:
// 1. Obtain the block hashes and ompute the longest hash chain..
// 2. Get blocks by hashes from computed hash chain.
// 3. Insert the blocks to blockchain.
func (d *Downloader) doShortRangeSync() (int, error) {
sh := &srHelper{
syncProtocol: d.syncProtocol,
ctx: d.ctx,
config: d.config,
logger: d.logger.With().Str("mode", "short range").Logger(),
}
if err := sh.checkPrerequisites(); err != nil {
return 0, errors.Wrap(err, "prerequisite")
}
curBN := d.bc.CurrentBlock().NumberU64()
hashChain, whitelist, err := sh.getHashChain(curBN)
if err != nil {
return 0, errors.Wrap(err, "getHashChain")
}
if len(hashChain) == 0 {
// short circuit for no sync is needed
return 0, nil
}
d.startSyncing()
expEndBN := curBN + uint64(len(hashChain)) - 1
d.status.setTargetBN(expEndBN)
defer d.finishSyncing()
blocks, err := sh.getBlocksByHashes(hashChain, whitelist)
if err != nil {
if !errors.Is(err, context.Canceled) {
sh.removeStreams(whitelist) // Remote nodes cannot provide blocks with target hashes
}
return 0, errors.Wrap(err, "getBlocksByHashes")
}
n, err := d.ih.verifyAndInsertBlocks(blocks)
if err != nil {
if !errors.As(err, &sigVerifyError{}) {
sh.removeStreams(whitelist) // Data provided by remote nodes is corrupted
}
return n, errors.Wrap(err, "InsertChain")
}
return len(blocks), nil
}
type srHelper struct {
syncProtocol syncProtocol
ctx context.Context
config Config
logger zerolog.Logger
}
func (sh *srHelper) getHashChain(curBN uint64) ([]common.Hash, []sttypes.StreamID, error) {
bns := sh.prepareBlockHashNumbers(curBN)
results := newBlockHashResults(bns)
var wg sync.WaitGroup
wg.Add(sh.config.Concurrency)
for i := 0; i != sh.config.Concurrency; i++ {
go func() {
defer wg.Done()
hashes, stid, err := sh.doGetBlockHashesRequest(bns)
if err != nil {
return
}
results.addResult(hashes, stid)
}()
}
wg.Wait()
select {
case <-sh.ctx.Done():
return nil, nil, sh.ctx.Err()
default:
}
hashChain, wl := results.computeLongestHashChain()
return hashChain, wl, nil
}
func (sh *srHelper) getBlocksByHashes(hashes []common.Hash, whitelist []sttypes.StreamID) ([]*types.Block, error) {
ctx, cancel := context.WithCancel(sh.ctx)
m := newGetBlocksByHashManager(hashes, whitelist)
var (
wg sync.WaitGroup
gErr error
errLock sync.Mutex
)
wg.Add(sh.config.Concurrency)
for i := 0; i != sh.config.Concurrency; i++ {
go func() {
defer wg.Done()
defer cancel() // it's ok to cancel context more than once
for {
if m.isDone() {
return
}
hashes, wl, err := m.getNextHashes()
if err != nil {
errLock.Lock()
gErr = err
errLock.Unlock()
return
}
if len(hashes) == 0 {
select {
case <-time.After(200 * time.Millisecond):
continue
case <-ctx.Done():
return
}
}
blocks, stid, err := sh.doGetBlocksByHashesRequest(ctx, hashes, wl)
if err != nil {
sh.logger.Err(err).Msg("getBlocksByHashes worker failed")
m.handleResultError(hashes, stid)
} else {
m.addResult(hashes, blocks, stid)
}
}
}()
}
wg.Wait()
if gErr != nil {
return nil, gErr
}
select {
case <-sh.ctx.Done():
return nil, sh.ctx.Err()
default:
}
return m.getResults()
}
func (sh *srHelper) checkPrerequisites() error {
if sh.syncProtocol.NumStreams() < sh.config.Concurrency {
return errors.New("not enough streams")
}
return nil
}
func (sh *srHelper) prepareBlockHashNumbers(curNumber uint64) []uint64 {
res := make([]uint64, 0, numBlockHashesPerRequest)
for bn := curNumber + 1; bn <= curNumber+uint64(numBlockHashesPerRequest); bn++ {
res = append(res, bn)
}
return res
}
func (sh *srHelper) doGetBlockHashesRequest(bns []uint64) ([]common.Hash, sttypes.StreamID, error) {
ctx, cancel := context.WithTimeout(sh.ctx, 1*time.Second)
defer cancel()
hashes, stid, err := sh.syncProtocol.GetBlockHashes(ctx, bns)
if err != nil {
return nil, stid, err
}
if len(hashes) != len(bns) {
