blockchain: Reorg reactor (#3561)

* go routines in blockchain reactor

* Added reference to the go routine diagram

* Initial commit

* cleanup

* Undo testing_logger change, committed by mistake

* Fix the test loggers

* pulled some fsm code into pool.go

* added pool tests

* changes to the design

added block requests under peer

moved the request trigger in the reactor poolRoutine, triggered now by a ticker

in general moved everything required for making block requests smarter in the poolRoutine

added a simple map of heights to keep track of what will need to be requested next

added a few more tests

* send errors to FSM in a different channel than blocks

send errors (RemovePeer) from switch on a different channel than the
one receiving blocks
renamed channels
added more pool tests

* more pool tests

* lint errors

* more tests

* more tests

* switch fast sync to new implementation

* fixed data race in tests

* cleanup

* finished fsm tests

* address golangci comments :)

* address golangci comments :)

* Added timeout on next block needed to advance

* updating docs and cleanup

* fix issue in test from previous cleanup

* cleanup

* Added termination scenarios, tests and more cleanup

* small fixes to adr, comments and cleanup

* Fix bug in sendRequest()

If we tried to send a request to a peer not present in the switch, a
missing continue statement caused the request to be blackholed in a peer
that was removed and never retried.

While this bug was manifesting, the reactor kept asking for other
blocks that would be stored and never consumed. Added the number of
unconsumed blocks in the math for requesting blocks ahead of current
processing height so eventually there will be no more blocks requested
until the already received ones are consumed.

* remove bpPeer's didTimeout field

* Use distinct err codes for peer timeout and FSM timeouts

* Don't allow peers to update with lower height

* review comments from Ethan and Zarko

* some cleanup, renaming, comments

* Move block execution in separate goroutine

* Remove pool's numPending

* review comments

* fix lint, remove old blockchain reactor and duplicates in fsm tests

* small reorg around peer after review comments

* add the reactor spec

* verify block only once

* review comments

* change to int for max number of pending requests

* cleanup and godoc

* Add configuration flag fast sync version

* golangci fixes

* fix config template

* move both reactor versions under blockchain

* cleanup, golint, renaming stuff

* updated documentation, fixed more golint warnings

* integrate with behavior package

* sync with master

* gofmt

* add changelog_pending entry

* move to improvments

* suggestion to changelog entry

CHANGELOG_PENDING.md

@@ -25,6 +25,7 @@ program](https://hackerone.com/tendermint).
 - [rpc] \#3818 Make `max_body_bytes` and `max_header_bytes` configurable
 - [p2p] \#3664 p2p/conn: reuse buffer when write/read from secret connection
 - [mempool] \#3826 Make `max_msg_bytes` configurable
+- [blockchain] \#3561 Add early version of the new blockchain reactor, which is supposed to be more modular and testable compared to the old version. To try it, you'll have to change `version` in the config file, [here](https://github.com/tendermint/tendermint/blob/master/config/toml.go#L303) NOTE: It's not ready for a production yet. For further information, see [ADR-40](https://github.com/tendermint/tendermint/blob/master/docs/architecture/adr-040-blockchain-reactor-refactor.md) & [ADR-43](https://github.com/tendermint/tendermint/blob/master/docs/architecture/adr-043-blockchain-riri-org.md)
 
 ### BUG FIXES:
 

blockchain/v1/peer.go

@@ -0,0 +1,209 @@
+package v1
+
+import (
+	"fmt"
+	"math"
+	"time"
+
+	flow "github.com/tendermint/tendermint/libs/flowrate"
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+//--------
+// Peer
+
+// BpPeerParams stores the peer parameters that are used when creating a peer.
+type BpPeerParams struct {
+	timeout     time.Duration
+	minRecvRate int64
+	sampleRate  time.Duration
+	windowSize  time.Duration
+}
+
+// BpPeer is the datastructure associated with a fast sync peer.
+type BpPeer struct {
+	logger log.Logger
+	ID     p2p.ID
+
+	Height                  int64                  // the peer reported height
+	NumPendingBlockRequests int                    // number of requests still waiting for block responses
+	blocks                  map[int64]*types.Block // blocks received or expected to be received from this peer
+	blockResponseTimer      *time.Timer
+	recvMonitor             *flow.Monitor
+	params                  *BpPeerParams // parameters for timer and monitor
+
+	onErr func(err error, peerID p2p.ID) // function to call on error
+}
+
+// NewBpPeer creates a new peer.
+func NewBpPeer(
+	peerID p2p.ID, height int64, onErr func(err error, peerID p2p.ID), params *BpPeerParams) *BpPeer {
+
+	if params == nil {
+		params = BpPeerDefaultParams()
+	}
+	return &BpPeer{
+		ID:     peerID,
+		Height: height,
+		blocks: make(map[int64]*types.Block, maxRequestsPerPeer),
+		logger: log.NewNopLogger(),
+		onErr:  onErr,
+		params: params,
+	}
+}
+
+// String returns a string representation of a peer.
+func (peer *BpPeer) String() string {
+	return fmt.Sprintf("peer: %v height: %v pending: %v", peer.ID, peer.Height, peer.NumPendingBlockRequests)
+}
+
+// SetLogger sets the logger of the peer.
+func (peer *BpPeer) SetLogger(l log.Logger) {
+	peer.logger = l
+}
+
+// Cleanup performs cleanup of the peer, removes blocks, requests, stops timer and monitor.
+func (peer *BpPeer) Cleanup() {
+	if peer.blockResponseTimer != nil {
+		peer.blockResponseTimer.Stop()
+	}
+	if peer.NumPendingBlockRequests != 0 {
+		peer.logger.Info("peer with pending requests is being cleaned", "peer", peer.ID)
+	}
+	if len(peer.blocks)-peer.NumPendingBlockRequests != 0 {
+		peer.logger.Info("peer with pending blocks is being cleaned", "peer", peer.ID)
+	}
+	for h := range peer.blocks {
+		delete(peer.blocks, h)
+	}
+	peer.NumPendingBlockRequests = 0
+	peer.recvMonitor = nil
+}
+
+// BlockAtHeight returns the block at a given height if available and errMissingBlock otherwise.
+func (peer *BpPeer) BlockAtHeight(height int64) (*types.Block, error) {
+	block, ok := peer.blocks[height]
+	if !ok {
+		return nil, errMissingBlock
+	}
+	if block == nil {
+		return nil, errMissingBlock
+	}
+	return peer.blocks[height], nil
+}
+
+// AddBlock adds a block at peer level. Block must be non-nil and recvSize a positive integer
+// The peer must have a pending request for this block.
+func (peer *BpPeer) AddBlock(block *types.Block, recvSize int) error {
+	if block == nil || recvSize < 0 {
+		panic("bad parameters")
+	}
+	existingBlock, ok := peer.blocks[block.Height]
+	if !ok {
+		peer.logger.Error("unsolicited block", "blockHeight", block.Height, "peer", peer.ID)
+		return errMissingBlock
+	}
+	if existingBlock != nil {
+		peer.logger.Error("already have a block for height", "height", block.Height)
+		return errDuplicateBlock
+	}
+	if peer.NumPendingBlockRequests == 0 {
+		panic("peer does not have pending requests")
+	}
+	peer.blocks[block.Height] = block
+	peer.NumPendingBlockRequests--
+	if peer.NumPendingBlockRequests == 0 {
+		peer.stopMonitor()
+		peer.stopBlockResponseTimer()
+	} else {
+		peer.recvMonitor.Update(recvSize)
+		peer.resetBlockResponseTimer()
+	}
+	return nil
+}
+
+// RemoveBlock removes the block of given height
+func (peer *BpPeer) RemoveBlock(height int64) {
+	delete(peer.blocks, height)
+}
+
+// RequestSent records that a request was sent, and starts the peer timer and monitor if needed.
+func (peer *BpPeer) RequestSent(height int64) {
+	peer.blocks[height] = nil
+
+	if peer.NumPendingBlockRequests == 0 {
+		peer.startMonitor()
+		peer.resetBlockResponseTimer()
+	}
+	peer.NumPendingBlockRequests++
+}
+
+// CheckRate verifies that the response rate of the peer is acceptable (higher than the minimum allowed).
+func (peer *BpPeer) CheckRate() error {
+	if peer.NumPendingBlockRequests == 0 {
+		return nil
+	}
+	curRate := peer.recvMonitor.Status().CurRate
+	// curRate can be 0 on start
+	if curRate != 0 && curRate < peer.params.minRecvRate {
+		err := errSlowPeer
+		peer.logger.Error("SendTimeout", "peer", peer,
+			"reason", err,
+			"curRate", fmt.Sprintf("%d KB/s", curRate/1024),
+			"minRate", fmt.Sprintf("%d KB/s", peer.params.minRecvRate/1024))
+		return err
+	}
+	return nil
+}
+
+func (peer *BpPeer) onTimeout() {
+	peer.onErr(errNoPeerResponse, peer.ID)
+}
+
+func (peer *BpPeer) stopMonitor() {
+	peer.recvMonitor.Done()
+	peer.recvMonitor = nil
+}
+
+func (peer *BpPeer) startMonitor() {
+	peer.recvMonitor = flow.New(peer.params.sampleRate, peer.params.windowSize)
+	initialValue := float64(peer.params.minRecvRate) * math.E
+	peer.recvMonitor.SetREMA(initialValue)
+}
+
+func (peer *BpPeer) resetBlockResponseTimer() {
+	if peer.blockResponseTimer == nil {
+		peer.blockResponseTimer = time.AfterFunc(peer.params.timeout, peer.onTimeout)
+	} else {
+		peer.blockResponseTimer.Reset(peer.params.timeout)
+	}
+}
+
+func (peer *BpPeer) stopBlockResponseTimer() bool {
+	if peer.blockResponseTimer == nil {
+		return false
+	}
+	return peer.blockResponseTimer.Stop()
+}
+
+// BpPeerDefaultParams returns the default peer parameters.
+func BpPeerDefaultParams() *BpPeerParams {
+	return &BpPeerParams{
+		// Timeout for a peer to respond to a block request.
+		timeout: 15 * time.Second,
+
+		// Minimum recv rate to ensure we're receiving blocks from a peer fast
+		// enough. If a peer is not sending data at at least that rate, we
+		// consider them to have timedout and we disconnect.
+		//
+		// Assuming a DSL connection (not a good choice) 128 Kbps (upload) ~ 15 KB/s,
+		// sending data across atlantic ~ 7.5 KB/s.
+		minRecvRate: int64(7680),
+
+		// Monitor parameters
+		sampleRate: time.Second,
+		windowSize: 40 * time.Second,
+	}
+}

blockchain/v1/peer_test.go

@@ -0,0 +1,278 @@
+package v1
+
+import (
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	cmn "github.com/tendermint/tendermint/libs/common"
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+func TestPeerMonitor(t *testing.T) {
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {},
+		nil)
+	peer.SetLogger(log.TestingLogger())
+	peer.startMonitor()
+	assert.NotNil(t, peer.recvMonitor)
+	peer.stopMonitor()
+	assert.Nil(t, peer.recvMonitor)
+}
+
+func TestPeerResetBlockResponseTimer(t *testing.T) {
+	var (
+		numErrFuncCalls int        // number of calls to the errFunc
+		lastErr         error      // last generated error
+		peerTestMtx     sync.Mutex // modifications of ^^ variables are also done from timer handler goroutine
+	)
+	params := &BpPeerParams{timeout: 2 * time.Millisecond}
+
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {
+			peerTestMtx.Lock()
+			defer peerTestMtx.Unlock()
+			lastErr = err
+			numErrFuncCalls++
+		},
+		params)
+
+	peer.SetLogger(log.TestingLogger())
+	checkByStoppingPeerTimer(t, peer, false)
+
+	// initial reset call with peer having a nil timer
+	peer.resetBlockResponseTimer()
+	assert.NotNil(t, peer.blockResponseTimer)
+	// make sure timer is running and stop it
+	checkByStoppingPeerTimer(t, peer, true)
+
+	// reset with running timer
+	peer.resetBlockResponseTimer()
+	time.Sleep(time.Millisecond)
+	peer.resetBlockResponseTimer()
+	assert.NotNil(t, peer.blockResponseTimer)
+
+	// let the timer expire and ...
+	time.Sleep(3 * time.Millisecond)
+	// ... check timer is not running
+	checkByStoppingPeerTimer(t, peer, false)
+
+	peerTestMtx.Lock()
+	// ... check errNoPeerResponse has been sent
+	assert.Equal(t, 1, numErrFuncCalls)
+	assert.Equal(t, lastErr, errNoPeerResponse)
+	peerTestMtx.Unlock()
+}
+
+func TestPeerRequestSent(t *testing.T) {
+	params := &BpPeerParams{timeout: 2 * time.Millisecond}
+
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {},
+		params)
+
+	peer.SetLogger(log.TestingLogger())
+
+	peer.RequestSent(1)
+	assert.NotNil(t, peer.recvMonitor)
+	assert.NotNil(t, peer.blockResponseTimer)
+	assert.Equal(t, 1, peer.NumPendingBlockRequests)
+
+	peer.RequestSent(1)
+	assert.NotNil(t, peer.recvMonitor)
+	assert.NotNil(t, peer.blockResponseTimer)
+	assert.Equal(t, 2, peer.NumPendingBlockRequests)
+}
+
+func TestPeerGetAndRemoveBlock(t *testing.T) {
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 100,
+		func(err error, _ p2p.ID) {},
+		nil)
+
+	// Change peer height
+	peer.Height = int64(10)
+	assert.Equal(t, int64(10), peer.Height)
+
+	// request some blocks and receive few of them
+	for i := 1; i <= 10; i++ {
+		peer.RequestSent(int64(i))
+		if i > 5 {
+			// only receive blocks 1..5
+			continue
+		}
+		_ = peer.AddBlock(makeSmallBlock(i), 10)
+	}
+
+	tests := []struct {
+		name         string
+		height       int64
+		wantErr      error
+		blockPresent bool
+	}{
+		{"no request", 100, errMissingBlock, false},
+		{"no block", 6, errMissingBlock, false},
+		{"block 1 present", 1, nil, true},
+		{"block max present", 5, nil, true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// try to get the block
+			b, err := peer.BlockAtHeight(tt.height)
+			assert.Equal(t, tt.wantErr, err)
+			assert.Equal(t, tt.blockPresent, b != nil)
+
+			// remove the block
+			peer.RemoveBlock(tt.height)
+			_, err = peer.BlockAtHeight(tt.height)
+			assert.Equal(t, errMissingBlock, err)
+		})
+	}
+}
+
+func TestPeerAddBlock(t *testing.T) {
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 100,
+		func(err error, _ p2p.ID) {},
+		nil)
+
+	// request some blocks, receive one
+	for i := 1; i <= 10; i++ {
+		peer.RequestSent(int64(i))
+		if i == 5 {
+			// receive block 5
+			_ = peer.AddBlock(makeSmallBlock(i), 10)
+		}
+	}
+
+	tests := []struct {
+		name         string
+		height       int64
+		wantErr      error
+		blockPresent bool
+	}{
+		{"no request", 50, errMissingBlock, false},
+		{"duplicate block", 5, errDuplicateBlock, true},
+		{"block 1 successfully received", 1, nil, true},
+		{"block max successfully received", 10, nil, true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// try to get the block
+			err := peer.AddBlock(makeSmallBlock(int(tt.height)), 10)
+			assert.Equal(t, tt.wantErr, err)
+			_, err = peer.BlockAtHeight(tt.height)
+			assert.Equal(t, tt.blockPresent, err == nil)
+		})
+	}
+}
+
+func TestPeerOnErrFuncCalledDueToExpiration(t *testing.T) {
+
+	params := &BpPeerParams{timeout: 2 * time.Millisecond}
+	var (
+		numErrFuncCalls int        // number of calls to the onErr function
+		lastErr         error      // last generated error
+		peerTestMtx     sync.Mutex // modifications of ^^ variables are also done from timer handler goroutine
+	)
+
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {
+			peerTestMtx.Lock()
+			defer peerTestMtx.Unlock()
+			lastErr = err
+			numErrFuncCalls++
+		},
+		params)
+
+	peer.SetLogger(log.TestingLogger())
+
+	peer.RequestSent(1)
+	time.Sleep(4 * time.Millisecond)
+	// timer should have expired by now, check that the on error function was called
+	peerTestMtx.Lock()
+	assert.Equal(t, 1, numErrFuncCalls)
+	assert.Equal(t, errNoPeerResponse, lastErr)
+	peerTestMtx.Unlock()
+}
+
+func TestPeerCheckRate(t *testing.T) {
+	params := &BpPeerParams{
+		timeout:     time.Second,
+		minRecvRate: int64(100), // 100 bytes/sec exponential moving average
+	}
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {},
+		params)
+	peer.SetLogger(log.TestingLogger())
+
+	require.Nil(t, peer.CheckRate())
+
+	for i := 0; i < 40; i++ {
+		peer.RequestSent(int64(i))
+	}
+
+	// monitor starts with a higher rEMA (~ 2*minRecvRate), wait for it to go down
+	time.Sleep(900 * time.Millisecond)
+
+	// normal peer - send a bit more than 100 bytes/sec, > 10 bytes/100msec, check peer is not considered slow
+	for i := 0; i < 10; i++ {
+		_ = peer.AddBlock(makeSmallBlock(i), 11)
+		time.Sleep(100 * time.Millisecond)
+		require.Nil(t, peer.CheckRate())
+	}
+
+	// slow peer - send a bit less than 10 bytes/100msec
+	for i := 10; i < 20; i++ {
+		_ = peer.AddBlock(makeSmallBlock(i), 9)
+		time.Sleep(100 * time.Millisecond)
+	}
+	// check peer is considered slow
+	assert.Equal(t, errSlowPeer, peer.CheckRate())
+}
+
+func TestPeerCleanup(t *testing.T) {
+	params := &BpPeerParams{timeout: 2 * time.Millisecond}
+
+	peer := NewBpPeer(
+		p2p.ID(cmn.RandStr(12)), 10,
+		func(err error, _ p2p.ID) {},
+		params)
+	peer.SetLogger(log.TestingLogger())
+
+	assert.Nil(t, peer.blockResponseTimer)
+	peer.RequestSent(1)
+	assert.NotNil(t, peer.blockResponseTimer)
+
+	peer.Cleanup()
+	checkByStoppingPeerTimer(t, peer, false)
+}
+
+// Check if peer timer is running or not (a running timer can be successfully stopped).
+// Note: stops the timer.
+func checkByStoppingPeerTimer(t *testing.T, peer *BpPeer, running bool) {
+	assert.NotPanics(t, func() {
+		stopped := peer.stopBlockResponseTimer()
+		if running {
+			assert.True(t, stopped)
+		} else {
+			assert.False(t, stopped)
+		}
+	})
+}
+
+func makeSmallBlock(height int) *types.Block {
+	return types.MakeBlock(int64(height), []types.Tx{types.Tx("foo")}, nil, nil)
+}

