libs: Call Flush() before rename #2428 (#2439)

* fix Group.RotateFile need call Flush() before rename. #2428
* fix some review issue. #2428
 refactor Group's config: replace  setting member with initial option
* fix a handwriting mistake
* fix a time window error between rename and write.
* fix a syntax mistake.
* change option name Get_ to With_
* fix review issue
* fix review issue

consensus/wal.go

@@ -73,13 +73,13 @@ type baseWAL struct {
 	enc *WALEncoder
 }
 
-func NewWAL(walFile string) (*baseWAL, error) {
+func NewWAL(walFile string, groupOptions ...func(*auto.Group)) (*baseWAL, error) {
 	err := cmn.EnsureDir(filepath.Dir(walFile), 0700)
 	if err != nil {
 		return nil, errors.Wrap(err, "failed to ensure WAL directory is in place")
 	}
 
-	group, err := auto.OpenGroup(walFile)
+	group, err := auto.OpenGroup(walFile, groupOptions...)
 	if err != nil {
 		return nil, err
 	}

consensus/wal_generator.go

@@ -4,6 +4,7 @@ import (
 	"bufio"
 	"bytes"
 	"fmt"
+	"io"
 	"os"
 	"path/filepath"
 	"strings"
@@ -23,12 +24,11 @@ import (
 	"github.com/tendermint/tendermint/types"
 )
 
-// WALWithNBlocks generates a consensus WAL. It does this by spining up a
+// WALGenerateNBlocks generates a consensus WAL. It does this by spining up a
 // stripped down version of node (proxy app, event bus, consensus state) with a
 // persistent kvstore application and special consensus wal instance
-// (byteBufferWAL) and waits until numBlocks are created. Then it returns a WAL
-// content. If the node fails to produce given numBlocks, it returns an error.
-func WALWithNBlocks(numBlocks int) (data []byte, err error) {
+// (byteBufferWAL) and waits until numBlocks are created. If the node fails to produce given numBlocks, it returns an error.
+func WALGenerateNBlocks(wr io.Writer, numBlocks int) (err error) {
 	config := getConfig()
 
 	app := kvstore.NewPersistentKVStoreApplication(filepath.Join(config.DBDir(), "wal_generator"))
@@ -43,26 +43,26 @@ func WALWithNBlocks(numBlocks int) (data []byte, err error) {
 	privValidator := privval.LoadOrGenFilePV(privValidatorFile)
 	genDoc, err := types.GenesisDocFromFile(config.GenesisFile())
 	if err != nil {
-		return nil, errors.Wrap(err, "failed to read genesis file")
+		return errors.Wrap(err, "failed to read genesis file")
 	}
 	stateDB := db.NewMemDB()
 	blockStoreDB := db.NewMemDB()
 	state, err := sm.MakeGenesisState(genDoc)
 	if err != nil {
-		return nil, errors.Wrap(err, "failed to make genesis state")
+		return errors.Wrap(err, "failed to make genesis state")
 	}
 	blockStore := bc.NewBlockStore(blockStoreDB)
 	proxyApp := proxy.NewAppConns(proxy.NewLocalClientCreator(app))
 	proxyApp.SetLogger(logger.With("module", "proxy"))
 	if err := proxyApp.Start(); err != nil {
-		return nil, errors.Wrap(err, "failed to start proxy app connections")
+		return errors.Wrap(err, "failed to start proxy app connections")
 	}
 	defer proxyApp.Stop()
 
 	eventBus := types.NewEventBus()
 	eventBus.SetLogger(logger.With("module", "events"))
 	if err := eventBus.Start(); err != nil {
-		return nil, errors.Wrap(err, "failed to start event bus")
+		return errors.Wrap(err, "failed to start event bus")
 	}
 	defer eventBus.Stop()
 	mempool := sm.MockMempool{}
@@ -78,8 +78,6 @@ func WALWithNBlocks(numBlocks int) (data []byte, err error) {
 	/////////////////////////////////////////////////////////////////////////////
 