err := errors.New("unexpected get block hashes result delivered")
sh.logger.Warn().Err(err).Str("stream", string(stid)).Msg("failed to doGetBlockHashesRequest")
sh.syncProtocol.RemoveStream(stid)
return nil, stid, err
}
return hashes, stid, nil
}
func (sh *srHelper) doGetBlocksByHashesRequest(ctx context.Context, hashes []common.Hash, wl []sttypes.StreamID) ([]*types.Block, sttypes.StreamID, error) {
ctx, cancel := context.WithTimeout(sh.ctx, 10*time.Second)
defer cancel()
blocks, stid, err := sh.syncProtocol.GetBlocksByHashes(ctx, hashes,
syncProto.WithWhitelist(wl))
if err != nil {
return nil, stid, err
}
if err := checkGetBlockByHashesResult(blocks, hashes); err != nil {
sh.logger.Warn().Err(err).Str("stream", string(stid)).Msg("failed to getBlockByHashes")
sh.syncProtocol.RemoveStream(stid)
return nil, stid, err
}
return blocks, stid, nil
}
func (sh *srHelper) removeStreams(sts []sttypes.StreamID) {
for _, st := range sts {
sh.syncProtocol.RemoveStream(st)
}
}
func checkGetBlockByHashesResult(blocks []*types.Block, hashes []common.Hash) error {
if len(blocks) != len(hashes) {
return errors.New("unexpected number of getBlocksByHashes result")
}
for i, block := range blocks {
if block == nil {
return errors.New("nil block found")
}
if block.Hash() != hashes[i] {
return fmt.Errorf("unexpected block hash: %x / %x", block.Hash(), hashes[i])
}
}
return nil
}
type (
blockHashResults struct {
bns []uint64
results []map[sttypes.StreamID]common.Hash
lock sync.Mutex
}
)
func newBlockHashResults(bns []uint64) *blockHashResults {
results := make([]map[sttypes.StreamID]common.Hash, 0, len(bns))
for range bns {
results = append(results, make(map[sttypes.StreamID]common.Hash))
}
return &blockHashResults{
bns: bns,
results: results,
}
}
func (res *blockHashResults) addResult(hashes []common.Hash, stid sttypes.StreamID) {
res.lock.Lock()
defer res.lock.Unlock()
for i, h := range hashes {
if h == emptyHash {
return // nil block hash reached
}
res.results[i][stid] = h
}
return
}
func (res *blockHashResults) computeLongestHashChain() ([]common.Hash, []sttypes.StreamID) {
var (
whitelist map[sttypes.StreamID]struct{}
hashChain []common.Hash
)
for _, result := range res.results {
hash, nextWl := countHashMaxVote(result, whitelist)
if hash == emptyHash {
break
}
hashChain = append(hashChain, hash)
whitelist = nextWl
}
sts := make([]sttypes.StreamID, 0, len(whitelist))
for st := range whitelist {
sts = append(sts, st)
}
return hashChain, sts
}
func countHashMaxVote(m map[sttypes.StreamID]common.Hash, whitelist map[sttypes.StreamID]struct{}) (common.Hash, map[sttypes.StreamID]struct{}) {
var (
voteM = make(map[common.Hash]int)
res common.Hash
maxCnt = 0
)
for st, h := range m {
if len(whitelist) != 0 {
if _, ok := whitelist[st]; !ok {
continue
}
}
if _, ok := voteM[h]; !ok {
voteM[h] = 0
}
voteM[h]++
if voteM[h] > maxCnt {
maxCnt = voteM[h]
res = h
}
}
nextWl := make(map[sttypes.StreamID]struct{})
for st, h := range m {
if h != res {
continue
}
if len(whitelist) != 0 {
if _, ok := whitelist[st]; ok {
nextWl[st] = struct{}{}
}
} else {
nextWl[st] = struct{}{}
}
}
return res, nextWl
}
type getBlocksByHashManager struct {
hashes []common.Hash
pendings map[common.Hash]struct{}
results map[common.Hash]blockResult
whitelist []sttypes.StreamID
lock sync.Mutex
}
func newGetBlocksByHashManager(hashes []common.Hash, whitelist []sttypes.StreamID) *getBlocksByHashManager {
return &getBlocksByHashManager{
hashes: hashes,
pendings: make(map[common.Hash]struct{}),
results: make(map[common.Hash]blockResult),
whitelist: whitelist,
}
}
func (m *getBlocksByHashManager) getNextHashes() ([]common.Hash, []sttypes.StreamID, error) {
m.lock.Lock()
defer m.lock.Unlock()
num := m.numBlocksPerRequest()
hashes := make([]common.Hash, 0, num)