blockchain/v1/pool.go

@@ -0,0 +1,369 @@
+package v1
+
+import (
+	"sort"
+
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+// BlockPool keeps track of the fast sync peers, block requests and block responses.
+type BlockPool struct {
+	logger log.Logger
+	// Set of peers that have sent status responses, with height bigger than pool.Height
+	peers map[p2p.ID]*BpPeer
+	// Set of block heights and the corresponding peers from where a block response is expected or has been received.
+	blocks map[int64]p2p.ID
+
+	plannedRequests   map[int64]struct{} // list of blocks to be assigned peers for blockRequest
+	nextRequestHeight int64              // next height to be added to plannedRequests
+
+	Height        int64 // height of next block to execute
+	MaxPeerHeight int64 // maximum height of all peers
+	toBcR         bcReactor
+}
+
+// NewBlockPool creates a new BlockPool.
+func NewBlockPool(height int64, toBcR bcReactor) *BlockPool {
+	return &BlockPool{
+		Height:            height,
+		MaxPeerHeight:     0,
+		peers:             make(map[p2p.ID]*BpPeer),
+		blocks:            make(map[int64]p2p.ID),
+		plannedRequests:   make(map[int64]struct{}),
+		nextRequestHeight: height,
+		toBcR:             toBcR,
+	}
+}
+
+// SetLogger sets the logger of the pool.
+func (pool *BlockPool) SetLogger(l log.Logger) {
+	pool.logger = l
+}
+
+// ReachedMaxHeight check if the pool has reached the maximum peer height.
+func (pool *BlockPool) ReachedMaxHeight() bool {
+	return pool.Height >= pool.MaxPeerHeight
+}
+
+func (pool *BlockPool) rescheduleRequest(peerID p2p.ID, height int64) {
+	pool.logger.Info("reschedule requests made to peer for height ", "peerID", peerID, "height", height)
+	pool.plannedRequests[height] = struct{}{}
+	delete(pool.blocks, height)
+	pool.peers[peerID].RemoveBlock(height)
+}
+
+// Updates the pool's max height. If no peers are left MaxPeerHeight is set to 0.
+func (pool *BlockPool) updateMaxPeerHeight() {
+	var newMax int64
+	for _, peer := range pool.peers {
+		peerHeight := peer.Height
+		if peerHeight > newMax {
+			newMax = peerHeight
+		}
+	}
+	pool.MaxPeerHeight = newMax
+}
+
+// UpdatePeer adds a new peer or updates an existing peer with a new height.
+// If a peer is short it is not added.
+func (pool *BlockPool) UpdatePeer(peerID p2p.ID, height int64) error {
+
+	peer := pool.peers[peerID]
+
+	if peer == nil {
+		if height < pool.Height {
+			pool.logger.Info("Peer height too small",
+				"peer", peerID, "height", height, "fsm_height", pool.Height)
+			return errPeerTooShort
+		}
+		// Add new peer.
+		peer = NewBpPeer(peerID, height, pool.toBcR.sendPeerError, nil)
+		peer.SetLogger(pool.logger.With("peer", peerID))
+		pool.peers[peerID] = peer
+		pool.logger.Info("added peer", "peerID", peerID, "height", height, "num_peers", len(pool.peers))
+	} else {
+		// Check if peer is lowering its height. This is not allowed.
+		if height < peer.Height {
+			pool.RemovePeer(peerID, errPeerLowersItsHeight)
+			return errPeerLowersItsHeight
+		}
+		// Update existing peer.
+		peer.Height = height
+	}
+
+	// Update the pool's MaxPeerHeight if needed.
+	pool.updateMaxPeerHeight()
+
+	return nil
+}
+
+// Cleans and deletes the peer. Recomputes the max peer height.
+func (pool *BlockPool) deletePeer(peer *BpPeer) {
+	if peer == nil {
+		return
+	}
+	peer.Cleanup()
+	delete(pool.peers, peer.ID)
+
+	if peer.Height == pool.MaxPeerHeight {
+		pool.updateMaxPeerHeight()
+	}
+}
+
+// RemovePeer removes the blocks and requests from the peer, reschedules them and deletes the peer.
+func (pool *BlockPool) RemovePeer(peerID p2p.ID, err error) {
+	peer := pool.peers[peerID]
+	if peer == nil {
+		return
+	}
+	pool.logger.Info("removing peer", "peerID", peerID, "error", err)
+
+	// Reschedule the block requests made to the peer, or received and not processed yet.
+	// Note that some of the requests may be removed further down.
+	for h := range pool.peers[peerID].blocks {
+		pool.rescheduleRequest(peerID, h)
+	}
+
+	oldMaxPeerHeight := pool.MaxPeerHeight
+	// Delete the peer. This operation may result in the pool's MaxPeerHeight being lowered.
+	pool.deletePeer(peer)
+
+	// Check if the pool's MaxPeerHeight has been lowered.
+	// This may happen if the tallest peer has been removed.
+	if oldMaxPeerHeight > pool.MaxPeerHeight {
+		// Remove any planned requests for heights over the new MaxPeerHeight.
+		for h := range pool.plannedRequests {
+			if h > pool.MaxPeerHeight {
+				delete(pool.plannedRequests, h)
+			}
+		}
+		// Adjust the nextRequestHeight to the new max plus one.
+		if pool.nextRequestHeight > pool.MaxPeerHeight {
+			pool.nextRequestHeight = pool.MaxPeerHeight + 1
+		}
+	}
+}
+
+func (pool *BlockPool) removeShortPeers() {
+	for _, peer := range pool.peers {
+		if peer.Height < pool.Height {
+			pool.RemovePeer(peer.ID, nil)
+		}
+	}
+}
+
+func (pool *BlockPool) removeBadPeers() {
+	pool.removeShortPeers()
+	for _, peer := range pool.peers {
+		if err := peer.CheckRate(); err != nil {
+			pool.RemovePeer(peer.ID, err)
+			pool.toBcR.sendPeerError(err, peer.ID)
+		}
+	}
+}
+
+// MakeNextRequests creates more requests if the block pool is running low.
+func (pool *BlockPool) MakeNextRequests(maxNumRequests int) {
+	heights := pool.makeRequestBatch(maxNumRequests)
+	if len(heights) != 0 {
+		pool.logger.Info("makeNextRequests will make following requests",
+			"number", len(heights), "heights", heights)
+	}
+
+	for _, height := range heights {
+		h := int64(height)
+		if !pool.sendRequest(h) {
+			// If a good peer was not found for sending the request at height h then return,
+			// as it shouldn't be possible to find a peer for h+1.
+			return
+		}
+		delete(pool.plannedRequests, h)
+	}
+}
+
+// Makes a batch of requests sorted by height such that the block pool has up to maxNumRequests entries.
+func (pool *BlockPool) makeRequestBatch(maxNumRequests int) []int {
+	pool.removeBadPeers()
+	// At this point pool.requests may include heights for requests to be redone due to removal of peers:
+	// - peers timed out or were removed by switch
+	// - FSM timed out on waiting to advance the block execution due to missing blocks at h or h+1
+	// Determine the number of requests needed by subtracting the number of requests already made from the maximum
+	// allowed
+	numNeeded := int(maxNumRequests) - len(pool.blocks)
+	for len(pool.plannedRequests) < numNeeded {
+		if pool.nextRequestHeight > pool.MaxPeerHeight {
+			break
+		}
+		pool.plannedRequests[pool.nextRequestHeight] = struct{}{}
+		pool.nextRequestHeight++
+	}
+
+	heights := make([]int, 0, len(pool.plannedRequests))
+	for k := range pool.plannedRequests {
+		heights = append(heights, int(k))
+	}
+	sort.Ints(heights)
+	return heights
+}
+
+func (pool *BlockPool) sendRequest(height int64) bool {
+	for _, peer := range pool.peers {
+		if peer.NumPendingBlockRequests >= maxRequestsPerPeer {
+			continue
+		}
+		if peer.Height < height {
+			continue
+		}
+
+		err := pool.toBcR.sendBlockRequest(peer.ID, height)
+		if err == errNilPeerForBlockRequest {
+			// Switch does not have this peer, remove it and continue to look for another peer.
+			pool.logger.Error("switch does not have peer..removing peer selected for height", "peer",
+				peer.ID, "height", height)
+			pool.RemovePeer(peer.ID, err)
+			continue
+		}
+
+		if err == errSendQueueFull {
+			pool.logger.Error("peer queue is full", "peer", peer.ID, "height", height)
+			continue
+		}
+
+		pool.logger.Info("assigned request to peer", "peer", peer.ID, "height", height)
+
+		pool.blocks[height] = peer.ID
+		peer.RequestSent(height)
+
+		return true
+	}
+	pool.logger.Error("could not find peer to send request for block at height", "height", height)
+	return false
+}
+
+// AddBlock validates that the block comes from the peer it was expected from and stores it in the 'blocks' map.
+func (pool *BlockPool) AddBlock(peerID p2p.ID, block *types.Block, blockSize int) error {
+	peer, ok := pool.peers[peerID]
+	if !ok {
+		pool.logger.Error("block from unknown peer", "height", block.Height, "peer", peerID)
+		return errBadDataFromPeer
+	}
+	if wantPeerID, ok := pool.blocks[block.Height]; ok && wantPeerID != peerID {
+		pool.logger.Error("block received from wrong peer", "height", block.Height,
+			"peer", peerID, "expected_peer", wantPeerID)
+		return errBadDataFromPeer
+	}
+
+	return peer.AddBlock(block, blockSize)
+}
+
+// BlockData stores the peer responsible to deliver a block and the actual block if delivered.
+type BlockData struct {
+	block *types.Block
+	peer  *BpPeer
+}
+
+// BlockAndPeerAtHeight retrieves the block and delivery peer at specified height.
+// Returns errMissingBlock if a block was not found
+func (pool *BlockPool) BlockAndPeerAtHeight(height int64) (bData *BlockData, err error) {
+	peerID := pool.blocks[height]
+	peer := pool.peers[peerID]
+	if peer == nil {
+		return nil, errMissingBlock
+	}
+
+	block, err := peer.BlockAtHeight(height)
+	if err != nil {
+		return nil, err
+	}
+
+	return &BlockData{peer: peer, block: block}, nil
+
+}
+
+// FirstTwoBlocksAndPeers returns the blocks and the delivery peers at pool's height H and H+1.
+func (pool *BlockPool) FirstTwoBlocksAndPeers() (first, second *BlockData, err error) {
+	first, err = pool.BlockAndPeerAtHeight(pool.Height)
+	second, err2 := pool.BlockAndPeerAtHeight(pool.Height + 1)
+	if err == nil {
+		err = err2
+	}
+	return
+}
+
+// InvalidateFirstTwoBlocks removes the peers that sent us the first two blocks, blocks are removed by RemovePeer().
+func (pool *BlockPool) InvalidateFirstTwoBlocks(err error) {
+	first, err1 := pool.BlockAndPeerAtHeight(pool.Height)
+	second, err2 := pool.BlockAndPeerAtHeight(pool.Height + 1)
+
+	if err1 == nil {
+		pool.RemovePeer(first.peer.ID, err)
+	}
+	if err2 == nil {
+		pool.RemovePeer(second.peer.ID, err)
+	}
+}
+
+// ProcessedCurrentHeightBlock performs cleanup after a block is processed. It removes block at pool height and
+// the peers that are now short.
+func (pool *BlockPool) ProcessedCurrentHeightBlock() {
+	peerID, peerOk := pool.blocks[pool.Height]
+	if peerOk {
+		pool.peers[peerID].RemoveBlock(pool.Height)
+	}
+	delete(pool.blocks, pool.Height)
+	pool.logger.Debug("removed block at height", "height", pool.Height)
+	pool.Height++
+	pool.removeShortPeers()
+}
+
+// RemovePeerAtCurrentHeights checks if a block at pool's height H exists and if not, it removes the
+// delivery peer and returns. If a block at height H exists then the check and peer removal is done for H+1.
+// This function is called when the FSM is not able to make progress for some time.
+// This happens if either the block H or H+1 have not been delivered.
+func (pool *BlockPool) RemovePeerAtCurrentHeights(err error) {
+	peerID := pool.blocks[pool.Height]
+	peer, ok := pool.peers[peerID]
+	if ok {
+		if _, err := peer.BlockAtHeight(pool.Height); err != nil {
+			pool.logger.Info("remove peer that hasn't sent block at pool.Height",
+				"peer", peerID, "height", pool.Height)
+			pool.RemovePeer(peerID, err)
+			return
+		}
+	}
+	peerID = pool.blocks[pool.Height+1]
+	peer, ok = pool.peers[peerID]
+	if ok {
+		if _, err := peer.BlockAtHeight(pool.Height + 1); err != nil {
+			pool.logger.Info("remove peer that hasn't sent block at pool.Height+1",
+				"peer", peerID, "height", pool.Height+1)
+			pool.RemovePeer(peerID, err)
+			return
+		}
+	}
+}
+
+// Cleanup performs pool and peer cleanup
+func (pool *BlockPool) Cleanup() {
+	for id, peer := range pool.peers {
+		peer.Cleanup()
+		delete(pool.peers, id)
+	}
+	pool.plannedRequests = make(map[int64]struct{})
+	pool.blocks = make(map[int64]p2p.ID)
+	pool.nextRequestHeight = 0
+	pool.Height = 0
+	pool.MaxPeerHeight = 0
+}
+
+// NumPeers returns the number of peers in the pool
+func (pool *BlockPool) NumPeers() int {
+	return len(pool.peers)
+}
+
+// NeedsBlocks returns true if more blocks are required.
+func (pool *BlockPool) NeedsBlocks() bool {
+	return len(pool.blocks) < maxNumRequests
+}