 	// set consensus wal to buffered WAL, which will write all incoming msgs to buffer
-	var b bytes.Buffer
-	wr := bufio.NewWriter(&b)
 	numBlocksWritten := make(chan struct{})
 	wal := newByteBufferWAL(logger, NewWALEncoder(wr), int64(numBlocks), numBlocksWritten)
 	// see wal.go#103
@@ -87,20 +85,32 @@ func WALWithNBlocks(numBlocks int) (data []byte, err error) {
 	consensusState.wal = wal
 
 	if err := consensusState.Start(); err != nil {
-		return nil, errors.Wrap(err, "failed to start consensus state")
+		return errors.Wrap(err, "failed to start consensus state")
 	}
 
 	select {
 	case <-numBlocksWritten:
 		consensusState.Stop()
-		wr.Flush()
-		return b.Bytes(), nil
+		return nil
 	case <-time.After(1 * time.Minute):
 		consensusState.Stop()
-		return []byte{}, fmt.Errorf("waited too long for tendermint to produce %d blocks (grep logs for `wal_generator`)", numBlocks)
+		return fmt.Errorf("waited too long for tendermint to produce %d blocks (grep logs for `wal_generator`)", numBlocks)
 	}
 }
 
+//WALWithNBlocks returns a WAL content with numBlocks.
+func WALWithNBlocks(numBlocks int) (data []byte, err error) {
+	var b bytes.Buffer
+	wr := bufio.NewWriter(&b)
+
+	if err := WALGenerateNBlocks(wr, numBlocks); err != nil {
+		return []byte{}, err
+	}
+
+	wr.Flush()
+	return b.Bytes(), nil
+}
+
 // f**ing long, but unique for each test
 func makePathname() string {
 	// get path

consensus/wal_test.go

@@ -4,11 +4,16 @@ import (
 	"bytes"
 	"crypto/rand"
 	"fmt"
+	"io/ioutil"
+	"os"
+	"path/filepath"
+
 	// "sync"
 	"testing"
 	"time"
 
 	"github.com/tendermint/tendermint/consensus/types"
+	"github.com/tendermint/tendermint/libs/autofile"
 	tmtypes "github.com/tendermint/tendermint/types"
 	tmtime "github.com/tendermint/tendermint/types/time"
 
@@ -16,6 +21,51 @@ import (
 	"github.com/stretchr/testify/require"
 )
 
+func TestWALTruncate(t *testing.T) {
+	walDir, err := ioutil.TempDir("", "wal")
+	if err != nil {
+		panic(fmt.Errorf("failed to create temp WAL file: %v", err))
+	}
+	defer os.RemoveAll(walDir)
+
+	walFile := filepath.Join(walDir, "wal")
+
+	//this magic number 4K can truncate the content when RotateFile. defaultHeadSizeLimit(10M) is hard to simulate.
+	//this magic number 1 * time.Millisecond make RotateFile check frequently. defaultGroupCheckDuration(5s) is hard to simulate.
+	wal, err := NewWAL(walFile, autofile.GroupHeadSizeLimit(4096), autofile.GroupCheckDuration(1*time.Millisecond))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	wal.Start()
+	defer wal.Stop()
+
+	//60 block's size nearly 70K, greater than group's headBuf size(4096 * 10), when headBuf is full, truncate content will Flush to the file.
+	//at this time, RotateFile is called, truncate content exist in each file.
+	err = WALGenerateNBlocks(wal.Group(), 60)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	time.Sleep(1 * time.Millisecond) //wait groupCheckDuration, make sure RotateFile run
+
+	wal.Group().Flush()
+
+	h := int64(50)
+	gr, found, err := wal.SearchForEndHeight(h, &WALSearchOptions{})
+	assert.NoError(t, err, fmt.Sprintf("expected not to err on height %d", h))
+	assert.True(t, found, fmt.Sprintf("expected to find end height for %d", h))
+	assert.NotNil(t, gr, "expected group not to be nil")
+	defer gr.Close()
+
+	dec := NewWALDecoder(gr)
+	msg, err := dec.Decode()
+	assert.NoError(t, err, "expected to decode a message")
+	rs, ok := msg.Msg.(tmtypes.EventDataRoundState)
+	assert.True(t, ok, "expected message of type EventDataRoundState")
+	assert.Equal(t, rs.Height, h+1, fmt.Sprintf("wrong height"))
+}
+
 func TestWALEncoderDecoder(t *testing.T) {
 	now := tmtime.Now()
 	msgs := []TimedWALMessage{