if len(m.whitelist) == 0 {
return nil, nil, errors.New("empty white list")
}
for _, hash := range m.hashes {
if len(hashes) == num {
break
}
_, ok1 := m.pendings[hash]
_, ok2 := m.results[hash]
if !ok1 && !ok2 {
hashes = append(hashes, hash)
}
}
sts := make([]sttypes.StreamID, len(m.whitelist))
copy(sts, m.whitelist)
return hashes, sts, nil
}
func (m *getBlocksByHashManager) numBlocksPerRequest() int {
val := divideCeil(len(m.hashes), len(m.whitelist))
if val < numBlocksByHashesLowerCap {
val = numBlocksByHashesLowerCap
}
if val > numBlocksByHashesUpperCap {
val = numBlocksByHashesUpperCap
}
return val
}
func (m *getBlocksByHashManager) addResult(hashes []common.Hash, blocks []*types.Block, stid sttypes.StreamID) {
m.lock.Lock()
defer m.lock.Unlock()
for i, hash := range hashes {
block := blocks[i]
delete(m.pendings, hash)
m.results[hash] = blockResult{
block: block,
stid: stid,
}
}
}
func (m *getBlocksByHashManager) handleResultError(hashes []common.Hash, stid sttypes.StreamID) {
m.lock.Lock()
defer m.lock.Unlock()
m.removeStreamID(stid)
for _, hash := range hashes {
delete(m.pendings, hash)
}
}
func (m *getBlocksByHashManager) getResults() ([]*types.Block, error) {
m.lock.Lock()
defer m.lock.Unlock()
blocks := make([]*types.Block, 0, len(m.hashes))
for _, hash := range m.hashes {
if m.results[hash].block == nil {
return nil, errors.New("SANITY: nil block found")
}
blocks = append(blocks, m.results[hash].block)
}
return blocks, nil
}
func (m *getBlocksByHashManager) isDone() bool {
m.lock.Lock()
defer m.lock.Unlock()
return len(m.results) == len(m.hashes)
}
func (m *getBlocksByHashManager) removeStreamID(target sttypes.StreamID) {
// O(n^2) complexity. But considering the whitelist size is small, should not
// have performance issue.
loop:
for i, stid := range m.whitelist {
if stid == target {
if i == len(m.whitelist) {
m.whitelist = m.whitelist[:i]
} else {
m.whitelist = append(m.whitelist[:i], m.whitelist[i+1:]...)
}
goto loop
}
}
return
}
func divideCeil(x, y int) int {
fVal := float64(x) / float64(y)
return int(math.Ceil(fVal))
}

@ -0,0 +1,438 @@
package downloader
import (
"context"
"errors"
"fmt"
"sync"
"testing"
"github.com/ethereum/go-ethereum/common"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
"github.com/rs/zerolog"
)
func TestDownloader_doShortRangeSync(t *testing.T) {
chain := newTestBlockChain(100, nil)
d := &Downloader{
bc: chain,
ih: &testInsertHelper{chain},
syncProtocol: newTestSyncProtocol(105, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
ctx: context.Background(),
logger: zerolog.Logger{},
}
n, err := d.doShortRangeSync()
if err != nil {
t.Error(err)
}
if n == 0 {
t.Error("not synced")
}
if curNum := d.bc.CurrentBlock().NumberU64(); curNum != 105 {
t.Errorf("unexpected block number after sync: %v / %v", curNum, 105)
}
}
func TestSrHelper_getHashChain(t *testing.T) {
tests := []struct {
curBN uint64
syncProtocol syncProtocol
config Config
expHashChainSize int
expStSize int
}{
{
curBN: 100,
syncProtocol: newTestSyncProtocol(1000, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: numBlockHashesPerRequest,
expStSize: 16, // Concurrency
},
{
curBN: 100,
syncProtocol: newTestSyncProtocol(100, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: 0,
expStSize: 0,
},
{
curBN: 100,
syncProtocol: newTestSyncProtocol(110, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: 10,
expStSize: 16,
},
{
// stream size is smaller than concurrency
curBN: 100,
syncProtocol: newTestSyncProtocol(1000, 10, nil),
config: Config{
Concurrency: 16,
MinStreams: 8,
},
expHashChainSize: numBlockHashesPerRequest,
expStSize: 10,
},
{
// one stream reports an error, else are fine
curBN: 100,
syncProtocol: newTestSyncProtocol(1000, 32, makeOnceErrorFunc()),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: numBlockHashesPerRequest,