blockchain/v1/pool_test.go

@@ -0,0 +1,650 @@
+package v1
+
+import (
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+type testPeer struct {
+	id     p2p.ID
+	height int64
+}
+
+type testBcR struct {
+	logger log.Logger
+}
+
+type testValues struct {
+	numRequestsSent int
+}
+
+var testResults testValues
+
+func resetPoolTestResults() {
+	testResults.numRequestsSent = 0
+}
+
+func (testR *testBcR) sendPeerError(err error, peerID p2p.ID) {
+}
+
+func (testR *testBcR) sendStatusRequest() {
+}
+
+func (testR *testBcR) sendBlockRequest(peerID p2p.ID, height int64) error {
+	testResults.numRequestsSent++
+	return nil
+}
+
+func (testR *testBcR) resetStateTimer(name string, timer **time.Timer, timeout time.Duration) {
+}
+
+func (testR *testBcR) switchToConsensus() {
+
+}
+
+func newTestBcR() *testBcR {
+	testBcR := &testBcR{logger: log.TestingLogger()}
+	return testBcR
+}
+
+type tPBlocks struct {
+	id     p2p.ID
+	create bool
+}
+
+// Makes a block pool with specified current height, list of peers, block requests and block responses
+func makeBlockPool(bcr *testBcR, height int64, peers []BpPeer, blocks map[int64]tPBlocks) *BlockPool {
+	bPool := NewBlockPool(height, bcr)
+	bPool.SetLogger(bcr.logger)
+
+	txs := []types.Tx{types.Tx("foo"), types.Tx("bar")}
+
+	var maxH int64
+	for _, p := range peers {
+		if p.Height > maxH {
+			maxH = p.Height
+		}
+		bPool.peers[p.ID] = NewBpPeer(p.ID, p.Height, bcr.sendPeerError, nil)
+		bPool.peers[p.ID].SetLogger(bcr.logger)
+
+	}
+	bPool.MaxPeerHeight = maxH
+	for h, p := range blocks {
+		bPool.blocks[h] = p.id
+		bPool.peers[p.id].RequestSent(int64(h))
+		if p.create {
+			// simulate that a block at height h has been received
+			_ = bPool.peers[p.id].AddBlock(types.MakeBlock(int64(h), txs, nil, nil), 100)
+		}
+	}
+	return bPool
+}
+
+func assertPeerSetsEquivalent(t *testing.T, set1 map[p2p.ID]*BpPeer, set2 map[p2p.ID]*BpPeer) {
+	assert.Equal(t, len(set1), len(set2))
+	for peerID, peer1 := range set1 {
+		peer2 := set2[peerID]
+		assert.NotNil(t, peer2)
+		assert.Equal(t, peer1.NumPendingBlockRequests, peer2.NumPendingBlockRequests)
+		assert.Equal(t, peer1.Height, peer2.Height)
+		assert.Equal(t, len(peer1.blocks), len(peer2.blocks))
+		for h, block1 := range peer1.blocks {
+			block2 := peer2.blocks[h]
+			// block1 and block2 could be nil if a request was made but no block was received
+			assert.Equal(t, block1, block2)
+		}
+	}
+}
+
+func assertBlockPoolEquivalent(t *testing.T, poolWanted, pool *BlockPool) {
+	assert.Equal(t, poolWanted.blocks, pool.blocks)
+	assertPeerSetsEquivalent(t, poolWanted.peers, pool.peers)
+	assert.Equal(t, poolWanted.MaxPeerHeight, pool.MaxPeerHeight)
+	assert.Equal(t, poolWanted.Height, pool.Height)
+
+}
+
+func TestBlockPoolUpdatePeer(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		args       testPeer
+		poolWanted *BlockPool
+		errWanted  error
+	}{
+		{
+			name:       "add a first short peer",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+			args:       testPeer{"P1", 50},
+			errWanted:  errPeerTooShort,
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "add a first good peer",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+			args:       testPeer{"P1", 101},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 101}}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "increase the height of P1 from 120 to 123",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}}, map[int64]tPBlocks{}),
+			args:       testPeer{"P1", 123},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 123}}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "decrease the height of P1 from 120 to 110",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}}, map[int64]tPBlocks{}),
+			args:       testPeer{"P1", 110},
+			errWanted:  errPeerLowersItsHeight,
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "decrease the height of P1 from 105 to 102 with blocks",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 105}},
+				map[int64]tPBlocks{
+					100: {"P1", true}, 101: {"P1", true}, 102: {"P1", true}}),
+			args:      testPeer{"P1", 102},
+			errWanted: errPeerLowersItsHeight,
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{},
+				map[int64]tPBlocks{}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			pool := tt.pool
+			err := pool.UpdatePeer(tt.args.id, tt.args.height)
+			assert.Equal(t, tt.errWanted, err)
+			assert.Equal(t, tt.poolWanted.blocks, tt.pool.blocks)
+			assertPeerSetsEquivalent(t, tt.poolWanted.peers, tt.pool.peers)
+			assert.Equal(t, tt.poolWanted.MaxPeerHeight, tt.pool.MaxPeerHeight)
+		})
+	}
+}
+
+func TestBlockPoolRemovePeer(t *testing.T) {
+	testBcR := newTestBcR()
+
+	type args struct {
+		peerID p2p.ID
+		err    error
+	}
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		args       args
+		poolWanted *BlockPool
+	}{
+		{
+			name:       "attempt to delete non-existing peer",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}}, map[int64]tPBlocks{}),
+			args:       args{"P99", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "delete the only peer without blocks",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}}, map[int64]tPBlocks{}),
+			args:       args{"P1", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "delete the shortest of two peers without blocks",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 120}}, map[int64]tPBlocks{}),
+			args:       args{"P1", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P2", Height: 120}}, map[int64]tPBlocks{}),
+		},
+		{
+			name:       "delete the tallest of two peers without blocks",
+			pool:       makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 120}}, map[int64]tPBlocks{}),
+			args:       args{"P2", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 100}}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "delete the only peer with block requests sent and blocks received",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}},
+				map[int64]tPBlocks{100: {"P1", true}, 101: {"P1", false}}),
+			args:       args{"P1", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "delete the shortest of two peers with block requests sent and blocks received",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 200}},
+				map[int64]tPBlocks{100: {"P1", true}, 101: {"P1", false}}),
+			args:       args{"P1", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P2", Height: 200}}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "delete the tallest of two peers with block requests sent and blocks received",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 110}},
+				map[int64]tPBlocks{100: {"P1", true}, 101: {"P1", false}}),
+			args:       args{"P1", nil},
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P2", Height: 110}}, map[int64]tPBlocks{}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.pool.RemovePeer(tt.args.peerID, tt.args.err)
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}
+
+func TestBlockPoolRemoveShortPeers(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		poolWanted *BlockPool
+	}{
+		{
+			name: "no short peers",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 110}, {ID: "P3", Height: 120}}, map[int64]tPBlocks{}),
+			poolWanted: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 110}, {ID: "P3", Height: 120}}, map[int64]tPBlocks{}),
+		},
+
+		{
+			name: "one short peer",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 90}, {ID: "P3", Height: 120}}, map[int64]tPBlocks{}),
+			poolWanted: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P3", Height: 120}}, map[int64]tPBlocks{}),
+		},
+
+		{
+			name: "all short peers",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 90}, {ID: "P2", Height: 91}, {ID: "P3", Height: 92}}, map[int64]tPBlocks{}),
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			pool := tt.pool
+			pool.removeShortPeers()
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}
+
+func TestBlockPoolSendRequestBatch(t *testing.T) {
+	type testPeerResult struct {
+		id                      p2p.ID
+		numPendingBlockRequests int
+	}
+
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name                       string
+		pool                       *BlockPool
+		maxRequestsPerPeer         int
+		expRequests                map[int64]bool
+		expPeerResults             []testPeerResult
+		expnumPendingBlockRequests int
+	}{
+		{
+			name:                       "one peer - send up to maxRequestsPerPeer block requests",
+			pool:                       makeBlockPool(testBcR, 10, []BpPeer{{ID: "P1", Height: 100}}, map[int64]tPBlocks{}),
+			maxRequestsPerPeer:         2,
+			expRequests:                map[int64]bool{10: true, 11: true},
+			expPeerResults:             []testPeerResult{{id: "P1", numPendingBlockRequests: 2}},
+			expnumPendingBlockRequests: 2,
+		},
+		{
+			name:               "n peers - send n*maxRequestsPerPeer block requests",
+			pool:               makeBlockPool(testBcR, 10, []BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}}, map[int64]tPBlocks{}),
+			maxRequestsPerPeer: 2,
+			expRequests:        map[int64]bool{10: true, 11: true},
+			expPeerResults: []testPeerResult{
+				{id: "P1", numPendingBlockRequests: 2},
+				{id: "P2", numPendingBlockRequests: 2}},
+			expnumPendingBlockRequests: 4,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			resetPoolTestResults()
+
+			var pool = tt.pool
+			maxRequestsPerPeer = tt.maxRequestsPerPeer
+			pool.MakeNextRequests(10)
+			assert.Equal(t, testResults.numRequestsSent, maxRequestsPerPeer*len(pool.peers))
+
+			for _, tPeer := range tt.expPeerResults {
+				var peer = pool.peers[tPeer.id]
+				assert.NotNil(t, peer)
+				assert.Equal(t, tPeer.numPendingBlockRequests, peer.NumPendingBlockRequests)
+			}
+			assert.Equal(t, testResults.numRequestsSent, maxRequestsPerPeer*len(pool.peers))
+
+		})
+	}
+}
+
+func TestBlockPoolAddBlock(t *testing.T) {
+	testBcR := newTestBcR()
+	txs := []types.Tx{types.Tx("foo"), types.Tx("bar")}
+
+	type args struct {
+		peerID    p2p.ID
+		block     *types.Block
+		blockSize int
+	}
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		args       args
+		poolWanted *BlockPool
+		errWanted  error
+	}{
+		{name: "block from unknown peer",
+			pool: makeBlockPool(testBcR, 10, []BpPeer{{ID: "P1", Height: 100}}, map[int64]tPBlocks{}),
+			args: args{
+				peerID:    "P2",
+				block:     types.MakeBlock(int64(10), txs, nil, nil),
+				blockSize: 100,
+			},
+			poolWanted: makeBlockPool(testBcR, 10, []BpPeer{{ID: "P1", Height: 100}}, map[int64]tPBlocks{}),
+			errWanted:  errBadDataFromPeer,
+		},
+		{name: "unexpected block 11 from known peer - waiting for 10",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", false}}),
+			args: args{
+				peerID:    "P1",
+				block:     types.MakeBlock(int64(11), txs, nil, nil),
+				blockSize: 100,
+			},
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", false}}),
+			errWanted: errMissingBlock,
+		},
+		{name: "unexpected block 10 from known peer - already have 10",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", true}, 11: {"P1", false}}),
+			args: args{
+				peerID:    "P1",
+				block:     types.MakeBlock(int64(10), txs, nil, nil),
+				blockSize: 100,
+			},
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", true}, 11: {"P1", false}}),
+			errWanted: errDuplicateBlock,
+		},
+		{name: "unexpected block 10 from known peer P2 - expected 10 to come from P1",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", false}}),
+			args: args{
+				peerID:    "P2",
+				block:     types.MakeBlock(int64(10), txs, nil, nil),
+				blockSize: 100,
+			},
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", false}}),
+			errWanted: errBadDataFromPeer,
+		},
+		{name: "expected block from known peer",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", false}}),
+			args: args{
+				peerID:    "P1",
+				block:     types.MakeBlock(int64(10), txs, nil, nil),
+				blockSize: 100,
+			},
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", true}}),
+			errWanted: nil,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			err := tt.pool.AddBlock(tt.args.peerID, tt.args.block, tt.args.blockSize)
+			assert.Equal(t, tt.errWanted, err)
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}
+
+func TestBlockPoolFirstTwoBlocksAndPeers(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name         string
+		pool         *BlockPool
+		firstWanted  int64
+		secondWanted int64
+		errWanted    error
+	}{
+		{
+			name: "both blocks missing",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{15: {"P1", true}, 16: {"P2", true}}),
+			errWanted: errMissingBlock,
+		},
+		{
+			name: "second block missing",
+			pool: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{15: {"P1", true}, 18: {"P2", true}}),
+			firstWanted: 15,
+			errWanted:   errMissingBlock,
+		},
+		{
+			name: "first block missing",
+			pool: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{16: {"P2", true}, 18: {"P2", true}}),
+			secondWanted: 16,
+			errWanted:    errMissingBlock,
+		},
+		{
+			name: "both blocks present",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", true}, 11: {"P2", true}}),
+			firstWanted:  10,
+			secondWanted: 11,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			pool := tt.pool
+			gotFirst, gotSecond, err := pool.FirstTwoBlocksAndPeers()
+			assert.Equal(t, tt.errWanted, err)
+
+			if tt.firstWanted != 0 {
+				peer := pool.blocks[tt.firstWanted]
+				block := pool.peers[peer].blocks[tt.firstWanted]
+				assert.Equal(t, block, gotFirst.block,
+					"BlockPool.FirstTwoBlocksAndPeers() gotFirst = %v, want %v",
+					tt.firstWanted, gotFirst.block.Height)
+			}
+
+			if tt.secondWanted != 0 {
+				peer := pool.blocks[tt.secondWanted]
+				block := pool.peers[peer].blocks[tt.secondWanted]
+				assert.Equal(t, block, gotSecond.block,
+					"BlockPool.FirstTwoBlocksAndPeers() gotFirst = %v, want %v",
+					tt.secondWanted, gotSecond.block.Height)
+			}
+		})
+	}
+}
+
+func TestBlockPoolInvalidateFirstTwoBlocks(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		poolWanted *BlockPool
+	}{
+		{
+			name: "both blocks missing",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{15: {"P1", true}, 16: {"P2", true}}),
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{15: {"P1", true}, 16: {"P2", true}}),
+		},
+		{
+			name: "second block missing",
+			pool: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{15: {"P1", true}, 18: {"P2", true}}),
+			poolWanted: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P2", Height: 100}},
+				map[int64]tPBlocks{18: {"P2", true}}),
+		},
+		{
+			name: "first block missing",
+			pool: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{18: {"P1", true}, 16: {"P2", true}}),
+			poolWanted: makeBlockPool(testBcR, 15,
+				[]BpPeer{{ID: "P1", Height: 100}},
+				map[int64]tPBlocks{18: {"P1", true}}),
+		},
+		{
+			name: "both blocks present",
+			pool: makeBlockPool(testBcR, 10,
+				[]BpPeer{{ID: "P1", Height: 100}, {ID: "P2", Height: 100}},
+				map[int64]tPBlocks{10: {"P1", true}, 11: {"P2", true}}),
+			poolWanted: makeBlockPool(testBcR, 10,
+				[]BpPeer{},
+				map[int64]tPBlocks{}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.pool.InvalidateFirstTwoBlocks(errNoPeerResponse)
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}
+
+func TestProcessedCurrentHeightBlock(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		poolWanted *BlockPool
+	}{
+		{
+			name: "one peer",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}},
+				map[int64]tPBlocks{100: {"P1", true}, 101: {"P1", true}}),
+			poolWanted: makeBlockPool(testBcR, 101, []BpPeer{{ID: "P1", Height: 120}},
+				map[int64]tPBlocks{101: {"P1", true}}),
+		},
+		{
+			name: "multiple peers",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					100: {"P1", true}, 104: {"P1", true}, 105: {"P1", false},
+					101: {"P2", true}, 103: {"P2", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+			poolWanted: makeBlockPool(testBcR, 101,
+				[]BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					104: {"P1", true}, 105: {"P1", false},
+					101: {"P2", true}, 103: {"P2", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.pool.ProcessedCurrentHeightBlock()
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}
+
+func TestRemovePeerAtCurrentHeight(t *testing.T) {
+	testBcR := newTestBcR()
+
+	tests := []struct {
+		name       string
+		pool       *BlockPool
+		poolWanted *BlockPool
+	}{
+		{
+			name: "one peer, remove peer for block at H",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}},
+				map[int64]tPBlocks{100: {"P1", false}, 101: {"P1", true}}),
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "one peer, remove peer for block at H+1",
+			pool: makeBlockPool(testBcR, 100, []BpPeer{{ID: "P1", Height: 120}},
+				map[int64]tPBlocks{100: {"P1", true}, 101: {"P1", false}}),
+			poolWanted: makeBlockPool(testBcR, 100, []BpPeer{}, map[int64]tPBlocks{}),
+		},
+		{
+			name: "multiple peers, remove peer for block at H",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					100: {"P1", false}, 104: {"P1", true}, 105: {"P1", false},
+					101: {"P2", true}, 103: {"P2", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+			poolWanted: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P2", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					101: {"P2", true}, 103: {"P2", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+		},
+		{
+			name: "multiple peers, remove peer for block at H+1",
+			pool: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 120}, {ID: "P2", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					100: {"P1", true}, 104: {"P1", true}, 105: {"P1", false},
+					101: {"P2", false}, 103: {"P2", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+			poolWanted: makeBlockPool(testBcR, 100,
+				[]BpPeer{{ID: "P1", Height: 120}, {ID: "P3", Height: 130}},
+				map[int64]tPBlocks{
+					100: {"P1", true}, 104: {"P1", true}, 105: {"P1", false},
+					102: {"P3", true}, 106: {"P3", true}}),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.pool.RemovePeerAtCurrentHeights(errNoPeerResponse)
+			assertBlockPoolEquivalent(t, tt.poolWanted, tt.pool)
+		})
+	}
+}