expStSize: 15, // Concurrency
},
{
// error happens at one block number, all stream removed
curBN: 100,
syncProtocol: newTestSyncProtocol(1000, 32, func(bn uint64) error {
if bn == 110 {
return errors.New("test error")
}
return nil
}),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: 0,
expStSize: 0,
},
{
curBN: 100,
syncProtocol: newTestSyncProtocol(1000, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expHashChainSize: numBlockHashesPerRequest,
expStSize: 16, // Concurrency
},
}
for i, test := range tests {
sh := &srHelper{
syncProtocol: test.syncProtocol,
ctx: context.Background(),
config: test.config,
}
hashChain, wl, err := sh.getHashChain(test.curBN)
if err != nil {
t.Error(err)
}
if len(hashChain) != test.expHashChainSize {
t.Errorf("Test %v: hash chain size unexpected: %v / %v", i, len(hashChain), test.expHashChainSize)
}
if len(wl) != test.expStSize {
t.Errorf("Test %v: whitelist size unexpected: %v / %v", i, len(wl), test.expStSize)
}
}
}
func TestSrHelper_GetBlocksByHashes(t *testing.T) {
tests := []struct {
hashes []common.Hash
syncProtocol syncProtocol
config Config
expBlockNumbers []uint64
expErr error
}{
{
hashes: testNumberToHashes([]uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110}),
syncProtocol: newTestSyncProtocol(1000, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expBlockNumbers: []uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110},
expErr: nil,
},
{
// remote node cannot give the block with the given hash
hashes: testNumberToHashes([]uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110}),
syncProtocol: newTestSyncProtocol(100, 32, nil),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expBlockNumbers: []uint64{},
expErr: errors.New("all streams are bad"),
},
{
// one request return an error, else are fine
hashes: testNumberToHashes([]uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110}),
syncProtocol: newTestSyncProtocol(1000, 32, makeOnceErrorFunc()),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expBlockNumbers: []uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110},
expErr: nil,
},
{
// All nodes encounter an error
hashes: testNumberToHashes([]uint64{101, 102, 103, 104, 105, 106, 107, 108, 109, 110}),
syncProtocol: newTestSyncProtocol(1000, 32, func(n uint64) error {
if n == 109 {
return errors.New("test error")
}
return nil
}),
config: Config{
Concurrency: 16,
MinStreams: 16,
},
expErr: errors.New("error expected"),
},
}
for i, test := range tests {
sh := &srHelper{
syncProtocol: test.syncProtocol,
ctx: context.Background(),
config: test.config,
}
blocks, err := sh.getBlocksByHashes(test.hashes, makeStreamIDs(5))
if (err == nil) != (test.expErr == nil) {
t.Errorf("Test %v: unexpected error %v / %v", i, err, test.expErr)
}
if len(blocks) != len(test.expBlockNumbers) {
t.Errorf("Test %v: unepxected block number size: %v / %v", i, len(blocks), len(test.expBlockNumbers))
}
for i, block := range blocks {
gotNum := testHashToNumber(block.Hash())
if gotNum != test.expBlockNumbers[i] {
t.Errorf("Test %v: unexpected block number", i)
}
}
}
}
func TestBlockHashResult_ComputeLongestHashChain(t *testing.T) {
tests := []struct {
bns []uint64
results map[sttypes.StreamID][]int64
expChain []uint64
expWhitelist map[sttypes.StreamID]struct{}
expErr error
}{
{
bns: []uint64{1, 2, 3, 4, 5, 6, 7, 8, 9, 10},
results: map[sttypes.StreamID][]int64{
makeStreamID(0): {1, 2, 3, 4, 5, 6, 7},
makeStreamID(1): {1, 2, 3, 4, 5, 6, 7},
makeStreamID(2): {1, 2, 3, 4, 5}, // left behind
},
expChain: []uint64{1, 2, 3, 4, 5, 6, 7},
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(0): {},
makeStreamID(1): {},
},
},
{
// minority fork
bns: []uint64{1, 2, 3, 4, 5, 6, 7, 8, 9, 10},