blockchain/v1/reactor.go

@@ -0,0 +1,620 @@
+package v1
+
+import (
+	"errors"
+	"fmt"
+	"reflect"
+	"time"
+
+	amino "github.com/tendermint/go-amino"
+	"github.com/tendermint/tendermint/behaviour"
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
+	"github.com/tendermint/tendermint/types"
+)
+
+const (
+	// BlockchainChannel is a channel for blocks and status updates (`BlockStore` height)
+	BlockchainChannel = byte(0x40)
+	trySyncIntervalMS = 10
+	trySendIntervalMS = 10
+
+	// ask for best height every 10s
+	statusUpdateIntervalSeconds = 10
+
+	// NOTE: keep up to date with bcBlockResponseMessage
+	bcBlockResponseMessagePrefixSize   = 4
+	bcBlockResponseMessageFieldKeySize = 1
+	maxMsgSize                         = types.MaxBlockSizeBytes +
+		bcBlockResponseMessagePrefixSize +
+		bcBlockResponseMessageFieldKeySize
+)
+
+var (
+	// Maximum number of requests that can be pending per peer, i.e. for which requests have been sent but blocks
+	// have not been received.
+	maxRequestsPerPeer = 20
+	// Maximum number of block requests for the reactor, pending or for which blocks have been received.
+	maxNumRequests = 64
+)
+
+type consensusReactor interface {
+	// for when we switch from blockchain reactor and fast sync to
+	// the consensus machine
+	SwitchToConsensus(sm.State, int)
+}
+
+// BlockchainReactor handles long-term catchup syncing.
+type BlockchainReactor struct {
+	p2p.BaseReactor
+
+	initialState sm.State // immutable
+	state        sm.State
+
+	blockExec *sm.BlockExecutor
+	store     *store.BlockStore
+
+	fastSync bool
+
+	fsm          *BcReactorFSM
+	blocksSynced int
+
+	// Receive goroutine forwards messages to this channel to be processed in the context of the poolRoutine.
+	messagesForFSMCh chan bcReactorMessage
+
+	// Switch goroutine may send RemovePeer to the blockchain reactor. This is an error message that is relayed
+	// to this channel to be processed in the context of the poolRoutine.
+	errorsForFSMCh chan bcReactorMessage
+
+	// This channel is used by the FSM and indirectly the block pool to report errors to the blockchain reactor and
+	// the switch.
+	eventsFromFSMCh chan bcFsmMessage
+
+	swReporter *behaviour.SwitchReporter
+}
+
+// NewBlockchainReactor returns new reactor instance.
+func NewBlockchainReactor(state sm.State, blockExec *sm.BlockExecutor, store *store.BlockStore,
+	fastSync bool) *BlockchainReactor {
+
+	if state.LastBlockHeight != store.Height() {
+		panic(fmt.Sprintf("state (%v) and store (%v) height mismatch", state.LastBlockHeight,
+			store.Height()))
+	}
+
+	const capacity = 1000
+	eventsFromFSMCh := make(chan bcFsmMessage, capacity)
+	messagesForFSMCh := make(chan bcReactorMessage, capacity)
+	errorsForFSMCh := make(chan bcReactorMessage, capacity)
+
+	startHeight := store.Height() + 1
+	bcR := &BlockchainReactor{
+		initialState:     state,
+		state:            state,
+		blockExec:        blockExec,
+		fastSync:         fastSync,
+		store:            store,
+		messagesForFSMCh: messagesForFSMCh,
+		eventsFromFSMCh:  eventsFromFSMCh,
+		errorsForFSMCh:   errorsForFSMCh,
+	}
+	fsm := NewFSM(startHeight, bcR)
+	bcR.fsm = fsm
+	bcR.BaseReactor = *p2p.NewBaseReactor("BlockchainReactor", bcR)
+	//bcR.swReporter = behaviour.NewSwitcReporter(bcR.BaseReactor.Switch)
+
+	return bcR
+}
+
+// bcReactorMessage is used by the reactor to send messages to the FSM.
+type bcReactorMessage struct {
+	event bReactorEvent
+	data  bReactorEventData
+}
+
+type bFsmEvent uint
+
+const (
+	// message type events
+	peerErrorEv = iota + 1
+	syncFinishedEv
+)
+
+type bFsmEventData struct {
+	peerID p2p.ID
+	err    error
+}
+
+// bcFsmMessage is used by the FSM to send messages to the reactor
+type bcFsmMessage struct {
+	event bFsmEvent
+	data  bFsmEventData
+}
+
+// SetLogger implements cmn.Service by setting the logger on reactor and pool.
+func (bcR *BlockchainReactor) SetLogger(l log.Logger) {
+	bcR.BaseService.Logger = l
+	bcR.fsm.SetLogger(l)
+}
+
+// OnStart implements cmn.Service.
+func (bcR *BlockchainReactor) OnStart() error {
+	bcR.swReporter = behaviour.NewSwitcReporter(bcR.BaseReactor.Switch)
+	if bcR.fastSync {
+		go bcR.poolRoutine()
+	}
+	return nil
+}
+
+// OnStop implements cmn.Service.
+func (bcR *BlockchainReactor) OnStop() {
+	_ = bcR.Stop()
+}
+
+// GetChannels implements Reactor
+func (bcR *BlockchainReactor) GetChannels() []*p2p.ChannelDescriptor {
+	return []*p2p.ChannelDescriptor{
+		{
+			ID:                  BlockchainChannel,
+			Priority:            10,
+			SendQueueCapacity:   2000,
+			RecvBufferCapacity:  50 * 4096,
+			RecvMessageCapacity: maxMsgSize,
+		},
+	}
+}
+
+// AddPeer implements Reactor by sending our state to peer.
+func (bcR *BlockchainReactor) AddPeer(peer p2p.Peer) {
+	msgBytes := cdc.MustMarshalBinaryBare(&bcStatusResponseMessage{bcR.store.Height()})
+	if !peer.Send(BlockchainChannel, msgBytes) {
+		// doing nothing, will try later in `poolRoutine`
+	}
+	// peer is added to the pool once we receive the first
+	// bcStatusResponseMessage from the peer and call pool.updatePeer()
+}
+
+// sendBlockToPeer loads a block and sends it to the requesting peer.
+// If the block doesn't exist a bcNoBlockResponseMessage is sent.
+// If all nodes are honest, no node should be requesting for a block that doesn't exist.
+func (bcR *BlockchainReactor) sendBlockToPeer(msg *bcBlockRequestMessage,
+	src p2p.Peer) (queued bool) {
+
+	block := bcR.store.LoadBlock(msg.Height)
+	if block != nil {
+		msgBytes := cdc.MustMarshalBinaryBare(&bcBlockResponseMessage{Block: block})
+		return src.TrySend(BlockchainChannel, msgBytes)
+	}
+
+	bcR.Logger.Info("peer asking for a block we don't have", "src", src, "height", msg.Height)
+
+	msgBytes := cdc.MustMarshalBinaryBare(&bcNoBlockResponseMessage{Height: msg.Height})
+	return src.TrySend(BlockchainChannel, msgBytes)
+}
+
+func (bcR *BlockchainReactor) sendStatusResponseToPeer(msg *bcStatusRequestMessage, src p2p.Peer) (queued bool) {
+	msgBytes := cdc.MustMarshalBinaryBare(&bcStatusResponseMessage{bcR.store.Height()})
+	return src.TrySend(BlockchainChannel, msgBytes)
+}
+
+// RemovePeer implements Reactor by removing peer from the pool.
+func (bcR *BlockchainReactor) RemovePeer(peer p2p.Peer, reason interface{}) {
+	msgData := bcReactorMessage{
+		event: peerRemoveEv,
+		data: bReactorEventData{
+			peerID: peer.ID(),
+			err:    errSwitchRemovesPeer,
+		},
+	}
+	bcR.errorsForFSMCh <- msgData
+}
+
+// Receive implements Reactor by handling 4 types of messages (look below).
+func (bcR *BlockchainReactor) Receive(chID byte, src p2p.Peer, msgBytes []byte) {
+	msg, err := decodeMsg(msgBytes)
+	if err != nil {
+		bcR.Logger.Error("error decoding message",
+			"src", src, "chId", chID, "msg", msg, "err", err, "bytes", msgBytes)
+		_ = bcR.swReporter.Report(behaviour.BadMessage(src.ID(), err.Error()))
+		return
+	}
+
+	if err = msg.ValidateBasic(); err != nil {
+		bcR.Logger.Error("peer sent us invalid msg", "peer", src, "msg", msg, "err", err)
+		_ = bcR.swReporter.Report(behaviour.BadMessage(src.ID(), err.Error()))
+		return
+	}
+
+	bcR.Logger.Debug("Receive", "src", src, "chID", chID, "msg", msg)
+
+	switch msg := msg.(type) {
+	case *bcBlockRequestMessage:
+		if queued := bcR.sendBlockToPeer(msg, src); !queued {
+			// Unfortunately not queued since the queue is full.
+			bcR.Logger.Error("Could not send block message to peer", "src", src, "height", msg.Height)
+		}
+
+	case *bcStatusRequestMessage:
+		// Send peer our state.
+		if queued := bcR.sendStatusResponseToPeer(msg, src); !queued {
+			// Unfortunately not queued since the queue is full.
+			bcR.Logger.Error("Could not send status message to peer", "src", src)
+		}
+
+	case *bcBlockResponseMessage:
+		msgForFSM := bcReactorMessage{
+			event: blockResponseEv,
+			data: bReactorEventData{
+				peerID: src.ID(),
+				height: msg.Block.Height,
+				block:  msg.Block,
+				length: len(msgBytes),
+			},
+		}
+		bcR.Logger.Info("Received", "src", src, "height", msg.Block.Height)
+		bcR.messagesForFSMCh <- msgForFSM
+
+	case *bcStatusResponseMessage:
+		// Got a peer status. Unverified.
+		msgForFSM := bcReactorMessage{
+			event: statusResponseEv,
+			data: bReactorEventData{
+				peerID: src.ID(),
+				height: msg.Height,
+				length: len(msgBytes),
+			},
+		}
+		bcR.messagesForFSMCh <- msgForFSM
+
+	default:
+		bcR.Logger.Error(fmt.Sprintf("unknown message type %v", reflect.TypeOf(msg)))
+	}
+}
+
+// processBlocksRoutine processes blocks until signlaed to stop over the stopProcessing channel
+func (bcR *BlockchainReactor) processBlocksRoutine(stopProcessing chan struct{}) {
+
+	processReceivedBlockTicker := time.NewTicker(trySyncIntervalMS * time.Millisecond)
+	doProcessBlockCh := make(chan struct{}, 1)
+
+	lastHundred := time.Now()
+	lastRate := 0.0
+
+ForLoop:
+	for {
+		select {
+		case <-stopProcessing:
+			bcR.Logger.Info("finishing block execution")
+			break ForLoop
+		case <-processReceivedBlockTicker.C: // try to execute blocks
+			select {
+			case doProcessBlockCh <- struct{}{}:
+			default:
+			}
+		case <-doProcessBlockCh:
+			for {
+				err := bcR.processBlock()
+				if err == errMissingBlock {
+					break
+				}
+				// Notify FSM of block processing result.
+				msgForFSM := bcReactorMessage{
+					event: processedBlockEv,
+					data: bReactorEventData{
+						err: err,
+					},
+				}
+				_ = bcR.fsm.Handle(&msgForFSM)
+
+				if err != nil {
+					break
+				}
+
+				bcR.blocksSynced++
+				if bcR.blocksSynced%100 == 0 {
+					lastRate = 0.9*lastRate + 0.1*(100/time.Since(lastHundred).Seconds())
+					height, maxPeerHeight := bcR.fsm.Status()
+					bcR.Logger.Info("Fast Sync Rate", "height", height,
+						"max_peer_height", maxPeerHeight, "blocks/s", lastRate)
+					lastHundred = time.Now()
+				}
+			}
+		}
+	}
+}
+
+// poolRoutine receives and handles messages from the Receive() routine and from the FSM.
+func (bcR *BlockchainReactor) poolRoutine() {
+
+	bcR.fsm.Start()
+
+	sendBlockRequestTicker := time.NewTicker(trySendIntervalMS * time.Millisecond)
+	statusUpdateTicker := time.NewTicker(statusUpdateIntervalSeconds * time.Second)
+
+	stopProcessing := make(chan struct{}, 1)
+	go bcR.processBlocksRoutine(stopProcessing)
+
+ForLoop:
+	for {
+		select {
+
+		case <-sendBlockRequestTicker.C:
+			if !bcR.fsm.NeedsBlocks() {
+				continue
+			}
+			_ = bcR.fsm.Handle(&bcReactorMessage{
+				event: makeRequestsEv,
+				data: bReactorEventData{
+					maxNumRequests: maxNumRequests}})
+
+		case <-statusUpdateTicker.C:
+			// Ask for status updates.
+			go bcR.sendStatusRequest()
+
+		case msg := <-bcR.messagesForFSMCh:
+			// Sent from the Receive() routine when status (statusResponseEv) and
+			// block (blockResponseEv) response events are received
+			_ = bcR.fsm.Handle(&msg)
+
+		case msg := <-bcR.errorsForFSMCh:
+			// Sent from the switch.RemovePeer() routine (RemovePeerEv) and
+			// FSM state timer expiry routine (stateTimeoutEv).
+			_ = bcR.fsm.Handle(&msg)
+
+		case msg := <-bcR.eventsFromFSMCh:
+			switch msg.event {
+			case syncFinishedEv:
+				stopProcessing <- struct{}{}
+				// Sent from the FSM when it enters finished state.
+				break ForLoop
+			case peerErrorEv:
+				// Sent from the FSM when it detects peer error
+				bcR.reportPeerErrorToSwitch(msg.data.err, msg.data.peerID)
+				if msg.data.err == errNoPeerResponse {
+					// Sent from the peer timeout handler routine
+					_ = bcR.fsm.Handle(&bcReactorMessage{
+						event: peerRemoveEv,
+						data: bReactorEventData{
+							peerID: msg.data.peerID,
+							err:    msg.data.err,
+						},
+					})
+				} else {
+					// For slow peers, or errors due to blocks received from wrong peer
+					// the FSM had already removed the peers
+				}
+			default:
+				bcR.Logger.Error("Event from FSM not supported", "type", msg.event)
+			}
+
+		case <-bcR.Quit():
+			break ForLoop
+		}
+	}
+}
+
+func (bcR *BlockchainReactor) reportPeerErrorToSwitch(err error, peerID p2p.ID) {
+	peer := bcR.Switch.Peers().Get(peerID)
+	if peer != nil {
+		_ = bcR.swReporter.Report(behaviour.BadMessage(peerID, err.Error()))
+	}
+}
+
+func (bcR *BlockchainReactor) processBlock() error {
+
+	first, second, err := bcR.fsm.FirstTwoBlocks()
+	if err != nil {
+		// We need both to sync the first block.
+		return err
+	}
+
+	chainID := bcR.initialState.ChainID
+
+	firstParts := first.MakePartSet(types.BlockPartSizeBytes)
+	firstPartsHeader := firstParts.Header()
+	firstID := types.BlockID{Hash: first.Hash(), PartsHeader: firstPartsHeader}
+	// Finally, verify the first block using the second's commit
+	// NOTE: we can probably make this more efficient, but note that calling
+	// first.Hash() doesn't verify the tx contents, so MakePartSet() is
+	// currently necessary.
+	err = bcR.state.Validators.VerifyCommit(chainID, firstID, first.Height, second.LastCommit)
+	if err != nil {
+		bcR.Logger.Error("error during commit verification", "err", err,
+			"first", first.Height, "second", second.Height)
+		return errBlockVerificationFailure
+	}
+
+	bcR.store.SaveBlock(first, firstParts, second.LastCommit)
+
+	bcR.state, err = bcR.blockExec.ApplyBlock(bcR.state, firstID, first)
+	if err != nil {
+		panic(fmt.Sprintf("failed to process committed block (%d:%X): %v", first.Height, first.Hash(), err))
+	}
+
+	return nil
+}
+
+// Implements bcRNotifier
+// sendStatusRequest broadcasts `BlockStore` height.
+func (bcR *BlockchainReactor) sendStatusRequest() {
+	msgBytes := cdc.MustMarshalBinaryBare(&bcStatusRequestMessage{bcR.store.Height()})
+	bcR.Switch.Broadcast(BlockchainChannel, msgBytes)
+}
+
+// Implements bcRNotifier
+// BlockRequest sends `BlockRequest` height.
+func (bcR *BlockchainReactor) sendBlockRequest(peerID p2p.ID, height int64) error {
+	peer := bcR.Switch.Peers().Get(peerID)
+	if peer == nil {
+		return errNilPeerForBlockRequest
+	}
+
+	msgBytes := cdc.MustMarshalBinaryBare(&bcBlockRequestMessage{height})
+	queued := peer.TrySend(BlockchainChannel, msgBytes)
+	if !queued {
+		return errSendQueueFull
+	}
+	return nil
+}
+
+// Implements bcRNotifier
+func (bcR *BlockchainReactor) switchToConsensus() {
+	conR, ok := bcR.Switch.Reactor("CONSENSUS").(consensusReactor)
+	if ok {
+		conR.SwitchToConsensus(bcR.state, bcR.blocksSynced)
+		bcR.eventsFromFSMCh <- bcFsmMessage{event: syncFinishedEv}
+	} else {
+		// Should only happen during testing.
+	}
+}
+
+// Implements bcRNotifier
+// Called by FSM and pool:
+// - pool calls when it detects slow peer or when peer times out
+// - FSM calls when:
+//    - adding a block (addBlock) fails
+//    - reactor processing of a block reports failure and FSM sends back the peers of first and second blocks
+func (bcR *BlockchainReactor) sendPeerError(err error, peerID p2p.ID) {
+	bcR.Logger.Info("sendPeerError:", "peer", peerID, "error", err)
+	msgData := bcFsmMessage{
+		event: peerErrorEv,
+		data: bFsmEventData{
+			peerID: peerID,
+			err:    err,
+		},
+	}
+	bcR.eventsFromFSMCh <- msgData
+}
+
+// Implements bcRNotifier
+func (bcR *BlockchainReactor) resetStateTimer(name string, timer **time.Timer, timeout time.Duration) {
+	if timer == nil {
+		panic("nil timer pointer parameter")
+	}
+	if *timer == nil {
+		*timer = time.AfterFunc(timeout, func() {
+			msg := bcReactorMessage{
+				event: stateTimeoutEv,
+				data: bReactorEventData{
+					stateName: name,
+				},
+			}
+			bcR.errorsForFSMCh <- msg
+		})
+	} else {
+		(*timer).Reset(timeout)
+	}
+}
+
+//-----------------------------------------------------------------------------
+// Messages
+
+// BlockchainMessage is a generic message for this reactor.
+type BlockchainMessage interface {
+	ValidateBasic() error
+}
+
+// RegisterBlockchainMessages registers the fast sync messages for amino encoding.
+func RegisterBlockchainMessages(cdc *amino.Codec) {
+	cdc.RegisterInterface((*BlockchainMessage)(nil), nil)
+	cdc.RegisterConcrete(&bcBlockRequestMessage{}, "tendermint/blockchain/BlockRequest", nil)
+	cdc.RegisterConcrete(&bcBlockResponseMessage{}, "tendermint/blockchain/BlockResponse", nil)
+	cdc.RegisterConcrete(&bcNoBlockResponseMessage{}, "tendermint/blockchain/NoBlockResponse", nil)
+	cdc.RegisterConcrete(&bcStatusResponseMessage{}, "tendermint/blockchain/StatusResponse", nil)
+	cdc.RegisterConcrete(&bcStatusRequestMessage{}, "tendermint/blockchain/StatusRequest", nil)
+}
+
+func decodeMsg(bz []byte) (msg BlockchainMessage, err error) {
+	if len(bz) > maxMsgSize {
+		return msg, fmt.Errorf("msg exceeds max size (%d > %d)", len(bz), maxMsgSize)
+	}
+	err = cdc.UnmarshalBinaryBare(bz, &msg)
+	return
+}
+
+//-------------------------------------
+
+type bcBlockRequestMessage struct {
+	Height int64
+}
+
+// ValidateBasic performs basic validation.
+func (m *bcBlockRequestMessage) ValidateBasic() error {
+	if m.Height < 0 {
+		return errors.New("negative Height")
+	}
+	return nil
+}
+
+func (m *bcBlockRequestMessage) String() string {
+	return fmt.Sprintf("[bcBlockRequestMessage %v]", m.Height)
+}
+
+type bcNoBlockResponseMessage struct {
+	Height int64
+}
+
+// ValidateBasic performs basic validation.
+func (m *bcNoBlockResponseMessage) ValidateBasic() error {
+	if m.Height < 0 {
+		return errors.New("negative Height")
+	}
+	return nil
+}
+
+func (m *bcNoBlockResponseMessage) String() string {
+	return fmt.Sprintf("[bcNoBlockResponseMessage %d]", m.Height)
+}
+
+//-------------------------------------
+
+type bcBlockResponseMessage struct {
+	Block *types.Block
+}
+
+// ValidateBasic performs basic validation.
+func (m *bcBlockResponseMessage) ValidateBasic() error {
+	return m.Block.ValidateBasic()
+}
+
+func (m *bcBlockResponseMessage) String() string {
+	return fmt.Sprintf("[bcBlockResponseMessage %v]", m.Block.Height)
+}
+
+//-------------------------------------
+
+type bcStatusRequestMessage struct {
+	Height int64
+}
+
+// ValidateBasic performs basic validation.
+func (m *bcStatusRequestMessage) ValidateBasic() error {
+	if m.Height < 0 {
+		return errors.New("negative Height")
+	}
+	return nil
+}
+
+func (m *bcStatusRequestMessage) String() string {
+	return fmt.Sprintf("[bcStatusRequestMessage %v]", m.Height)
+}
+
+//-------------------------------------
+
+type bcStatusResponseMessage struct {
+	Height int64
+}
+
+// ValidateBasic performs basic validation.
+func (m *bcStatusResponseMessage) ValidateBasic() error {
+	if m.Height < 0 {
+		return errors.New("negative Height")
+	}
+	return nil
+}
+
+func (m *bcStatusResponseMessage) String() string {
+	return fmt.Sprintf("[bcStatusResponseMessage %v]", m.Height)
+}

blockchain/v1/reactor_fsm.go

@@ -0,0 +1,450 @@
+package v1
+
+import (
+	"errors"
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+// Blockchain Reactor State
+type bcReactorFSMState struct {
+	name string
+
+	// called when transitioning out of current state
+	handle func(*BcReactorFSM, bReactorEvent, bReactorEventData) (next *bcReactorFSMState, err error)
+	// called when entering the state
+	enter func(fsm *BcReactorFSM)
+
+	// timeout to ensure FSM is not stuck in a state forever
+	// the timer is owned and run by the fsm instance
+	timeout time.Duration
+}
+
+func (s *bcReactorFSMState) String() string {
+	return s.name
+}
+
+// BcReactorFSM is the datastructure for the Blockchain Reactor State Machine
+type BcReactorFSM struct {
+	logger log.Logger
+	mtx    sync.Mutex
+
+	startTime time.Time
+
+	state      *bcReactorFSMState
+	stateTimer *time.Timer
+	pool       *BlockPool
+
+	// interface used to call the Blockchain reactor to send StatusRequest, BlockRequest, reporting errors, etc.
+	toBcR bcReactor
+}
+
+// NewFSM creates a new reactor FSM.
+func NewFSM(height int64, toBcR bcReactor) *BcReactorFSM {
+	return &BcReactorFSM{
+		state:     unknown,
+		startTime: time.Now(),
+		pool:      NewBlockPool(height, toBcR),
+		toBcR:     toBcR,
+	}
+}
+
+// bReactorEventData is part of the message sent by the reactor to the FSM and used by the state handlers.
+type bReactorEventData struct {
+	peerID         p2p.ID
+	err            error        // for peer error: timeout, slow; for processed block event if error occurred
+	height         int64        // for status response; for processed block event
+	block          *types.Block // for block response
+	stateName      string       // for state timeout events
+	length         int          // for block response event, length of received block, used to detect slow peers
+	maxNumRequests int          // for request needed event, maximum number of pending requests
+}
+
+// Blockchain Reactor Events (the input to the state machine)
+type bReactorEvent uint
+
+const (
+	// message type events
+	startFSMEv = iota + 1
+	statusResponseEv
+	blockResponseEv
+	processedBlockEv
+	makeRequestsEv
+	stopFSMEv
+
+	// other events
+	peerRemoveEv = iota + 256
+	stateTimeoutEv
+)
+
+func (msg *bcReactorMessage) String() string {
+	var dataStr string
+
+	switch msg.event {
+	case startFSMEv:
+		dataStr = ""
+	case statusResponseEv:
+		dataStr = fmt.Sprintf("peer=%v height=%v", msg.data.peerID, msg.data.height)
+	case blockResponseEv:
+		dataStr = fmt.Sprintf("peer=%v block.height=%v length=%v",
+			msg.data.peerID, msg.data.block.Height, msg.data.length)
+	case processedBlockEv:
+		dataStr = fmt.Sprintf("error=%v", msg.data.err)
+	case makeRequestsEv:
+		dataStr = ""
+	case stopFSMEv:
+		dataStr = ""
+	case peerRemoveEv:
+		dataStr = fmt.Sprintf("peer: %v is being removed by the switch", msg.data.peerID)
+	case stateTimeoutEv:
+		dataStr = fmt.Sprintf("state=%v", msg.data.stateName)
+	default:
+		dataStr = fmt.Sprintf("cannot interpret message data")
+	}
+
+	return fmt.Sprintf("%v: %v", msg.event, dataStr)
+}
+
+func (ev bReactorEvent) String() string {
+	switch ev {
+	case startFSMEv:
+		return "startFSMEv"
+	case statusResponseEv:
+		return "statusResponseEv"
+	case blockResponseEv:
+		return "blockResponseEv"
+	case processedBlockEv:
+		return "processedBlockEv"
+	case makeRequestsEv:
+		return "makeRequestsEv"
+	case stopFSMEv:
+		return "stopFSMEv"
+	case peerRemoveEv:
+		return "peerRemoveEv"
+	case stateTimeoutEv:
+		return "stateTimeoutEv"
+	default:
+		return "event unknown"
+	}
+
+}
+
+// states
+var (
+	unknown      *bcReactorFSMState
+	waitForPeer  *bcReactorFSMState
+	waitForBlock *bcReactorFSMState
+	finished     *bcReactorFSMState
+)
+
+// timeouts for state timers
+const (
+	waitForPeerTimeout                 = 3 * time.Second
+	waitForBlockAtCurrentHeightTimeout = 10 * time.Second
+)
+
+// errors
+var (
+	// internal to the package
+	errNoErrorFinished        = errors.New("fast sync is finished")
+	errInvalidEvent           = errors.New("invalid event in current state")
+	errMissingBlock           = errors.New("missing blocks")
+	errNilPeerForBlockRequest = errors.New("peer for block request does not exist in the switch")
+	errSendQueueFull          = errors.New("block request not made, send-queue is full")
+	errPeerTooShort           = errors.New("peer height too low, old peer removed/ new peer not added")
+	errSwitchRemovesPeer      = errors.New("switch is removing peer")
+	errTimeoutEventWrongState = errors.New("timeout event for a state different than the current one")
+	errNoTallerPeer           = errors.New("fast sync timed out on waiting for a peer taller than this node")
+
+	// reported eventually to the switch
+	errPeerLowersItsHeight             = errors.New("fast sync peer reports a height lower than previous")            // handle return
+	errNoPeerResponseForCurrentHeights = errors.New("fast sync timed out on peer block response for current heights") // handle return
+	errNoPeerResponse                  = errors.New("fast sync timed out on peer block response")                     // xx
+	errBadDataFromPeer                 = errors.New("fast sync received block from wrong peer or block is bad")       // xx
+	errDuplicateBlock                  = errors.New("fast sync received duplicate block from peer")
+	errBlockVerificationFailure        = errors.New("fast sync block verification failure")              // xx
+	errSlowPeer                        = errors.New("fast sync peer is not sending us data fast enough") // xx
+
+)
+
+func init() {
+	unknown = &bcReactorFSMState{
+		name: "unknown",
+		handle: func(fsm *BcReactorFSM, ev bReactorEvent, data bReactorEventData) (*bcReactorFSMState, error) {
+			switch ev {
+			case startFSMEv:
+				// Broadcast Status message. Currently doesn't return non-nil error.
+				fsm.toBcR.sendStatusRequest()
+				return waitForPeer, nil
+
+			case stopFSMEv:
+				return finished, errNoErrorFinished
+
+			default:
+				return unknown, errInvalidEvent
+			}
+		},
+	}
+
+	waitForPeer = &bcReactorFSMState{
+		name:    "waitForPeer",
+		timeout: waitForPeerTimeout,
+		enter: func(fsm *BcReactorFSM) {
+			// Stop when leaving the state.
+			fsm.resetStateTimer()
+		},
+		handle: func(fsm *BcReactorFSM, ev bReactorEvent, data bReactorEventData) (*bcReactorFSMState, error) {
+			switch ev {
+			case stateTimeoutEv:
+				if data.stateName != "waitForPeer" {
+					fsm.logger.Error("received a state timeout event for different state",
+						"state", data.stateName)
+					return waitForPeer, errTimeoutEventWrongState
+				}
+				// There was no statusResponse received from any peer.
+				// Should we send status request again?
+				return finished, errNoTallerPeer
+
+			case statusResponseEv:
+				if err := fsm.pool.UpdatePeer(data.peerID, data.height); err != nil {
+					if fsm.pool.NumPeers() == 0 {
+						return waitForPeer, err
+					}
+				}
+				if fsm.stateTimer != nil {
+					fsm.stateTimer.Stop()
+				}
+				return waitForBlock, nil
+
+			case stopFSMEv:
+				if fsm.stateTimer != nil {
+					fsm.stateTimer.Stop()
+				}
+				return finished, errNoErrorFinished
+
+			default:
+				return waitForPeer, errInvalidEvent
+			}
+		},
+	}
+
+	waitForBlock = &bcReactorFSMState{
+		name:    "waitForBlock",
+		timeout: waitForBlockAtCurrentHeightTimeout,
+		enter: func(fsm *BcReactorFSM) {
+			// Stop when leaving the state.
+			fsm.resetStateTimer()
+		},
+		handle: func(fsm *BcReactorFSM, ev bReactorEvent, data bReactorEventData) (*bcReactorFSMState, error) {
+			switch ev {
+
+			case statusResponseEv:
+				err := fsm.pool.UpdatePeer(data.peerID, data.height)
+				if fsm.pool.NumPeers() == 0 {
+					return waitForPeer, err
+				}
+				if fsm.pool.ReachedMaxHeight() {
+					return finished, err
+				}
+				return waitForBlock, err
+
+			case blockResponseEv:
+				fsm.logger.Debug("blockResponseEv", "H", data.block.Height)
+				err := fsm.pool.AddBlock(data.peerID, data.block, data.length)
+				if err != nil {
+					// A block was received that was unsolicited, from unexpected peer, or that we already have it.
+					// Ignore block, remove peer and send error to switch.
+					fsm.pool.RemovePeer(data.peerID, err)
+					fsm.toBcR.sendPeerError(err, data.peerID)
+				}
+				if fsm.pool.NumPeers() == 0 {
+					return waitForPeer, err
+				}
+				return waitForBlock, err
+
+			case processedBlockEv:
+				if data.err != nil {
+					first, second, _ := fsm.pool.FirstTwoBlocksAndPeers()
+					fsm.logger.Error("error processing block", "err", data.err,
+						"first", first.block.Height, "second", second.block.Height)
+					fsm.logger.Error("send peer error for", "peer", first.peer.ID)
+					fsm.toBcR.sendPeerError(data.err, first.peer.ID)
+					fsm.logger.Error("send peer error for", "peer", second.peer.ID)
+					fsm.toBcR.sendPeerError(data.err, second.peer.ID)
+					// Remove the first two blocks. This will also remove the peers
+					fsm.pool.InvalidateFirstTwoBlocks(data.err)
+				} else {
+					fsm.pool.ProcessedCurrentHeightBlock()
+					// Since we advanced one block reset the state timer
+					fsm.resetStateTimer()
+				}
+
+				// Both cases above may result in achieving maximum height.
+				if fsm.pool.ReachedMaxHeight() {
+					return finished, nil
+				}
+
+				return waitForBlock, data.err
+
+			case peerRemoveEv:
+				// This event is sent by the switch to remove disconnected and errored peers.
+				fsm.pool.RemovePeer(data.peerID, data.err)
+				if fsm.pool.NumPeers() == 0 {
+					return waitForPeer, nil
+				}
+				if fsm.pool.ReachedMaxHeight() {
+					return finished, nil
+				}
+				return waitForBlock, nil
+
+			case makeRequestsEv:
+				fsm.makeNextRequests(data.maxNumRequests)
+				return waitForBlock, nil
+
+			case stateTimeoutEv:
+				if data.stateName != "waitForBlock" {
+					fsm.logger.Error("received a state timeout event for different state",
+						"state", data.stateName)
+					return waitForBlock, errTimeoutEventWrongState
+				}
+				// We haven't received the block at current height or height+1. Remove peer.
+				fsm.pool.RemovePeerAtCurrentHeights(errNoPeerResponseForCurrentHeights)
+				fsm.resetStateTimer()
+				if fsm.pool.NumPeers() == 0 {
+					return waitForPeer, errNoPeerResponseForCurrentHeights
+				}
+				if fsm.pool.ReachedMaxHeight() {
+					return finished, nil
+				}
+				return waitForBlock, errNoPeerResponseForCurrentHeights
+
+			case stopFSMEv:
+				if fsm.stateTimer != nil {
+					fsm.stateTimer.Stop()
+				}
+				return finished, errNoErrorFinished
+
+			default:
+				return waitForBlock, errInvalidEvent
+			}
+		},
+	}
+
+	finished = &bcReactorFSMState{
+		name: "finished",
+		enter: func(fsm *BcReactorFSM) {
+			fsm.logger.Info("Time to switch to consensus reactor!", "height", fsm.pool.Height)
+			fsm.toBcR.switchToConsensus()
+			fsm.cleanup()
+		},
+		handle: func(fsm *BcReactorFSM, ev bReactorEvent, data bReactorEventData) (*bcReactorFSMState, error) {
+			return finished, nil
+		},
+	}
+}
+
+// Interface used by FSM for sending Block and Status requests,
+// informing of peer errors and state timeouts
+// Implemented by BlockchainReactor and tests
+type bcReactor interface {
+	sendStatusRequest()
+	sendBlockRequest(peerID p2p.ID, height int64) error
+	sendPeerError(err error, peerID p2p.ID)
+	resetStateTimer(name string, timer **time.Timer, timeout time.Duration)
+	switchToConsensus()
+}
+
+// SetLogger sets the FSM logger.
+func (fsm *BcReactorFSM) SetLogger(l log.Logger) {
+	fsm.logger = l
+	fsm.pool.SetLogger(l)
+}
+
+// Start starts the FSM.
+func (fsm *BcReactorFSM) Start() {
+	_ = fsm.Handle(&bcReactorMessage{event: startFSMEv})
+}
+
+// Handle processes messages and events sent to the FSM.
+func (fsm *BcReactorFSM) Handle(msg *bcReactorMessage) error {
+	fsm.mtx.Lock()
+	defer fsm.mtx.Unlock()
+	fsm.logger.Debug("FSM received", "event", msg, "state", fsm.state)
+
+	if fsm.state == nil {
+		fsm.state = unknown
+	}
+	next, err := fsm.state.handle(fsm, msg.event, msg.data)
+	if err != nil {
+		fsm.logger.Error("FSM event handler returned", "err", err,
+			"state", fsm.state, "event", msg.event)
+	}
+
+	oldState := fsm.state.name
+	fsm.transition(next)
+	if oldState != fsm.state.name {
+		fsm.logger.Info("FSM changed state", "new_state", fsm.state)
+	}
+	return err
+}
+
+func (fsm *BcReactorFSM) transition(next *bcReactorFSMState) {
+	if next == nil {
+		return
+	}
+	if fsm.state != next {
+		fsm.state = next
+		if next.enter != nil {
+			next.enter(fsm)
+		}
+	}
+}
+
+// Called when entering an FSM state in order to detect lack of progress in the state machine.
+// Note the use of the 'bcr' interface to facilitate testing without timer expiring.
+func (fsm *BcReactorFSM) resetStateTimer() {
+	fsm.toBcR.resetStateTimer(fsm.state.name, &fsm.stateTimer, fsm.state.timeout)
+}
+
+func (fsm *BcReactorFSM) isCaughtUp() bool {
+	return fsm.state == finished
+}
+
+func (fsm *BcReactorFSM) makeNextRequests(maxNumRequests int) {
+	fsm.pool.MakeNextRequests(maxNumRequests)
+}
+
+func (fsm *BcReactorFSM) cleanup() {
+	fsm.pool.Cleanup()
+}
+
+// NeedsBlocks checks if more block requests are required.
+func (fsm *BcReactorFSM) NeedsBlocks() bool {
+	fsm.mtx.Lock()
+	defer fsm.mtx.Unlock()
+	return fsm.state.name == "waitForBlock" && fsm.pool.NeedsBlocks()
+}
+
+// FirstTwoBlocks returns the two blocks at pool height and height+1
+func (fsm *BcReactorFSM) FirstTwoBlocks() (first, second *types.Block, err error) {
+	fsm.mtx.Lock()
+	defer fsm.mtx.Unlock()
+	firstBP, secondBP, err := fsm.pool.FirstTwoBlocksAndPeers()
+	if err == nil {
+		first = firstBP.block
+		second = secondBP.block
+	}
+	return
+}
+
+// Status returns the pool's height and the maximum peer height.
+func (fsm *BcReactorFSM) Status() (height, maxPeerHeight int64) {
+	fsm.mtx.Lock()
+	defer fsm.mtx.Unlock()
+	return fsm.pool.Height, fsm.pool.MaxPeerHeight
+}