results: map[sttypes.StreamID][]int64{
makeStreamID(0): {1, 2, 3, 4, 5, 6, 7},
makeStreamID(1): {1, 2, 3, 4, 5, 6, 7},
makeStreamID(2): {1, 2, 3, 4, 5, 7, 8, 9},
},
expChain: []uint64{1, 2, 3, 4, 5, 6, 7},
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(0): {},
makeStreamID(1): {},
},
}, {
// nil block
bns: []uint64{1, 2, 3, 4, 5, 6, 7, 8, 9, 10},
results: map[sttypes.StreamID][]int64{
makeStreamID(0): {},
makeStreamID(1): {},
makeStreamID(2): {},
},
expChain: nil,
expWhitelist: nil,
}, {
// not continuous block
bns: []uint64{1, 2, 3, 4, 5, 6, 7, 8, 9, 10},
results: map[sttypes.StreamID][]int64{
makeStreamID(0): {1, 2, 3, 4, 5, 6, 7, -1, 9},
makeStreamID(1): {1, 2, 3, 4, 5, 6, 7},
makeStreamID(2): {1, 2, 3, 4, 5, 7, 8, 9},
},
expChain: []uint64{1, 2, 3, 4, 5, 6, 7},
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(0): {},
makeStreamID(1): {},
},
},
{
// not continuous block
bns: []uint64{1, 2, 3, 4, 5, 6, 7, 8, 9, 10},
results: map[sttypes.StreamID][]int64{},
expErr: errors.New("zero result"),
},
}
for i, test := range tests {
res := newBlockHashResults(test.bns)
for st, hs := range test.results {
res.addResult(makeTestBlockHashes(hs), st)
}
chain, wl := res.computeLongestHashChain()
if err := checkHashChainResult(chain, test.expChain); err != nil {
t.Errorf("Test %v: %v", i, err)
}
if err := checkStreamSetEqual(streamIDListToMap(wl), test.expWhitelist); err != nil {
t.Errorf("Test %v: %v", i, err)
}
}
}
func checkHashChainResult(gots []common.Hash, exps []uint64) error {
if len(gots) != len(exps) {
return errors.New("unexpected size")
}
for i, got := range gots {
exp := exps[i]
if got != makeTestBlockHash(exp) {
return errors.New("unexpected block hash")
}
}
return nil
}
func TestHashMaxVote(t *testing.T) {
tests := []struct {
m map[sttypes.StreamID]common.Hash
whitelist map[sttypes.StreamID]struct{}
expRes common.Hash
expWhitelist map[sttypes.StreamID]struct{}
}{
{
m: map[sttypes.StreamID]common.Hash{
makeStreamID(0): makeTestBlockHash(0),
makeStreamID(1): makeTestBlockHash(1),
makeStreamID(2): makeTestBlockHash(1),
},
whitelist: map[sttypes.StreamID]struct{}{
makeStreamID(0): {},
makeStreamID(1): {},
makeStreamID(2): {},
},
expRes: makeTestBlockHash(1),
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(1): {},
makeStreamID(2): {},
},
}, {
m: map[sttypes.StreamID]common.Hash{
makeStreamID(0): makeTestBlockHash(0),
makeStreamID(1): makeTestBlockHash(1),
makeStreamID(2): makeTestBlockHash(1),
},
whitelist: nil,
expRes: makeTestBlockHash(1),
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(1): {},
makeStreamID(2): {},
},
}, {
m: map[sttypes.StreamID]common.Hash{
makeStreamID(0): makeTestBlockHash(0),
makeStreamID(1): makeTestBlockHash(1),
makeStreamID(2): makeTestBlockHash(1),
makeStreamID(3): makeTestBlockHash(0),
makeStreamID(4): makeTestBlockHash(0),
},
whitelist: map[sttypes.StreamID]struct{}{
makeStreamID(0): {},
makeStreamID(1): {},
makeStreamID(2): {},
},
expRes: makeTestBlockHash(1),
expWhitelist: map[sttypes.StreamID]struct{}{
makeStreamID(1): {},
makeStreamID(2): {},
},
},
}
for i, test := range tests {
h, wl := countHashMaxVote(test.m, test.whitelist)
if h != test.expRes {
t.Errorf("Test %v: unexpected hash: %x / %x", i, h, test.expRes)
}
if err := checkStreamSetEqual(wl, test.expWhitelist); err != nil {
t.Errorf("Test %v: %v", i, err)
}
}
}
func checkStreamSetEqual(m1, m2 map[sttypes.StreamID]struct{}) error {
if len(m1) != len(m2) {
return fmt.Errorf("unexpected size: %v / %v", len(m1), len(m2))
}
for st := range m1 {
if _, ok := m2[st]; !ok {
return errors.New("not equal")
}
}
return nil
}
func makeTestBlockHashes(bns []int64) []common.Hash {
hs := make([]common.Hash, 0, len(bns))
for _, bn := range bns {
if bn < 0 {
hs = append(hs, emptyHash)
} else {
hs = append(hs, makeTestBlockHash(uint64(bn)))
}
}
return hs
}
func streamIDListToMap(sts []sttypes.StreamID) map[sttypes.StreamID]struct{} {
res := make(map[sttypes.StreamID]struct{})
for _, st := range sts {
res[st] = struct{}{}
}
return res
}
func makeTestBlockHash(bn uint64) common.Hash {
return makeTestBlock(bn).Hash()
}
func makeOnceErrorFunc() func(num uint64) error {
var once sync.Once
return func(num uint64) error {
var err error
once.Do(func() {
err = errors.New("test error expected")
})
return err
}
}

@ -0,0 +1,292 @@
package downloader
import (
"container/heap"
"sync"
"github.com/ethereum/go-ethereum/common"
"github.com/harmony-one/harmony/core/types"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
)
var (
emptyHash common.Hash
)
type status struct {
isSyncing bool
targetBN uint64
lock sync.Mutex
}
func newStatus() status {
return status{}
}
func (s *status) startSyncing() {
s.lock.Lock()
defer s.lock.Unlock()
s.isSyncing = true
}
func (s *status) setTargetBN(val uint64) {
s.lock.Lock()
defer s.lock.Unlock()
s.targetBN = val
}
func (s *status) finishSyncing() {
s.lock.Lock()
defer s.lock.Unlock()
s.isSyncing = false
s.targetBN = 0
}
func (s *status) get() (bool, uint64) {
s.lock.Lock()
defer s.lock.Unlock()
return s.isSyncing, s.targetBN
}
type getBlocksResult struct {
bns []uint64
blocks []*types.Block
stid sttypes.StreamID
}
type resultQueue struct {
results *priorityQueue
lock sync.Mutex
}
func newResultQueue() *resultQueue {
pq := make(priorityQueue, 0, 200) // 200 - rough estimate
heap.Init(&pq)
return &resultQueue{
results: &pq,
}
}
// addBlockResults adds the blocks to the result queue to be processed by insertChainLoop.
// If a nil block is detected in the block list, will not process further blocks.
func (rq *resultQueue) addBlockResults(blocks []*types.Block, stid sttypes.StreamID) {
rq.lock.Lock()
defer rq.lock.Unlock()
for _, block := range blocks {
if block == nil {
continue
}
heap.Push(rq.results, &blockResult{
block: block,
stid: stid,
})
}
return
}
// popBlockResults pop a continuous list of blocks starting at expStartBN with capped size.
// Return the stale block numbers as the second return value
func (rq *resultQueue) popBlockResults(expStartBN uint64, cap int) ([]*blockResult, []uint64) {
rq.lock.Lock()
defer rq.lock.Unlock()
var (
res = make([]*blockResult, 0, cap)
stales []uint64
)
for cnt := 0; rq.results.Len() > 0 && cnt < cap; cnt++ {
br := heap.Pop(rq.results).(*blockResult)
// stale block number
if br.block.NumberU64() < expStartBN {
stales = append(stales, br.block.NumberU64())
continue
}
if br.block.NumberU64() != expStartBN {
heap.Push(rq.results, br)
return res, stales
}
res = append(res, br)
expStartBN++
}
return res, stales
}
// removeResultsByStreamID remove the block results of the given stream, return the block
// number removed from the queue
func (rq *resultQueue) removeResultsByStreamID(stid sttypes.StreamID) []uint64 {
rq.lock.Lock()
defer rq.lock.Unlock()
var removed []uint64
Loop:
for {
for i, res := range *rq.results {
blockRes := res.(*blockResult)
if blockRes.stid == stid {
rq.removeByIndex(i)
removed = append(removed, blockRes.block.NumberU64())
goto Loop
}
}
break
}
return removed
}
func (rq *resultQueue) length() int {
return len(*rq.results)
}
func (rq *resultQueue) removeByIndex(index int) {
heap.Remove(rq.results, index)
}
// bnPrioritizedItem is the item which uses block number to determine its priority
type bnPrioritizedItem interface {
getBlockNumber() uint64
}
type blockResult struct {
block *types.Block
stid sttypes.StreamID
}
func (br *blockResult) getBlockNumber() uint64 {
return br.block.NumberU64()
}
func blockResultsToBlocks(results []*blockResult) []*types.Block {
blocks := make([]*types.Block, 0, len(results))