blockchain/v1/reactor_fsm_test.go

@@ -0,0 +1,938 @@
+package v1
+
+import (
+	"fmt"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	cmn "github.com/tendermint/tendermint/libs/common"
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/types"
+)
+
+type lastBlockRequestT struct {
+	peerID p2p.ID
+	height int64
+}
+
+type lastPeerErrorT struct {
+	peerID p2p.ID
+	err    error
+}
+
+// reactor for FSM testing
+type testReactor struct {
+	logger            log.Logger
+	fsm               *BcReactorFSM
+	numStatusRequests int
+	numBlockRequests  int
+	lastBlockRequest  lastBlockRequestT
+	lastPeerError     lastPeerErrorT
+	stateTimerStarts  map[string]int
+}
+
+func sendEventToFSM(fsm *BcReactorFSM, ev bReactorEvent, data bReactorEventData) error {
+	return fsm.Handle(&bcReactorMessage{event: ev, data: data})
+}
+
+type fsmStepTestValues struct {
+	currentState string
+	event        bReactorEvent
+	data         bReactorEventData
+
+	wantErr           error
+	wantState         string
+	wantStatusReqSent bool
+	wantReqIncreased  bool
+	wantNewBlocks     []int64
+	wantRemovedPeers  []p2p.ID
+}
+
+// ---------------------------------------------------------------------------
+// helper test function for different FSM events, state and expected behavior
+func sStopFSMEv(current, expected string) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        stopFSMEv,
+		wantState:    expected,
+		wantErr:      errNoErrorFinished}
+}
+
+func sUnknownFSMEv(current string) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        1234,
+		wantState:    current,
+		wantErr:      errInvalidEvent}
+}
+
+func sStartFSMEv() fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState:      "unknown",
+		event:             startFSMEv,
+		wantState:         "waitForPeer",
+		wantStatusReqSent: true}
+}
+
+func sStateTimeoutEv(current, expected string, timedoutState string, wantErr error) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        stateTimeoutEv,
+		data: bReactorEventData{
+			stateName: timedoutState,
+		},
+		wantState: expected,
+		wantErr:   wantErr,
+	}
+}
+
+func sProcessedBlockEv(current, expected string, reactorError error) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        processedBlockEv,
+		data: bReactorEventData{
+			err: reactorError,
+		},
+		wantState: expected,
+		wantErr:   reactorError,
+	}
+}
+
+func sStatusEv(current, expected string, peerID p2p.ID, height int64, err error) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        statusResponseEv,
+		data:         bReactorEventData{peerID: peerID, height: height},
+		wantState:    expected,
+		wantErr:      err}
+}
+
+func sMakeRequestsEv(current, expected string, maxPendingRequests int) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState:     current,
+		event:            makeRequestsEv,
+		data:             bReactorEventData{maxNumRequests: maxPendingRequests},
+		wantState:        expected,
+		wantReqIncreased: true,
+	}
+}
+
+func sMakeRequestsEvErrored(current, expected string,
+	maxPendingRequests int, err error, peersRemoved []p2p.ID) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState:     current,
+		event:            makeRequestsEv,
+		data:             bReactorEventData{maxNumRequests: maxPendingRequests},
+		wantState:        expected,
+		wantErr:          err,
+		wantRemovedPeers: peersRemoved,
+		wantReqIncreased: true,
+	}
+}
+
+func sBlockRespEv(current, expected string, peerID p2p.ID, height int64, prevBlocks []int64) fsmStepTestValues {
+	txs := []types.Tx{types.Tx("foo"), types.Tx("bar")}
+	return fsmStepTestValues{
+		currentState: current,
+		event:        blockResponseEv,
+		data: bReactorEventData{
+			peerID: peerID,
+			height: height,
+			block:  types.MakeBlock(int64(height), txs, nil, nil),
+			length: 100},
+		wantState:     expected,
+		wantNewBlocks: append(prevBlocks, height),
+	}
+}
+
+func sBlockRespEvErrored(current, expected string,
+	peerID p2p.ID, height int64, prevBlocks []int64, wantErr error, peersRemoved []p2p.ID) fsmStepTestValues {
+	txs := []types.Tx{types.Tx("foo"), types.Tx("bar")}
+
+	return fsmStepTestValues{
+		currentState: current,
+		event:        blockResponseEv,
+		data: bReactorEventData{
+			peerID: peerID,
+			height: height,
+			block:  types.MakeBlock(int64(height), txs, nil, nil),
+			length: 100},
+		wantState:        expected,
+		wantErr:          wantErr,
+		wantRemovedPeers: peersRemoved,
+		wantNewBlocks:    prevBlocks,
+	}
+}
+
+func sPeerRemoveEv(current, expected string, peerID p2p.ID, err error, peersRemoved []p2p.ID) fsmStepTestValues {
+	return fsmStepTestValues{
+		currentState: current,
+		event:        peerRemoveEv,
+		data: bReactorEventData{
+			peerID: peerID,
+			err:    err,
+		},
+		wantState:        expected,
+		wantRemovedPeers: peersRemoved,
+	}
+}
+
+// --------------------------------------------
+
+func newTestReactor(height int64) *testReactor {
+	testBcR := &testReactor{logger: log.TestingLogger(), stateTimerStarts: make(map[string]int)}
+	testBcR.fsm = NewFSM(height, testBcR)
+	testBcR.fsm.SetLogger(testBcR.logger)
+	return testBcR
+}
+
+func fixBlockResponseEvStep(step *fsmStepTestValues, testBcR *testReactor) {
+	// There is currently no good way to know to which peer a block request was sent.
+	// So in some cases where it does not matter, before we simulate a block response
+	// we cheat and look where it is expected from.
+	if step.event == blockResponseEv {
+		height := step.data.height
+		peerID, ok := testBcR.fsm.pool.blocks[height]
+		if ok {
+			step.data.peerID = peerID
+		}
+	}
+}
+
+type testFields struct {
+	name               string
+	startingHeight     int64
+	maxRequestsPerPeer int
+	maxPendingRequests int
+	steps              []fsmStepTestValues
+}
+
+func executeFSMTests(t *testing.T, tests []testFields, matchRespToReq bool) {
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// Create test reactor
+			testBcR := newTestReactor(tt.startingHeight)
+
+			if tt.maxRequestsPerPeer != 0 {
+				maxRequestsPerPeer = tt.maxRequestsPerPeer
+			}
+
+			for _, step := range tt.steps {
+				assert.Equal(t, step.currentState, testBcR.fsm.state.name)
+
+				var heightBefore int64
+				if step.event == processedBlockEv && step.data.err == errBlockVerificationFailure {
+					heightBefore = testBcR.fsm.pool.Height
+				}
+				oldNumStatusRequests := testBcR.numStatusRequests
+				oldNumBlockRequests := testBcR.numBlockRequests
+				if matchRespToReq {
+					fixBlockResponseEvStep(&step, testBcR)
+				}
+
+				fsmErr := sendEventToFSM(testBcR.fsm, step.event, step.data)
+				assert.Equal(t, step.wantErr, fsmErr)
+
+				if step.wantStatusReqSent {
+					assert.Equal(t, oldNumStatusRequests+1, testBcR.numStatusRequests)
+				} else {
+					assert.Equal(t, oldNumStatusRequests, testBcR.numStatusRequests)
+				}
+
+				if step.wantReqIncreased {
+					assert.True(t, oldNumBlockRequests < testBcR.numBlockRequests)
+				} else {
+					assert.Equal(t, oldNumBlockRequests, testBcR.numBlockRequests)
+				}
+
+				for _, height := range step.wantNewBlocks {
+					_, err := testBcR.fsm.pool.BlockAndPeerAtHeight(height)
+					assert.Nil(t, err)
+				}
+				if step.event == processedBlockEv && step.data.err == errBlockVerificationFailure {
+					heightAfter := testBcR.fsm.pool.Height
+					assert.Equal(t, heightBefore, heightAfter)
+					firstAfter, err1 := testBcR.fsm.pool.BlockAndPeerAtHeight(testBcR.fsm.pool.Height)
+					secondAfter, err2 := testBcR.fsm.pool.BlockAndPeerAtHeight(testBcR.fsm.pool.Height + 1)
+					assert.NotNil(t, err1)
+					assert.NotNil(t, err2)
+					assert.Nil(t, firstAfter)
+					assert.Nil(t, secondAfter)
+				}
+
+				assert.Equal(t, step.wantState, testBcR.fsm.state.name)
+
+				if step.wantState == "finished" {
+					assert.True(t, testBcR.fsm.isCaughtUp())
+				}
+			}
+		})
+	}
+}
+
+func TestFSMBasic(t *testing.T) {
+	tests := []testFields{
+		{
+			name:               "one block, one peer - TS2",
+			startingHeight:     1,
+			maxRequestsPerPeer: 2,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 2, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 2, []int64{1}),
+				sProcessedBlockEv("waitForBlock", "finished", nil),
+			},
+		},
+		{
+			name:               "multi block, multi peer - TS2",
+			startingHeight:     1,
+			maxRequestsPerPeer: 2,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 4, nil),
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 4, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 2, []int64{1}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 3, []int64{1, 2}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 4, []int64{1, 2, 3}),
+
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+				sProcessedBlockEv("waitForBlock", "finished", nil),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, true)
+}
+
+func TestFSMBlockVerificationFailure(t *testing.T) {
+	tests := []testFields{
+		{
+			name:               "block verification failure - TS2 variant",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+
+				// add P1 and get blocks 1-3 from it
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 2, []int64{1}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 3, []int64{1, 2}),
+
+				// add P2
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, nil),
+
+				// process block failure, should remove P1 and all blocks
+				sProcessedBlockEv("waitForBlock", "waitForBlock", errBlockVerificationFailure),
+
+				// get blocks 1-3 from P2
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 2, []int64{1}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 3, []int64{1, 2}),
+
+				// finish after processing blocks 1 and 2
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+				sProcessedBlockEv("waitForBlock", "finished", nil),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, false)
+}
+
+func TestFSMBadBlockFromPeer(t *testing.T) {
+	tests := []testFields{
+		{
+			name:               "block we haven't asked for",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1 and ask for blocks 1-3
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 300, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				// blockResponseEv for height 100 should cause an error
+				sBlockRespEvErrored("waitForBlock", "waitForPeer",
+					"P1", 100, []int64{}, errMissingBlock, []p2p.ID{}),
+			},
+		},
+		{
+			name:               "block we already have",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1 and get block 1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 100, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock",
+					"P1", 1, []int64{}),
+
+				// Get block 1 again. Since peer is removed together with block 1,
+				// the blocks present in the pool should be {}
+				sBlockRespEvErrored("waitForBlock", "waitForPeer",
+					"P1", 1, []int64{}, errDuplicateBlock, []p2p.ID{"P1"}),
+			},
+		},
+		{
+			name:               "block from unknown peer",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1 and get block 1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+
+				// get block 1 from unknown peer P2
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEvErrored("waitForBlock", "waitForBlock",
+					"P2", 1, []int64{}, errBadDataFromPeer, []p2p.ID{"P2"}),
+			},
+		},
+		{
+			name:               "block from wrong peer",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1, make requests for blocks 1-3 to P1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				// add P2
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, nil),
+
+				// receive block 1 from P2
+				sBlockRespEvErrored("waitForBlock", "waitForBlock",
+					"P2", 1, []int64{}, errBadDataFromPeer, []p2p.ID{"P2"}),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, false)
+}
+
+func TestFSMBlockAtCurrentHeightDoesNotArriveInTime(t *testing.T) {
+	tests := []testFields{
+		{
+			name:               "block at current height undelivered - TS5",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1, get blocks 1 and 2, process block 1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock",
+					"P1", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock",
+					"P1", 2, []int64{1}),
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+
+				// add P2
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, nil),
+
+				// timeout on block 3, P1 should be removed
+				sStateTimeoutEv("waitForBlock", "waitForBlock", "waitForBlock", errNoPeerResponseForCurrentHeights),
+
+				// make requests and finish by receiving blocks 2 and 3 from P2
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 2, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P2", 3, []int64{2}),
+				sProcessedBlockEv("waitForBlock", "finished", nil),
+			},
+		},
+		{
+			name:               "block at current height undelivered, at maxPeerHeight after peer removal - TS3",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1, request blocks 1-3 from P1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				// add P2 (tallest)
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 30, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				// receive blocks 1-3 from P1
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 1, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 2, []int64{1}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 3, []int64{1, 2}),
+
+				// process blocks at heights 1 and 2
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+
+				// timeout on block at height 4
+				sStateTimeoutEv("waitForBlock", "finished", "waitForBlock", nil),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, true)
+}
+
+func TestFSMPeerRelatedEvents(t *testing.T) {
+	tests := []testFields{
+		{
+			name:           "peer remove event with no blocks",
+			startingHeight: 1,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1, P2, P3
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, nil),
+				sStatusEv("waitForBlock", "waitForBlock", "P3", 3, nil),
+
+				// switch removes P2
+				sPeerRemoveEv("waitForBlock", "waitForBlock", "P2", errSwitchRemovesPeer, []p2p.ID{"P2"}),
+			},
+		},
+		{
+			name:           "only peer removed while in waitForBlock state",
+			startingHeight: 100,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 200, nil),
+
+				// switch removes P1
+				sPeerRemoveEv("waitForBlock", "waitForPeer", "P1", errSwitchRemovesPeer, []p2p.ID{"P1"}),
+			},
+		},
+		{
+			name:               "highest peer removed while in waitForBlock state, node reaches maxPeerHeight - TS4 ",
+			startingHeight:     100,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1 and make requests
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 101, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				// add P2
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 200, nil),
+
+				// get blocks 100 and 101 from P1 and process block at height 100
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 100, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 101, []int64{100}),
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+
+				// switch removes peer P1, should be finished
+				sPeerRemoveEv("waitForBlock", "finished", "P2", errSwitchRemovesPeer, []p2p.ID{"P2"}),
+			},
+		},
+		{
+			name:               "highest peer lowers its height in waitForBlock state, node reaches maxPeerHeight - TS4",
+			startingHeight:     100,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1 and make requests
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 101, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+
+				// add P2
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 200, nil),
+
+				// get blocks 100 and 101 from P1
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 100, []int64{}),
+				sBlockRespEv("waitForBlock", "waitForBlock", "P1", 101, []int64{100}),
+
+				// processed block at heights 100
+				sProcessedBlockEv("waitForBlock", "waitForBlock", nil),
+
+				// P2 becomes short
+				sStatusEv("waitForBlock", "finished", "P2", 100, errPeerLowersItsHeight),
+			},
+		},
+		{
+			name:           "new short peer while in waitForPeer state",
+			startingHeight: 100,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForPeer", "P1", 3, errPeerTooShort),
+			},
+		},
+		{
+			name:           "new short peer while in waitForBlock state",
+			startingHeight: 100,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 200, nil),
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, errPeerTooShort),
+			},
+		},
+		{
+			name:           "only peer updated with low height while in waitForBlock state",
+			startingHeight: 100,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 200, nil),
+				sStatusEv("waitForBlock", "waitForPeer", "P1", 3, errPeerLowersItsHeight),
+			},
+		},
+		{
+			name:               "peer does not exist in the switch",
+			startingHeight:     9999999,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				// add P1
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 20000000, nil),
+				// send request for block 9999999
+				// Note: For this block request the "switch missing the peer" error is simulated,
+				// see implementation of bcReactor interface, sendBlockRequest(), in this file.
+				sMakeRequestsEvErrored("waitForBlock", "waitForBlock",
+					maxNumRequests, nil, []p2p.ID{"P1"}),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, true)
+}
+
+func TestFSMStopFSM(t *testing.T) {
+	tests := []testFields{
+		{
+			name: "stopFSMEv in unknown",
+			steps: []fsmStepTestValues{
+				sStopFSMEv("unknown", "finished"),
+			},
+		},
+		{
+			name:           "stopFSMEv in waitForPeer",
+			startingHeight: 1,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStopFSMEv("waitForPeer", "finished"),
+			},
+		},
+		{
+			name:           "stopFSMEv in waitForBlock",
+			startingHeight: 1,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sStopFSMEv("waitForBlock", "finished"),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, false)
+}
+
+func TestFSMUnknownElements(t *testing.T) {
+	tests := []testFields{
+		{
+			name: "unknown event for state unknown",
+			steps: []fsmStepTestValues{
+				sUnknownFSMEv("unknown"),
+			},
+		},
+		{
+			name: "unknown event for state waitForPeer",
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sUnknownFSMEv("waitForPeer"),
+			},
+		},
+		{
+			name:           "unknown event for state waitForBlock",
+			startingHeight: 1,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sUnknownFSMEv("waitForBlock"),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, false)
+}
+
+func TestFSMPeerStateTimeoutEvent(t *testing.T) {
+	tests := []testFields{
+		{
+			name:               "timeout event for state waitForPeer while in state waitForPeer - TS1",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStateTimeoutEv("waitForPeer", "finished", "waitForPeer", errNoTallerPeer),
+			},
+		},
+		{
+			name:               "timeout event for state waitForPeer while in a state != waitForPeer",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStateTimeoutEv("waitForPeer", "waitForPeer", "waitForBlock", errTimeoutEventWrongState),
+			},
+		},
+		{
+			name:               "timeout event for state waitForBlock while in state waitForBlock ",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sStateTimeoutEv("waitForBlock", "waitForPeer", "waitForBlock", errNoPeerResponseForCurrentHeights),
+			},
+		},
+		{
+			name:               "timeout event for state waitForBlock while in a state != waitForBlock",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sStateTimeoutEv("waitForBlock", "waitForBlock", "waitForPeer", errTimeoutEventWrongState),
+			},
+		},
+		{
+			name:               "timeout event for state waitForBlock with multiple peers",
+			startingHeight:     1,
+			maxRequestsPerPeer: 3,
+			steps: []fsmStepTestValues{
+				sStartFSMEv(),
+				sStatusEv("waitForPeer", "waitForBlock", "P1", 3, nil),
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+				sStatusEv("waitForBlock", "waitForBlock", "P2", 3, nil),
+				sStateTimeoutEv("waitForBlock", "waitForBlock", "waitForBlock", errNoPeerResponseForCurrentHeights),
+			},
+		},
+	}
+
+	executeFSMTests(t, tests, false)
+}
+
+func makeCorrectTransitionSequence(startingHeight int64, numBlocks int64, numPeers int, randomPeerHeights bool,
+	maxRequestsPerPeer int, maxPendingRequests int) testFields {
+
+	// Generate numPeers peers with random or numBlocks heights according to the randomPeerHeights flag.
+	peerHeights := make([]int64, numPeers)
+	for i := 0; i < numPeers; i++ {
+		if i == 0 {
+			peerHeights[0] = numBlocks
+			continue
+		}
+		if randomPeerHeights {
+			peerHeights[i] = int64(cmn.MaxInt(cmn.RandIntn(int(numBlocks)), int(startingHeight)+1))
+		} else {
+			peerHeights[i] = numBlocks
+		}
+	}
+
+	// Approximate the slice capacity to save time for appends.
+	testSteps := make([]fsmStepTestValues, 0, 3*numBlocks+int64(numPeers))
+
+	testName := fmt.Sprintf("%v-blocks %v-startingHeight %v-peers %v-maxRequestsPerPeer %v-maxNumRequests",
+		numBlocks, startingHeight, numPeers, maxRequestsPerPeer, maxPendingRequests)
+
+	// Add startFSMEv step.
+	testSteps = append(testSteps, sStartFSMEv())
+
+	// For each peer, add statusResponseEv step.
+	for i := 0; i < numPeers; i++ {
+		peerName := fmt.Sprintf("P%d", i)
+		if i == 0 {
+			testSteps = append(
+				testSteps,
+				sStatusEv("waitForPeer", "waitForBlock", p2p.ID(peerName), peerHeights[i], nil))
+		} else {
+			testSteps = append(testSteps,
+				sStatusEv("waitForBlock", "waitForBlock", p2p.ID(peerName), peerHeights[i], nil))
+		}
+	}
+
+	height := startingHeight
+	numBlocksReceived := 0
+	prevBlocks := make([]int64, 0, maxPendingRequests)
+
+forLoop:
+	for i := 0; i < int(numBlocks); i++ {
+
+		// Add the makeRequestEv step periodically.
+		if i%int(maxRequestsPerPeer) == 0 {
+			testSteps = append(
+				testSteps,
+				sMakeRequestsEv("waitForBlock", "waitForBlock", maxNumRequests),
+			)
+		}
+
+		// Add the blockRespEv step
+		testSteps = append(
+			testSteps,
+			sBlockRespEv("waitForBlock", "waitForBlock",
+				"P0", height, prevBlocks))
+		prevBlocks = append(prevBlocks, height)
+		height++
+		numBlocksReceived++
+
+		// Add the processedBlockEv step periodically.
+		if numBlocksReceived >= int(maxRequestsPerPeer) || height >= numBlocks {
+			for j := int(height) - numBlocksReceived; j < int(height); j++ {
+				if j >= int(numBlocks) {
+					// This is the last block that is processed, we should be in "finished" state.
+					testSteps = append(
+						testSteps,
+						sProcessedBlockEv("waitForBlock", "finished", nil))
+					break forLoop
+				}
+				testSteps = append(
+					testSteps,
+					sProcessedBlockEv("waitForBlock", "waitForBlock", nil))
+			}
+			numBlocksReceived = 0
+			prevBlocks = make([]int64, 0, maxPendingRequests)
+		}
+	}
+
+	return testFields{
+		name:               testName,
+		startingHeight:     startingHeight,
+		maxRequestsPerPeer: maxRequestsPerPeer,
+		maxPendingRequests: maxPendingRequests,
+		steps:              testSteps,
+	}
+}
+
+const (
+	maxStartingHeightTest       = 100
+	maxRequestsPerPeerTest      = 20
+	maxTotalPendingRequestsTest = 600
+	maxNumPeersTest             = 1000
+	maxNumBlocksInChainTest     = 10000 //should be smaller than 9999999
+)
+
+func makeCorrectTransitionSequenceWithRandomParameters() testFields {
+	// Generate a starting height for fast sync.
+	startingHeight := int64(cmn.RandIntn(maxStartingHeightTest) + 1)
+
+	// Generate the number of requests per peer.
+	maxRequestsPerPeer := cmn.RandIntn(maxRequestsPerPeerTest) + 1
+
+	// Generate the maximum number of total pending requests, >= maxRequestsPerPeer.
+	maxPendingRequests := cmn.RandIntn(maxTotalPendingRequestsTest-int(maxRequestsPerPeer)) + maxRequestsPerPeer
+
+	// Generate the number of blocks to be synced.
+	numBlocks := int64(cmn.RandIntn(maxNumBlocksInChainTest)) + startingHeight
+
+	// Generate a number of peers.
+	numPeers := cmn.RandIntn(maxNumPeersTest) + 1
+
+	return makeCorrectTransitionSequence(startingHeight, numBlocks, numPeers, true, maxRequestsPerPeer, maxPendingRequests)
+}
+
+func shouldApplyProcessedBlockEvStep(step *fsmStepTestValues, testBcR *testReactor) bool {
+	if step.event == processedBlockEv {
+		_, err := testBcR.fsm.pool.BlockAndPeerAtHeight(testBcR.fsm.pool.Height)
+		if err == errMissingBlock {
+			return false
+		}
+		_, err = testBcR.fsm.pool.BlockAndPeerAtHeight(testBcR.fsm.pool.Height + 1)
+		if err == errMissingBlock {
+			return false
+		}
+	}
+	return true
+}
+
+func TestFSMCorrectTransitionSequences(t *testing.T) {
+
+	tests := []testFields{
+		makeCorrectTransitionSequence(1, 100, 10, true, 10, 40),
+		makeCorrectTransitionSequenceWithRandomParameters(),
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// Create test reactor
+			testBcR := newTestReactor(tt.startingHeight)
+
+			if tt.maxRequestsPerPeer != 0 {
+				maxRequestsPerPeer = tt.maxRequestsPerPeer
+			}
+
+			for _, step := range tt.steps {
+				assert.Equal(t, step.currentState, testBcR.fsm.state.name)
+
+				oldNumStatusRequests := testBcR.numStatusRequests
+				fixBlockResponseEvStep(&step, testBcR)
+				if !shouldApplyProcessedBlockEvStep(&step, testBcR) {
+					continue
+				}
+
+				fsmErr := sendEventToFSM(testBcR.fsm, step.event, step.data)
+				assert.Equal(t, step.wantErr, fsmErr)
+
+				if step.wantStatusReqSent {
+					assert.Equal(t, oldNumStatusRequests+1, testBcR.numStatusRequests)
+				} else {
+					assert.Equal(t, oldNumStatusRequests, testBcR.numStatusRequests)
+				}
+
+				assert.Equal(t, step.wantState, testBcR.fsm.state.name)
+				if step.wantState == "finished" {
+					assert.True(t, testBcR.fsm.isCaughtUp())
+				}
+			}
+
+		})
+	}
+}
+
+// ----------------------------------------
+// implements the bcRNotifier
+func (testR *testReactor) sendPeerError(err error, peerID p2p.ID) {
+	testR.logger.Info("Reactor received sendPeerError call from FSM", "peer", peerID, "err", err)
+	testR.lastPeerError.peerID = peerID
+	testR.lastPeerError.err = err
+}
+
+func (testR *testReactor) sendStatusRequest() {
+	testR.logger.Info("Reactor received sendStatusRequest call from FSM")
+	testR.numStatusRequests++
+}
+
+func (testR *testReactor) sendBlockRequest(peerID p2p.ID, height int64) error {
+	testR.logger.Info("Reactor received sendBlockRequest call from FSM", "peer", peerID, "height", height)
+	testR.numBlockRequests++
+	testR.lastBlockRequest.peerID = peerID
+	testR.lastBlockRequest.height = height
+	if height == 9999999 {
+		// simulate switch does not have peer
+		return errNilPeerForBlockRequest
+	}
+	return nil
+}
+
+func (testR *testReactor) resetStateTimer(name string, timer **time.Timer, timeout time.Duration) {
+	testR.logger.Info("Reactor received resetStateTimer call from FSM", "state", name, "timeout", timeout)
+	if _, ok := testR.stateTimerStarts[name]; !ok {
+		testR.stateTimerStarts[name] = 1
+	} else {
+		testR.stateTimerStarts[name]++
+	}
+}
+
+func (testR *testReactor) switchToConsensus() {
+}
+
+// ----------------------------------------