for _, result := range results {
blocks = append(blocks, result.block)
}
return blocks
}
type (
prioritizedNumber uint64
prioritizedNumbers struct {
q *priorityQueue
}
)
func (b prioritizedNumber) getBlockNumber() uint64 {
return uint64(b)
}
func newPrioritizedNumbers() *prioritizedNumbers {
pqs := make(priorityQueue, 0)
heap.Init(&pqs)
return &prioritizedNumbers{
q: &pqs,
}
}
func (pbs *prioritizedNumbers) push(bn uint64) {
heap.Push(pbs.q, prioritizedNumber(bn))
}
func (pbs *prioritizedNumbers) pop() uint64 {
if pbs.q.Len() == 0 {
return 0
}
item := heap.Pop(pbs.q)
return uint64(item.(prioritizedNumber))
}
func (pbs *prioritizedNumbers) length() int {
return len(*pbs.q)
}
type (
blockByNumber types.Block
// blocksByNumber is the priority queue ordered by number
blocksByNumber struct {
q *priorityQueue
cap int
}
)
func (b *blockByNumber) getBlockNumber() uint64 {
raw := (*types.Block)(b)
return raw.NumberU64()
}
func newBlocksByNumber(cap int) *blocksByNumber {
pqs := make(priorityQueue, 0)
heap.Init(&pqs)
return &blocksByNumber{
q: &pqs,
cap: cap,
}
}
func (bs *blocksByNumber) push(b *types.Block) {
heap.Push(bs.q, (*blockByNumber)(b))
for bs.q.Len() > bs.cap {
heap.Pop(bs.q)
}
}
func (bs *blocksByNumber) pop() *types.Block {
if bs.q.Len() == 0 {
return nil
}
item := heap.Pop(bs.q)
return (*types.Block)(item.(*blockByNumber))
}
func (bs *blocksByNumber) len() int {
return bs.q.Len()
}
// priorityQueue is a priorityQueue with lowest block number with highest priority
type priorityQueue []bnPrioritizedItem
// resultQueue implements heap interface
func (q priorityQueue) Len() int {
return len(q)
}
// resultQueue implements heap interface
func (q priorityQueue) Less(i, j int) bool {
bn1 := q[i].getBlockNumber()
bn2 := q[j].getBlockNumber()
return bn1 < bn2 // small block number has higher priority
}
// resultQueue implements heap interface
func (q priorityQueue) Swap(i, j int) {
q[i], q[j] = q[j], q[i]
}
// resultQueue implements heap interface
func (q *priorityQueue) Push(x interface{}) {
item, ok := x.(bnPrioritizedItem)
if !ok {
panic("wrong type of getBlockNumber interface")
}
*q = append(*q, item)
}
// resultQueue implements heap interface
func (q *priorityQueue) Pop() interface{} {
prev := *q
n := len(prev)
if n == 0 {
return nil
}
res := prev[n-1]
*q = prev[0 : n-1]
return res
}

@ -0,0 +1,261 @@
package downloader
import (
"container/heap"
"fmt"
"math/big"
"strings"
"testing"
"github.com/harmony-one/harmony/block"
headerV3 "github.com/harmony-one/harmony/block/v3"
"github.com/harmony-one/harmony/core/types"
sttypes "github.com/harmony-one/harmony/p2p/stream/types"
)
func TestResultQueue_AddBlockResults(t *testing.T) {
tests := []struct {
initBNs []uint64
addBNs []uint64
expSize int
}{
{
initBNs: []uint64{},
addBNs: []uint64{1, 2, 3, 4},
expSize: 4,
},
{
initBNs: []uint64{1, 2, 3, 4},
addBNs: []uint64{5, 6, 7, 8},
expSize: 8,
},
}
for i, test := range tests {
rq := makeTestResultQueue(test.initBNs)
rq.addBlockResults(makeTestBlocks(test.addBNs), "")
if rq.results.Len() != test.expSize {
t.Errorf("Test %v: unexpected size: %v / %v", i, rq.results.Len(), test.expSize)
}
}
}
func TestResultQueue_PopBlockResults(t *testing.T) {
tests := []struct {
initBNs []uint64
cap int
expStart uint64
expSize int
staleSize int
}{
{
initBNs: []uint64{1, 2, 3, 4, 5},
cap: 3,
expStart: 1,
expSize: 3,
staleSize: 0,
},
{
initBNs: []uint64{1, 2, 3, 4, 5},
cap: 10,
expStart: 1,
expSize: 5,
staleSize: 0,
},
{
initBNs: []uint64{1, 3, 4, 5},
cap: 10,
expStart: 1,
expSize: 1,
staleSize: 0,
},
{
initBNs: []uint64{1, 2, 3, 4, 5},
cap: 10,
expStart: 0,
expSize: 0,
staleSize: 0,
},
{
initBNs: []uint64{1, 1, 1, 1, 2},
cap: 10,
expStart: 1,
expSize: 2,
staleSize: 3,
},
{
initBNs: []uint64{1, 2, 3, 4, 5},