blockchain/v1/reactor_test.go

@@ -0,0 +1,337 @@
+package v1
+
+import (
+	"fmt"
+	"os"
+	"sort"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	abci "github.com/tendermint/tendermint/abci/types"
+	cfg "github.com/tendermint/tendermint/config"
+	cmn "github.com/tendermint/tendermint/libs/common"
+	"github.com/tendermint/tendermint/libs/log"
+	"github.com/tendermint/tendermint/mock"
+	"github.com/tendermint/tendermint/p2p"
+	"github.com/tendermint/tendermint/proxy"
+	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
+	"github.com/tendermint/tendermint/types"
+	tmtime "github.com/tendermint/tendermint/types/time"
+	dbm "github.com/tendermint/tm-cmn/db"
+)
+
+var config *cfg.Config
+
+func randGenesisDoc(numValidators int, randPower bool, minPower int64) (*types.GenesisDoc, []types.PrivValidator) {
+	validators := make([]types.GenesisValidator, numValidators)
+	privValidators := make([]types.PrivValidator, numValidators)
+	for i := 0; i < numValidators; i++ {
+		val, privVal := types.RandValidator(randPower, minPower)
+		validators[i] = types.GenesisValidator{
+			PubKey: val.PubKey,
+			Power:  val.VotingPower,
+		}
+		privValidators[i] = privVal
+	}
+	sort.Sort(types.PrivValidatorsByAddress(privValidators))
+
+	return &types.GenesisDoc{
+		GenesisTime: tmtime.Now(),
+		ChainID:     config.ChainID(),
+		Validators:  validators,
+	}, privValidators
+}
+
+func makeVote(header *types.Header, blockID types.BlockID, valset *types.ValidatorSet, privVal types.PrivValidator) *types.Vote {
+	addr := privVal.GetPubKey().Address()
+	idx, _ := valset.GetByAddress(addr)
+	vote := &types.Vote{
+		ValidatorAddress: addr,
+		ValidatorIndex:   idx,
+		Height:           header.Height,
+		Round:            1,
+		Timestamp:        tmtime.Now(),
+		Type:             types.PrecommitType,
+		BlockID:          blockID,
+	}
+
+	_ = privVal.SignVote(header.ChainID, vote)
+
+	return vote
+}
+
+type BlockchainReactorPair struct {
+	bcR  *BlockchainReactor
+	conR *consensusReactorTest
+}
+
+func newBlockchainReactor(logger log.Logger, genDoc *types.GenesisDoc, privVals []types.PrivValidator, maxBlockHeight int64) *BlockchainReactor {
+	if len(privVals) != 1 {
+		panic("only support one validator")
+	}
+
+	app := &testApp{}
+	cc := proxy.NewLocalClientCreator(app)
+	proxyApp := proxy.NewAppConns(cc)
+	err := proxyApp.Start()
+	if err != nil {
+		panic(cmn.ErrorWrap(err, "error start app"))
+	}
+
+	blockDB := dbm.NewMemDB()
+	stateDB := dbm.NewMemDB()
+	blockStore := store.NewBlockStore(blockDB)
+
+	state, err := sm.LoadStateFromDBOrGenesisDoc(stateDB, genDoc)
+	if err != nil {
+		panic(cmn.ErrorWrap(err, "error constructing state from genesis file"))
+	}
+
+	// Make the BlockchainReactor itself.
+	// NOTE we have to create and commit the blocks first because
+	// pool.height is determined from the store.
+	fastSync := true
+	db := dbm.NewMemDB()
+	blockExec := sm.NewBlockExecutor(db, log.TestingLogger(), proxyApp.Consensus(),
+		mock.Mempool{}, sm.MockEvidencePool{})
+	sm.SaveState(db, state)
+
+	// let's add some blocks in
+	for blockHeight := int64(1); blockHeight <= maxBlockHeight; blockHeight++ {
+		lastCommit := types.NewCommit(types.BlockID{}, nil)
+		if blockHeight > 1 {
+			lastBlockMeta := blockStore.LoadBlockMeta(blockHeight - 1)
+			lastBlock := blockStore.LoadBlock(blockHeight - 1)
+
+			vote := makeVote(&lastBlock.Header, lastBlockMeta.BlockID, state.Validators, privVals[0]).CommitSig()
+			lastCommit = types.NewCommit(lastBlockMeta.BlockID, []*types.CommitSig{vote})
+		}
+
+		thisBlock := makeBlock(blockHeight, state, lastCommit)
+
+		thisParts := thisBlock.MakePartSet(types.BlockPartSizeBytes)
+		blockID := types.BlockID{Hash: thisBlock.Hash(), PartsHeader: thisParts.Header()}
+
+		state, err = blockExec.ApplyBlock(state, blockID, thisBlock)
+		if err != nil {
+			panic(cmn.ErrorWrap(err, "error apply block"))
+		}
+
+		blockStore.SaveBlock(thisBlock, thisParts, lastCommit)
+	}
+
+	bcReactor := NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
+	bcReactor.SetLogger(logger.With("module", "blockchain"))
+
+	return bcReactor
+}
+
+func newBlockchainReactorPair(logger log.Logger, genDoc *types.GenesisDoc, privVals []types.PrivValidator, maxBlockHeight int64) BlockchainReactorPair {
+
+	consensusReactor := &consensusReactorTest{}
+	consensusReactor.BaseReactor = *p2p.NewBaseReactor("Consensus reactor", consensusReactor)
+
+	return BlockchainReactorPair{
+		newBlockchainReactor(logger, genDoc, privVals, maxBlockHeight),
+		consensusReactor}
+}
+
+type consensusReactorTest struct {
+	p2p.BaseReactor     // BaseService + p2p.Switch
+	switchedToConsensus bool
+	mtx                 sync.Mutex
+}
+
+func (conR *consensusReactorTest) SwitchToConsensus(state sm.State, blocksSynced int) {
+	conR.mtx.Lock()
+	defer conR.mtx.Unlock()
+	conR.switchedToConsensus = true
+}
+
+func TestFastSyncNoBlockResponse(t *testing.T) {
+
+	config = cfg.ResetTestRoot("blockchain_new_reactor_test")
+	defer os.RemoveAll(config.RootDir)
+	genDoc, privVals := randGenesisDoc(1, false, 30)
+
+	maxBlockHeight := int64(65)
+
+	reactorPairs := make([]BlockchainReactorPair, 2)
+
+	logger := log.TestingLogger()
+	reactorPairs[0] = newBlockchainReactorPair(logger, genDoc, privVals, maxBlockHeight)
+	reactorPairs[1] = newBlockchainReactorPair(logger, genDoc, privVals, 0)
+
+	p2p.MakeConnectedSwitches(config.P2P, 2, func(i int, s *p2p.Switch) *p2p.Switch {
+		s.AddReactor("BLOCKCHAIN", reactorPairs[i].bcR)
+		s.AddReactor("CONSENSUS", reactorPairs[i].conR)
+		moduleName := fmt.Sprintf("blockchain-%v", i)
+		reactorPairs[i].bcR.SetLogger(logger.With("module", moduleName))
+
+		return s
+
+	}, p2p.Connect2Switches)
+
+	defer func() {
+		for _, r := range reactorPairs {
+			_ = r.bcR.Stop()
+			_ = r.conR.Stop()
+		}
+	}()
+
+	tests := []struct {
+		height   int64
+		existent bool
+	}{
+		{maxBlockHeight + 2, false},
+		{10, true},
+		{1, true},
+		{maxBlockHeight + 100, false},
+	}
+
+	for {
+		time.Sleep(10 * time.Millisecond)
+		reactorPairs[1].conR.mtx.Lock()
+		if reactorPairs[1].conR.switchedToConsensus {
+			reactorPairs[1].conR.mtx.Unlock()
+			break
+		}
+		reactorPairs[1].conR.mtx.Unlock()
+	}
+
+	assert.Equal(t, maxBlockHeight, reactorPairs[0].bcR.store.Height())
+
+	for _, tt := range tests {
+		block := reactorPairs[1].bcR.store.LoadBlock(tt.height)
+		if tt.existent {
+			assert.True(t, block != nil)
+		} else {
+			assert.True(t, block == nil)
+		}
+	}
+}
+
+// NOTE: This is too hard to test without
+// an easy way to add test peer to switch
+// or without significant refactoring of the module.
+// Alternatively we could actually dial a TCP conn but
+// that seems extreme.
+func TestFastSyncBadBlockStopsPeer(t *testing.T) {
+	numNodes := 4
+	maxBlockHeight := int64(148)
+
+	config = cfg.ResetTestRoot("blockchain_reactor_test")
+	defer os.RemoveAll(config.RootDir)
+	genDoc, privVals := randGenesisDoc(1, false, 30)
+
+	otherChain := newBlockchainReactorPair(log.TestingLogger(), genDoc, privVals, maxBlockHeight)
+	defer func() {
+		_ = otherChain.bcR.Stop()
+		_ = otherChain.conR.Stop()
+	}()
+
+	reactorPairs := make([]BlockchainReactorPair, numNodes)
+	logger := make([]log.Logger, numNodes)
+
+	for i := 0; i < numNodes; i++ {
+		logger[i] = log.TestingLogger()
+		height := int64(0)
+		if i == 0 {
+			height = maxBlockHeight
+		}
+		reactorPairs[i] = newBlockchainReactorPair(logger[i], genDoc, privVals, height)
+	}
+
+	switches := p2p.MakeConnectedSwitches(config.P2P, numNodes, func(i int, s *p2p.Switch) *p2p.Switch {
+		reactorPairs[i].conR.mtx.Lock()
+		s.AddReactor("BLOCKCHAIN", reactorPairs[i].bcR)
+		s.AddReactor("CONSENSUS", reactorPairs[i].conR)
+		moduleName := fmt.Sprintf("blockchain-%v", i)
+		reactorPairs[i].bcR.SetLogger(logger[i].With("module", moduleName))
+		reactorPairs[i].conR.mtx.Unlock()
+		return s
+
+	}, p2p.Connect2Switches)
+
+	defer func() {
+		for _, r := range reactorPairs {
+			_ = r.bcR.Stop()
+			_ = r.conR.Stop()
+		}
+	}()
+
+outerFor:
+	for {
+		time.Sleep(10 * time.Millisecond)
+		for i := 0; i < numNodes; i++ {
+			reactorPairs[i].conR.mtx.Lock()
+			if !reactorPairs[i].conR.switchedToConsensus {
+				reactorPairs[i].conR.mtx.Unlock()
+				continue outerFor
+			}
+			reactorPairs[i].conR.mtx.Unlock()
+		}
+		break
+	}
+
+	//at this time, reactors[0-3] is the newest
+	assert.Equal(t, numNodes-1, reactorPairs[1].bcR.Switch.Peers().Size())
+
+	//mark last reactorPair as an invalid peer
+	reactorPairs[numNodes-1].bcR.store = otherChain.bcR.store
+
+	lastLogger := log.TestingLogger()
+	lastReactorPair := newBlockchainReactorPair(lastLogger, genDoc, privVals, 0)
+	reactorPairs = append(reactorPairs, lastReactorPair)
+
+	switches = append(switches, p2p.MakeConnectedSwitches(config.P2P, 1, func(i int, s *p2p.Switch) *p2p.Switch {
+		s.AddReactor("BLOCKCHAIN", reactorPairs[len(reactorPairs)-1].bcR)
+		s.AddReactor("CONSENSUS", reactorPairs[len(reactorPairs)-1].conR)
+		moduleName := fmt.Sprintf("blockchain-%v", len(reactorPairs)-1)
+		reactorPairs[len(reactorPairs)-1].bcR.SetLogger(lastLogger.With("module", moduleName))
+		return s
+
+	}, p2p.Connect2Switches)...)
+
+	for i := 0; i < len(reactorPairs)-1; i++ {
+		p2p.Connect2Switches(switches, i, len(reactorPairs)-1)
+	}
+
+	for {
+		time.Sleep(1 * time.Second)
+		lastReactorPair.conR.mtx.Lock()
+		if lastReactorPair.conR.switchedToConsensus {
+			lastReactorPair.conR.mtx.Unlock()
+			break
+		}
+		lastReactorPair.conR.mtx.Unlock()
+
+		if lastReactorPair.bcR.Switch.Peers().Size() == 0 {
+			break
+		}
+	}
+
+	assert.True(t, lastReactorPair.bcR.Switch.Peers().Size() < len(reactorPairs)-1)
+}
+
+//----------------------------------------------
+// utility funcs
+
+func makeTxs(height int64) (txs []types.Tx) {
+	for i := 0; i < 10; i++ {
+		txs = append(txs, types.Tx([]byte{byte(height), byte(i)}))
+	}
+	return txs
+}
+
+func makeBlock(height int64, state sm.State, lastCommit *types.Commit) *types.Block {
+	block, _ := state.MakeBlock(height, makeTxs(height), lastCommit, nil, state.Validators.GetProposer().Address)
+	return block
+}
+
+type testApp struct {
+	abci.BaseApplication
+}

blockchain/v1/wire.go

@@ -0,0 +1,13 @@
+package v1
+
+import (
+	amino "github.com/tendermint/go-amino"
+	"github.com/tendermint/tendermint/types"
+)
+
+var cdc = amino.NewCodec()
+
+func init() {
+	RegisterBlockchainMessages(cdc)
+	types.RegisterBlockAmino(cdc)
+}

cmd/tendermint/commands/run_node.go

@@ -19,7 +19,7 @@ func AddNodeFlags(cmd *cobra.Command) {
 	cmd.Flags().String("priv_validator_laddr", config.PrivValidatorListenAddr, "Socket address to listen on for connections from external priv_validator process")
 
 	// node flags
-	cmd.Flags().Bool("fast_sync", config.FastSync, "Fast blockchain syncing")
+	cmd.Flags().Bool("fast_sync", config.FastSyncMode, "Fast blockchain syncing")
 
 	// abci flags
 	cmd.Flags().String("proxy_app", config.ProxyApp, "Proxy app address, or one of: 'kvstore', 'persistent_kvstore', 'counter', 'counter_serial' or 'noop' for local testing.")

config/config.go

@@ -64,6 +64,7 @@ type Config struct {
 	RPC             *RPCConfig             `mapstructure:"rpc"`
 	P2P             *P2PConfig             `mapstructure:"p2p"`
 	Mempool         *MempoolConfig         `mapstructure:"mempool"`
+	FastSync        *FastSyncConfig        `mapstructure:"fastsync"`
 	Consensus       *ConsensusConfig       `mapstructure:"consensus"`
 	TxIndex         *TxIndexConfig         `mapstructure:"tx_index"`
 	Instrumentation *InstrumentationConfig `mapstructure:"instrumentation"`
@@ -76,6 +77,7 @@ func DefaultConfig() *Config {
 		RPC:             DefaultRPCConfig(),
 		P2P:             DefaultP2PConfig(),
 		Mempool:         DefaultMempoolConfig(),
+		FastSync:        DefaultFastSyncConfig(),
 		Consensus:       DefaultConsensusConfig(),
 		TxIndex:         DefaultTxIndexConfig(),
 		Instrumentation: DefaultInstrumentationConfig(),
@@ -89,6 +91,7 @@ func TestConfig() *Config {
 		RPC:             TestRPCConfig(),
 		P2P:             TestP2PConfig(),
 		Mempool:         TestMempoolConfig(),
+		FastSync:        TestFastSyncConfig(),
 		Consensus:       TestConsensusConfig(),
 		TxIndex:         TestTxIndexConfig(),
 		Instrumentation: TestInstrumentationConfig(),
@@ -120,6 +123,9 @@ func (cfg *Config) ValidateBasic() error {
 	if err := cfg.Mempool.ValidateBasic(); err != nil {
 		return errors.Wrap(err, "Error in [mempool] section")
 	}
+	if err := cfg.FastSync.ValidateBasic(); err != nil {
+		return errors.Wrap(err, "Error in [fastsync] section")
+	}
 	if err := cfg.Consensus.ValidateBasic(); err != nil {
 		return errors.Wrap(err, "Error in [consensus] section")
 	}
@@ -151,7 +157,7 @@ type BaseConfig struct {
 	// If this node is many blocks behind the tip of the chain, FastSync
 	// allows them to catchup quickly by downloading blocks in parallel
 	// and verifying their commits
-	FastSync bool `mapstructure:"fast_sync"`
+	FastSyncMode bool `mapstructure:"fast_sync"`
 
 	// Database backend: goleveldb | cleveldb | boltdb
 	// * goleveldb (github.com/syndtr/goleveldb - most popular implementation)
@@ -216,7 +222,7 @@ func DefaultBaseConfig() BaseConfig {
 		LogLevel:           DefaultPackageLogLevels(),
 		LogFormat:          LogFormatPlain,
 		ProfListenAddress:  "",
-		FastSync:           true,
+		FastSyncMode:       true,
 		FilterPeers:        false,
 		DBBackend:          "goleveldb",
 		DBPath:             "data",
@@ -228,7 +234,7 @@ func TestBaseConfig() BaseConfig {
 	cfg := DefaultBaseConfig()
 	cfg.chainID = "tendermint_test"
 	cfg.ProxyApp = "kvstore"
-	cfg.FastSync = false
+	cfg.FastSyncMode = false
 	cfg.DBBackend = "memdb"
 	return cfg
 }
@@ -684,6 +690,38 @@ func (cfg *MempoolConfig) ValidateBasic() error {
 	return nil
 }
 
+//-----------------------------------------------------------------------------
+// FastSyncConfig
+
+// FastSyncConfig defines the configuration for the Tendermint fast sync service
+type FastSyncConfig struct {
+	Version string `mapstructure:"version"`
+}
+
+// DefaultFastSyncConfig returns a default configuration for the fast sync service
+func DefaultFastSyncConfig() *FastSyncConfig {
+	return &FastSyncConfig{
+		Version: "v0",
+	}
+}
+
+// TestFastSyncConfig returns a default configuration for the fast sync.
+func TestFastSyncConfig() *FastSyncConfig {
+	return DefaultFastSyncConfig()
+}
+
+// ValidateBasic performs basic validation.
+func (cfg *FastSyncConfig) ValidateBasic() error {
+	switch cfg.Version {
+	case "v0":
+		return nil
+	case "v1":
+		return nil
+	default:
+		return fmt.Errorf("unknown fastsync version %s", cfg.Version)
+	}
+}
+
 //-----------------------------------------------------------------------------
 // ConsensusConfig
 

config/toml.go

@@ -79,7 +79,7 @@ moniker = "{{ .BaseConfig.Moniker }}"
 # If this node is many blocks behind the tip of the chain, FastSync
 # allows them to catchup quickly by downloading blocks in parallel
 # and verifying their commits
-fast_sync = {{ .BaseConfig.FastSync }}
+fast_sync = {{ .BaseConfig.FastSyncMode }}
 
 # Database backend: goleveldb | cleveldb | boltdb
 # * goleveldb (github.com/syndtr/goleveldb - most popular implementation)
@@ -294,6 +294,14 @@ max_txs_bytes = {{ .Mempool.MaxTxsBytes }}
 # Size of the cache (used to filter transactions we saw earlier) in transactions
 cache_size = {{ .Mempool.CacheSize }}
 
+##### fast sync configuration options #####
+[fastsync]
+
+# Fast Sync version to use:
+#   1) "v0" (default) - the legacy fast sync implementation
+#   2) "v1" - refactor of v0 version for better testability
+version = "{{ .FastSync.Version }}"
+
 # Limit the size of TxMessage
 max_msg_bytes = {{ .Mempool.MaxMsgBytes }}
 

consensus/common_test.go

@@ -20,7 +20,6 @@ import (
 	"github.com/tendermint/tendermint/abci/example/counter"
 	"github.com/tendermint/tendermint/abci/example/kvstore"
 	abci "github.com/tendermint/tendermint/abci/types"
-	bc "github.com/tendermint/tendermint/blockchain"
 	cfg "github.com/tendermint/tendermint/config"
 	cstypes "github.com/tendermint/tendermint/consensus/types"
 	cmn "github.com/tendermint/tendermint/libs/common"
@@ -30,6 +29,7 @@ import (
 	"github.com/tendermint/tendermint/p2p"
 	"github.com/tendermint/tendermint/privval"
 	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
 	"github.com/tendermint/tendermint/types"
 	tmtime "github.com/tendermint/tendermint/types/time"
 	dbm "github.com/tendermint/tm-cmn/db"
@@ -280,7 +280,7 @@ func newConsensusStateWithConfig(thisConfig *cfg.Config, state sm.State, pv type
 
 func newConsensusStateWithConfigAndBlockStore(thisConfig *cfg.Config, state sm.State, pv types.PrivValidator, app abci.Application, blockDB dbm.DB) *ConsensusState {
 	// Get BlockStore
-	blockStore := bc.NewBlockStore(blockDB)
+	blockStore := store.NewBlockStore(blockDB)
 
 	// one for mempool, one for consensus
 	mtx := new(sync.Mutex)

consensus/reactor_test.go

@@ -17,13 +17,13 @@ import (
 	abcicli "github.com/tendermint/tendermint/abci/client"
 	"github.com/tendermint/tendermint/abci/example/kvstore"
 	abci "github.com/tendermint/tendermint/abci/types"
-	bc "github.com/tendermint/tendermint/blockchain"
 	cfg "github.com/tendermint/tendermint/config"
 	"github.com/tendermint/tendermint/libs/log"
 	mempl "github.com/tendermint/tendermint/mempool"
 	"github.com/tendermint/tendermint/p2p"
 	"github.com/tendermint/tendermint/p2p/mock"
 	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
 	"github.com/tendermint/tendermint/types"
 	dbm "github.com/tendermint/tm-cmn/db"
 )
@@ -133,7 +133,7 @@ func TestReactorWithEvidence(t *testing.T) {
 		// css[i] = newConsensusStateWithConfig(thisConfig, state, privVals[i], app)
 
 		blockDB := dbm.NewMemDB()
-		blockStore := bc.NewBlockStore(blockDB)
+		blockStore := store.NewBlockStore(blockDB)
 
 		// one for mempool, one for consensus
 		mtx := new(sync.Mutex)

consensus/replay_file.go

@@ -12,13 +12,13 @@ import (
 	"github.com/pkg/errors"
 	dbm "github.com/tendermint/tm-cmn/db"
 
-	bc "github.com/tendermint/tendermint/blockchain"
 	cfg "github.com/tendermint/tendermint/config"
 	cmn "github.com/tendermint/tendermint/libs/common"
 	"github.com/tendermint/tendermint/libs/log"
 	"github.com/tendermint/tendermint/mock"
 	"github.com/tendermint/tendermint/proxy"
 	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
 	"github.com/tendermint/tendermint/types"
 )
 
@@ -280,7 +280,7 @@ func newConsensusStateForReplay(config cfg.BaseConfig, csConfig *cfg.ConsensusCo
 	dbType := dbm.DBBackendType(config.DBBackend)
 	// Get BlockStore
 	blockStoreDB := dbm.NewDB("blockstore", dbType, config.DBDir())
-	blockStore := bc.NewBlockStore(blockStoreDB)
+	blockStore := store.NewBlockStore(blockStoreDB)
 
 	// Get State
 	stateDB := dbm.NewDB("state", dbType, config.DBDir())

consensus/wal_generator.go

@@ -12,7 +12,6 @@ import (
 	"github.com/pkg/errors"
 
 	"github.com/tendermint/tendermint/abci/example/kvstore"
-	bc "github.com/tendermint/tendermint/blockchain"
 	cfg "github.com/tendermint/tendermint/config"
 	cmn "github.com/tendermint/tendermint/libs/common"
 	"github.com/tendermint/tendermint/libs/log"
@@ -20,6 +19,7 @@ import (
 	"github.com/tendermint/tendermint/privval"
 	"github.com/tendermint/tendermint/proxy"
 	sm "github.com/tendermint/tendermint/state"
+	"github.com/tendermint/tendermint/store"
 	"github.com/tendermint/tendermint/types"
 	"github.com/tendermint/tm-cmn/db"
 )
@@ -55,7 +55,8 @@ func WALGenerateNBlocks(t *testing.T, wr io.Writer, numBlocks int) (err error) {
 	}
 	state.Version.Consensus.App = kvstore.ProtocolVersion
 	sm.SaveState(stateDB, state)
-	blockStore := bc.NewBlockStore(blockStoreDB)
+	blockStore := store.NewBlockStore(blockStoreDB)
+
 	proxyApp := proxy.NewAppConns(proxy.NewLocalClientCreator(app))
 	proxyApp.SetLogger(logger.With("module", "proxy"))
 	if err := proxyApp.Start(); err != nil {