cap: 10,
expStart: 2,
expSize: 4,
staleSize: 1,
},
}
for i, test := range tests {
rq := makeTestResultQueue(test.initBNs)
res, stales := rq.popBlockResults(test.expStart, test.cap)
if len(res) != test.expSize {
t.Errorf("Test %v: unexpect size %v / %v", i, len(res), test.expSize)
}
if len(stales) != test.staleSize {
t.Errorf("Test %v: unexpect stale size %v / %v", i, len(stales), test.staleSize)
}
}
}
func TestResultQueue_RemoveResultsByStreamID(t *testing.T) {
tests := []struct {
rq *resultQueue
rmStreamID sttypes.StreamID
removed int
expSize int
}{
{
rq: makeTestResultQueue([]uint64{1, 2, 3, 4}),
rmStreamID: "test stream id",
removed: 4,
expSize: 0,
},
{
rq: func() *resultQueue {
rq := makeTestResultQueue([]uint64{2, 3, 4, 5})
rq.addBlockResults([]*types.Block{
makeTestBlock(1),
makeTestBlock(5),
makeTestBlock(6),
}, "another test stream id")
return rq
}(),
rmStreamID: "test stream id",
removed: 4,
expSize: 3,
},
{
rq: func() *resultQueue {
rq := makeTestResultQueue([]uint64{2, 3, 4, 5})
rq.addBlockResults([]*types.Block{
makeTestBlock(1),
makeTestBlock(5),
makeTestBlock(6),
}, "another test stream id")
return rq
}(),
rmStreamID: "another test stream id",
removed: 3,
expSize: 4,
},
}
for i, test := range tests {
res := test.rq.removeResultsByStreamID(test.rmStreamID)
if len(res) != test.removed {
t.Errorf("Test %v: unexpected number removed %v / %v", i, len(res), test.removed)
}
if gotSize := test.rq.results.Len(); gotSize != test.expSize {
t.Errorf("Test %v: unexpected number after removal %v / %v", i, gotSize, test.expSize)
}
}
}
func makeTestResultQueue(bns []uint64) *resultQueue {
rq := newResultQueue()
for _, bn := range bns {
heap.Push(rq.results, &blockResult{
block: makeTestBlock(bn),
stid: "test stream id",
})
}
return rq
}
func TestPrioritizedBlocks(t *testing.T) {
addBNs := []uint64{4, 7, 6, 9}
bns := newPrioritizedNumbers()
for _, bn := range addBNs {
bns.push(bn)
}
prevBN := uint64(0)
for len(*bns.q) > 0 {
b := bns.pop()
if b < prevBN {
t.Errorf("number not incrementing")
}
prevBN = b
}
if last := bns.pop(); last != 0 {
t.Errorf("last elem is not 0")
}
}
func TestBlocksByNumber(t *testing.T) {
addBNs := []uint64{4, 7, 6, 9}
bns := newBlocksByNumber(10)
for _, bn := range addBNs {
bns.push(makeTestBlock(bn))
}
if bns.len() != len(addBNs) {
t.Errorf("size unexpected: %v / %v", bns.len(), len(addBNs))
}
prevBN := uint64(0)
for len(*bns.q) > 0 {
b := bns.pop()
if b.NumberU64() < prevBN {
t.Errorf("number not incrementing")
}
prevBN = b.NumberU64()
}
if lastBlock := bns.pop(); lastBlock != nil {
t.Errorf("last block is not nil")
}
}
func TestPriorityQueue(t *testing.T) {
testBNs := []uint64{1, 9, 2, 4, 5, 12}
pq := make(priorityQueue, 0, 10)
heap.Init(&pq)
for _, bn := range testBNs {
heap.Push(&pq, &blockResult{
block: makeTestBlock(bn),
stid: "",
})
}
cmpBN := uint64(0)
for pq.Len() > 0 {
bn := heap.Pop(&pq).(*blockResult).block.NumberU64()
if bn < cmpBN {
t.Errorf("not incrementing")
}
cmpBN = bn
}
if pq.Len() != 0 {
t.Errorf("after poping, size not 0")
}
}
func makeTestBlocks(bns []uint64) []*types.Block {
blocks := make([]*types.Block, 0, len(bns))
for _, bn := range bns {
blocks = append(blocks, makeTestBlock(bn))
}
return blocks
}
func makeTestBlock(bn uint64) *types.Block {
testHeader := &block.Header{Header: headerV3.NewHeader()}
testHeader.SetNumber(big.NewInt(int64(bn)))
return types.NewBlockWithHeader(testHeader)
}
func assertError(got, expect error) error {
if (got == nil) != (expect == nil) {
return fmt.Errorf("unexpected error [%v] / [%v]", got, expect)
}
if (got == nil) || (expect == nil) {
return nil
}
if !strings.Contains(got.Error(), expect.Error()) {
return fmt.Errorf("unexpected error [%v] / [%v]", got, expect)
}
return nil
}
Loading…
Cancel
Save