docs/spec/reactors/block_sync/bcv1/impl-v1.md

@@ -0,0 +1,237 @@
+# Blockchain Reactor v1
+
+### Data Structures
+The data structures used are illustrated below.
+
+![Data Structures](img/bc-reactor-new-datastructs.png)
+
+#### BlockchainReactor
+- is a `p2p.BaseReactor`.
+- has a `store.BlockStore` for persistence.
+- executes blocks using an `sm.BlockExecutor`.
+- starts the FSM and the `poolRoutine()`.
+- relays the fast-sync responses and switch messages to the FSM.
+- handles errors from the FSM and when necessarily reports them to the switch.
+- implements the blockchain reactor interface used by the FSM to send requests, errors to the switch and state timer resets.
+- registers all the concrete types and interfaces for serialisation.
+
+```go
+type BlockchainReactor struct {
+	p2p.BaseReactor
+
+	initialState sm.State // immutable
+	state        sm.State
+
+	blockExec *sm.BlockExecutor
+	store     *store.BlockStore
+
+	fastSync bool
+
+	fsm          *BcReactorFSM
+	blocksSynced int
+
+	// Receive goroutine forwards messages to this channel to be processed in the context of the poolRoutine.
+	messagesForFSMCh chan bcReactorMessage
+
+	// Switch goroutine may send RemovePeer to the blockchain reactor. This is an error message that is relayed
+	// to this channel to be processed in the context of the poolRoutine.
+	errorsForFSMCh chan bcReactorMessage
+
+	// This channel is used by the FSM and indirectly the block pool to report errors to the blockchain reactor and
+	// the switch.
+	eventsFromFSMCh chan bcFsmMessage
+}
+```
+
+#### BcReactorFSM
+- implements a simple finite state machine.
+- has a state and a state timer.
+- has a `BlockPool` to keep track of block requests sent to peers and blocks received from peers.
+- uses an interface to send status requests, block requests and reporting errors. The interface is implemented by the `BlockchainReactor` and tests.
+
+```go
+type BcReactorFSM struct {
+	logger log.Logger
+	mtx    sync.Mutex
+
+	startTime time.Time
+
+	state      *bcReactorFSMState
+	stateTimer *time.Timer
+	pool       *BlockPool
+
+	// interface used to call the Blockchain reactor to send StatusRequest, BlockRequest, reporting errors, etc.
+	toBcR bcReactor
+}
+```
+
+#### BlockPool
+- maintains a peer set, implemented as a map of peer ID to `BpPeer`.
+- maintains a set of requests made to peers, implemented as a map of block request heights to peer IDs.
+- maintains a list of future block requests needed to advance the fast-sync. This is a list of block heights. 
+- keeps track of the maximum height of the peers in the set.
+- uses an interface to send requests and report errors to the reactor (via FSM).
+
+```go
+type BlockPool struct {
+	logger log.Logger
+	// Set of peers that have sent status responses, with height bigger than pool.Height
+	peers map[p2p.ID]*BpPeer
+	// Set of block heights and the corresponding peers from where a block response is expected or has been received.
+	blocks map[int64]p2p.ID
+
+	plannedRequests   map[int64]struct{} // list of blocks to be assigned peers for blockRequest
+	nextRequestHeight int64              // next height to be added to plannedRequests
+
+	Height        int64 // height of next block to execute
+	MaxPeerHeight int64 // maximum height of all peers
+	toBcR         bcReactor
+}
+```
+Some reasons for the `BlockPool` data structure content:
+1. If a peer is removed by the switch fast access is required to the peer and the block requests made to that peer in order to redo them.
+2. When block verification fails fast access is required from the block height to the peer and the block requests made to that peer in order to redo them.
+3. The `BlockchainReactor` main routine decides when the block pool is running low and asks the `BlockPool` (via FSM) to make more requests. The `BlockPool` creates a list of requests and triggers the sending of the block requests (via the interface). The reason it maintains a list of requests is the redo operations that may occur during error handling. These are redone when the `BlockchainReactor` requires more blocks.
+
+#### BpPeer
+- keeps track of a single peer, with height bigger than the initial height.
+- maintains the block requests made to the peer and the blocks received from the peer until they are executed.
+- monitors the peer speed when there are pending requests.
+- it has an active timer when pending requests are present and reports error on timeout.
+
+```go
+type BpPeer struct {
+	logger log.Logger
+	ID     p2p.ID
+
+	Height                  int64                  // the peer reported height
+	NumPendingBlockRequests int                    // number of requests still waiting for block responses
+	blocks                  map[int64]*types.Block // blocks received or expected to be received from this peer
+	blockResponseTimer      *time.Timer
+	recvMonitor             *flow.Monitor
+	params                  *BpPeerParams // parameters for timer and monitor
+
+	onErr func(err error, peerID p2p.ID) // function to call on error
+}
+```
+
+### Concurrency Model
+
+The diagram below shows the goroutines (depicted by the gray blocks), timers (shown on the left with their values) and channels (colored rectangles). The FSM box shows some of the functionality and it is not a separate goroutine.
+
+The interface used by the FSM is shown in light red with the `IF` block. This is used to:
+- send block requests 
+- report peer errors to the switch - this results in the reactor calling `switch.StopPeerForError()` and, if triggered by the peer timeout routine, a `removePeerEv` is sent to the FSM and action is taken from the context of the `poolRoutine()`
+- ask the reactor to reset the state timers. The timers are owned by the FSM while the timeout routine is defined by the reactor. This was done in order to avoid running timers in tests and will change in the next revision.
+ 
+There are two main goroutines implemented by the blockchain reactor. All I/O operations are performed from the `poolRoutine()` context while the CPU intensive operations related to the block execution are performed from the context of the `executeBlocksRoutine()`. All goroutines are detailed in the next sections.
+
+![Go Routines Diagram](img/bc-reactor-new-goroutines.png)
+
+#### Receive()
+Fast-sync messages from peers are received by this goroutine. It performs basic validation and:
+- in helper mode (i.e. for request message) it replies immediately. This is different than the proposal in adr-040 that specifies having the FSM handling these.
+- forwards response messages to the `poolRoutine()`.
+
+#### poolRoutine()
+(named kept as in the previous reactor). 
+It starts the `executeBlocksRoutine()` and the FSM. It then waits in a loop for events. These are received from the following channels:
+- `sendBlockRequestTicker.C` - every 10msec the reactor asks FSM to make more block requests up to a maximum. Note: currently this value is constant but could be changed based on low/ high watermark thresholds for the number of blocks received and waiting to be processed, the number of blockResponse messages waiting in messagesForFSMCh, etc.
+- `statusUpdateTicker.C` - every 10 seconds the reactor broadcasts status requests to peers. While adr-040 specifies this to run within the FSM, at this point this functionality is kept in the reactor.
+- `messagesForFSMCh` - the `Receive()` goroutine sends status and block response messages to this channel and the reactor calls FSM to handle them.
+- `errorsForFSMCh` - this channel receives the following events: 
+    - peer remove - when the switch removes a peer
+    - sate timeout event - when FSM state timers trigger
+  The reactor forwards this messages to the FSM.
+- `eventsFromFSMCh` - there are two type of events sent over this channel:
+    - `syncFinishedEv` - triggered when FSM enters `finished` state and calls the switchToConsensus() interface function.
+    - `peerErrorEv`- peer timer expiry goroutine sends this event over the channel for processing from poolRoutine() context.
+    
+#### executeBlocksRoutine()
+Started by the `poolRoutine()`, it retrieves blocks from the pool and executes them:
+- `processReceivedBlockTicker.C` - a ticker event is received over the channel every 10msec and its handling results in a signal being sent to the doProcessBlockCh channel.
+- doProcessBlockCh - events are received on this channel as described as above and upon processing blocks are retrieved from the pool and executed. 
+
+
+### FSM
+
+![fsm](img/bc-reactor-new-fsm.png)
+
+#### States
+##### init (aka unknown)
+The FSM is created in `unknown` state. When started, by the reactor (`startFSMEv`), it broadcasts Status requests and transitions to `waitForPeer` state.
+
+##### waitForPeer
+In this state, the FSM waits for a Status responses from a "tall" peer. A timer is running in this state to allow the FSM to finish if there are no useful peers.
+
+If the timer expires, it moves to `finished` state and calls the reactor to switch to consensus.
+If a Status response is received from a peer within the timeout, the FSM transitions to `waitForBlock` state.
+
+##### waitForBlock
+In this state the FSM makes Block requests (triggered by a ticker in reactor) and waits for Block responses. There is a timer running in this state to detect if a peer is not sending the block at current processing height. If the timer expires, the FSM removes the peer where the request was sent and all requests made to that peer are redone.
+
+As blocks are received they are stored by the pool. Block execution is independently performed by the reactor and the result reported to the FSM:
+- if there are no errors, the FSM increases the pool height and resets the state timer.
+- if there are errors, the peers that delivered the two blocks (at height and height+1) are removed and the requests redone.
+
+In this state the FSM may receive peer remove events in any of the following scenarios: 
+- the switch is removing a peer
+- a peer is penalized because it has not responded to some block requests for a long time
+- a peer is penalized for being slow
+
+When processing of the last block (the one with height equal to the highest peer height minus one) is successful, the FSM transitions to `finished` state.
+If after a peer update or removal the pool height is same as maxPeerHeight, the FSM transitions to `finished` state.
+
+##### finished
+When entering this state, the FSM calls the reactor to switch to consensus and performs cleanup.
+
+#### Events
+
+The following events are handled by the FSM:
+
+```go
+const (
+	startFSMEv = iota + 1
+	statusResponseEv
+	blockResponseEv
+	processedBlockEv
+	makeRequestsEv
+	stopFSMEv
+	peerRemoveEv = iota + 256
+	stateTimeoutEv
+)
+```
+
+### Examples of Scenarios and Termination Handling
+A few scenarios are covered in this section together with the current/ proposed handling.
+In general, the scenarios involving faulty peers are made worse by the fact that they may quickly be re-added.
+
+#### 1. No Tall Peers
+
+S: In this scenario a node is started and while there are status responses received, none of the peers are at a height higher than this node.
+
+H: The FSM times out in `waitForPeer` state, moves to `finished` state where it calls the reactor to switch to consensus.
+
+#### 2. Typical Fast Sync
+
+S: A node fast syncs blocks from honest peers and eventually downloads and executes the penultimate block.
+
+H: The FSM in `waitForBlock` state will receive the processedBlockEv from the reactor and detect that the termination height is achieved.
+
+#### 3. Peer Claims Big Height but no Blocks
+
+S: In this scenario a faulty peer claims a big height (for which there are no blocks).
+
+H: The requests for the non-existing block will timeout, the peer removed and the pool's `MaxPeerHeight` updated. FSM checks if the termination height is achieved when peers are removed.
+
+#### 4. Highest Peer Removed or Updated to Short
+
+S: The fast sync node is caught up with all peers except one tall peer. The tall peer is removed or it sends status response with low height.
+
+H: FSM checks termination condition on peer removal and updates.
+
+#### 5. Block At Current Height Delayed
+
+S: A peer can block the progress of fast sync by delaying indefinitely the block response for the current processing height (h1).
+
+H: Currently, given h1 < h2, there is no enforcement at peer level that the response for h1 should be received before h2. So a peer will timeout only after delivering all blocks except h1. However the `waitForBlock` state timer fires if the block for current processing height is not received within a timeout. The peer is removed and the requests to that peer (including the one for current height) redone.

docs/spec/reactors/block_sync/impl.md

@@ -1,4 +1,6 @@
-## Blockchain Reactor
+## Blockchain Reactor v0 Modules
+
+### Blockchain Reactor
 
 - coordinates the pool for syncing
 - coordinates the store for persistence
@@ -8,7 +10,7 @@
 - starts the pool.Start() and its poolRoutine()
 - registers all the concrete types and interfaces for serialisation
 
-### poolRoutine
+#### poolRoutine
 
 - listens to these channels:
   - pool requests blocks from a specific peer by posting to requestsCh, block reactor then sends
@@ -22,7 +24,7 @@
 - implements Receive which is called by the switch/peer
   - calls AddBlock on the pool when it receives a new block from a peer
 
-## Block Pool
+### Block Pool
 
 - responsible for downloading blocks from peers
 - makeRequestersRoutine()
@@ -36,6 +38,7 @@
     - we receive a block
       - gotBlockCh is strange
 
-## Block Store
 
-- persists blocks to disk
+### Go Routines in Blockchain Reactor
+
+![Go Routines Diagram](img/bc-reactor-routines.png)

docs/tendermint-core/configuration.md

@@ -240,6 +240,14 @@ max_txs_bytes = 1073741824
 # Size of the cache (used to filter transactions we saw earlier) in transactions
 cache_size = 10000
 
+##### fast sync configuration options #####
+[fastsync]
+
+# Fast Sync version to use:
+#   1) "v0" (default) - the legacy fast sync implementation
+#   2) "v1" - refactor of v0 version for better testability
+version = "v0"
+
 # Limit the size of TxMessage
 max_msg_bytes = 1048576
 

go.sum

@@ -39,7 +39,6 @@ github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b h1:VKtxabqXZkF25pY9ekf
 github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b/go.mod h1:SBH7ygxi8pfUlaOkMMuAQtPIUF8ecWP5IEl/CR7VP2Q=
 github.com/golang/mock v1.1.1/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
 github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
-github.com/golang/protobuf v1.3.0/go.mod h1:Qd/q+1AKNOZr9uGQzbzCmRO6sUih6GTPZv6a1/R87v0=
 github.com/golang/protobuf v1.3.2 h1:6nsPYzhq5kReh6QImI3k5qWzO4PEbvbIW2cwSfR/6xs=
 github.com/golang/protobuf v1.3.2/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
 github.com/golang/snappy v0.0.1 h1:Qgr9rKW7uDUkrbSmQeiDsGa8SjGyCOGtuasMWwvp2P4=
@@ -79,7 +78,6 @@ github.com/onsi/gomega v1.4.3 h1:RE1xgDvH7imwFD45h+u2SgIfERHlS2yNG4DObb5BSKU=
 github.com/onsi/gomega v1.4.3/go.mod h1:ex+gbHU/CVuBBDIJjb2X0qEXbFg53c61hWP/1CpauHY=
 github.com/pelletier/go-toml v1.2.0 h1:T5zMGML61Wp+FlcbWjRDT7yAxhJNAiPPLOFECq181zc=
 github.com/pelletier/go-toml v1.2.0/go.mod h1:5z9KED0ma1S8pY6P1sdut58dfprrGBbd/94hg7ilaic=
-github.com/pkg/errors v0.8.0/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I=
 github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
@@ -122,7 +120,6 @@ github.com/tendermint/tm-cmn v0.0.0-20190716080004-dfcde30d5acb/go.mod h1:SLI3Mc
 go.etcd.io/bbolt v1.3.3 h1:MUGmc65QhB3pIlaQ5bB4LwqSj6GIonVJXpZiaKNyaKk=
 go.etcd.io/bbolt v1.3.3/go.mod h1:IbVyRI1SCnLcuJnV2u8VeU0CEYM7e686BmAb1XKL+uU=
 golang.org/x/crypto v0.0.0-20170930174604-9419663f5a44/go.mod h1:6SG95UA2DQfeDnfUPMdvaQW0Q7yPrPDi9nlGo2tz2b4=
-golang.org/x/crypto v0.0.0-20190228161510-8dd112bcdc25/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2 h1:VklqNMn3ovrHsnt90PveolxSbWFaJdECFbxSq0Mqo2M=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
@@ -145,10 +142,8 @@ golang.org/x/tools v0.0.0-20190311212946-11955173bddd/go.mod h1:LCzVGOaR6xXOjkQ3
 golang.org/x/tools v0.0.0-20190524140312-2c0ae7006135/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
 google.golang.org/appengine v1.1.0/go.mod h1:EbEs0AVv82hx2wNQdGPgUI5lhzA/G0D9YwlJXL52JkM=
 google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
-google.golang.org/genproto v0.0.0-20180831171423-11092d34479b/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
 google.golang.org/genproto v0.0.0-20181029155118-b69ba1387ce2 h1:67iHsV9djwGdZpdZNbLuQj6FOzCaZe3w+vhLjn5AcFA=
 google.golang.org/genproto v0.0.0-20181029155118-b69ba1387ce2/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
-google.golang.org/grpc v1.13.0/go.mod h1:yo6s7OP7yaDglbqo1J04qKzAhqBH6lvTonzMVmEdcZw=
 google.golang.org/grpc v1.22.0 h1:J0UbZOIrCAl+fpTOf8YLs4dJo8L/owV4LYVtAXQoPkw=
 google.golang.org/grpc v1.22.0/go.mod h1:Y5yQAOtifL1yxbo5wqy6BxZv8vAUGQwXBOALyacEbxg=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=

node/node.go

@@ -18,8 +18,8 @@ import (
 
 	amino "github.com/tendermint/go-amino"
 	abci "github.com/tendermint/tendermint/abci/types"
-	"github.com/tendermint/tendermint/blockchain"
-	bc "github.com/tendermint/tendermint/blockchain"
+	bcv0 "github.com/tendermint/tendermint/blockchain/v0"
+	bcv1 "github.com/tendermint/tendermint/blockchain/v1"
 	cfg "github.com/tendermint/tendermint/config"
 	"github.com/tendermint/tendermint/consensus"
 	cs "github.com/tendermint/tendermint/consensus"
@@ -41,6 +41,7 @@ import (
 	"github.com/tendermint/tendermint/state/txindex"
 	"github.com/tendermint/tendermint/state/txindex/kv"
 	"github.com/tendermint/tendermint/state/txindex/null"
+	"github.com/tendermint/tendermint/store"
 	"github.com/tendermint/tendermint/types"
 	tmtime "github.com/tendermint/tendermint/types/time"
 	"github.com/tendermint/tendermint/version"
@@ -175,9 +176,9 @@ type Node struct {
 	// services
 	eventBus         *types.EventBus // pub/sub for services
 	stateDB          dbm.DB
-	blockStore       *bc.BlockStore        // store the blockchain to disk
-	bcReactor        *bc.BlockchainReactor // for fast-syncing
-	mempoolReactor   *mempl.Reactor        // for gossipping transactions
+	blockStore       *store.BlockStore // store the blockchain to disk
+	bcReactor        p2p.Reactor       // for fast-syncing
+	mempoolReactor   *mempl.Reactor    // for gossipping transactions
 	mempool          mempl.Mempool
 	consensusState   *cs.ConsensusState     // latest consensus state
 	consensusReactor *cs.ConsensusReactor   // for participating in the consensus
@@ -190,13 +191,13 @@ type Node struct {
 	prometheusSrv    *http.Server
 }
 
-func initDBs(config *cfg.Config, dbProvider DBProvider) (blockStore *bc.BlockStore, stateDB dbm.DB, err error) {
+func initDBs(config *cfg.Config, dbProvider DBProvider) (blockStore *store.BlockStore, stateDB dbm.DB, err error) {
 	var blockStoreDB dbm.DB
 	blockStoreDB, err = dbProvider(&DBContext{"blockstore", config})
 	if err != nil {
 		return
 	}
-	blockStore = bc.NewBlockStore(blockStoreDB)
+	blockStore = store.NewBlockStore(blockStoreDB)
 
 	stateDB, err = dbProvider(&DBContext{"state", config})
 	if err != nil {
@@ -337,6 +338,26 @@ func createEvidenceReactor(config *cfg.Config, dbProvider DBProvider,
 	return evidenceReactor, evidencePool, nil
 }
 
+func createBlockchainReactor(config *cfg.Config,
+	state sm.State,
+	blockExec *sm.BlockExecutor,
+	blockStore *store.BlockStore,
+	fastSync bool,
+	logger log.Logger) (bcReactor p2p.Reactor, err error) {
+
+	switch config.FastSync.Version {
+	case "v0":
+		bcReactor = bcv0.NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
+	case "v1":
+		bcReactor = bcv1.NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
+	default:
+		return nil, fmt.Errorf("unknown fastsync version %s", config.FastSync.Version)
+	}
+
+	bcReactor.SetLogger(logger.With("module", "blockchain"))
+	return bcReactor, nil
+}
+
 func createConsensusReactor(config *cfg.Config,
 	state sm.State,
 	blockExec *sm.BlockExecutor,
@@ -431,7 +452,7 @@ func createSwitch(config *cfg.Config,
 	p2pMetrics *p2p.Metrics,
 	peerFilters []p2p.PeerFilterFunc,
 	mempoolReactor *mempl.Reactor,
-	bcReactor *blockchain.BlockchainReactor,
+	bcReactor p2p.Reactor,
 	consensusReactor *consensus.ConsensusReactor,
 	evidenceReactor *evidence.EvidenceReactor,
 	nodeInfo p2p.NodeInfo,
@@ -572,7 +593,7 @@ func NewNode(config *cfg.Config,
 
 	// Decide whether to fast-sync or not
 	// We don't fast-sync when the only validator is us.
-	fastSync := config.FastSync && !onlyValidatorIsUs(state, privValidator)
+	fastSync := config.FastSyncMode && !onlyValidatorIsUs(state, privValidator)
 
 	csMetrics, p2pMetrics, memplMetrics, smMetrics := metricsProvider(genDoc.ChainID)
 
@@ -596,8 +617,10 @@ func NewNode(config *cfg.Config,
 	)
 
 	// Make BlockchainReactor
-	bcReactor := bc.NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
-	bcReactor.SetLogger(logger.With("module", "blockchain"))
+	bcReactor, err := createBlockchainReactor(config, state, blockExec, blockStore, fastSync, logger)
+	if err != nil {
+		return nil, errors.Wrap(err, "could not create blockchain reactor")
+	}
 
 	// Make ConsensusReactor
 	consensusReactor, consensusState := createConsensusReactor(
@@ -930,7 +953,7 @@ func (n *Node) Switch() *p2p.Switch {
 }
 
 // BlockStore returns the Node's BlockStore.
-func (n *Node) BlockStore() *bc.BlockStore {
+func (n *Node) BlockStore() *store.BlockStore {
 	return n.blockStore
 }
 
@@ -1018,6 +1041,17 @@ func makeNodeInfo(
 	if _, ok := txIndexer.(*null.TxIndex); ok {
 		txIndexerStatus = "off"
 	}
+
+	var bcChannel byte
+	switch config.FastSync.Version {
+	case "v0":
+		bcChannel = bcv0.BlockchainChannel
+	case "v1":
+		bcChannel = bcv1.BlockchainChannel
+	default:
+		return nil, fmt.Errorf("unknown fastsync version %s", config.FastSync.Version)
+	}
+
 	nodeInfo := p2p.DefaultNodeInfo{
 		ProtocolVersion: p2p.NewProtocolVersion(
 			version.P2PProtocol, // global
@@ -1028,7 +1062,7 @@ func makeNodeInfo(
 		Network: genDoc.ChainID,
 		Version: version.TMCoreSemVer,
 		Channels: []byte{
-			bc.BlockchainChannel,
+			bcChannel,
 			cs.StateChannel, cs.DataChannel, cs.VoteChannel, cs.VoteSetBitsChannel,
 			mempl.MempoolChannel,
 			evidence.EvidenceChannel,

store/wire.go

@@ -0,0 +1,12 @@
+package store
+
+import (
+	amino "github.com/tendermint/go-amino"
+	cryptoAmino "github.com/tendermint/tendermint/crypto/encoding/amino"
+)
+
+var cdc = amino.NewCodec()
+
+func init() {
+	cryptoAmino.RegisterAmino(cdc)
+}