Compact database during connectBlocksParallel

blockbook.go

@@ -113,7 +113,7 @@ func main() {
 	}
 
 	if *synchronize {
-		if err := resyncIndex(); err != nil {
+		if err := resyncIndex(true); err != nil {
 			glog.Fatal("resyncIndex ", err)
 		}
 		go syncIndexLoop()
@@ -219,7 +219,7 @@ func syncIndexLoop() {
 	glog.Info("syncIndexLoop starting")
 	// resync index about every 15 minutes if there are no chanSyncIndex requests, with debounce 1 second
 	tickAndDebounce(935093*time.Millisecond, 1009*time.Millisecond, chanSyncIndex, func() {
-		if err := resyncIndex(); err != nil {
+		if err := resyncIndex(false); err != nil {
 			glog.Error("syncIndexLoop", err)
 		}
 	})
@@ -279,7 +279,7 @@ func printResult(txid string, vout uint32, isOutput bool) error {
 	return nil
 }
 
-func resyncIndex() error {
+func resyncIndex(bulk bool) error {
 	remote, err := chain.GetBestBlockHash()
 	if err != nil {
 		return err
@@ -334,7 +334,7 @@ func resyncIndex() error {
 			if err != nil {
 				return err
 			}
-			return resyncIndex()
+			return resyncIndex(false)
 		}
 	}
 
@@ -359,7 +359,7 @@ func resyncIndex() error {
 
 	// if parallel operation is enabled and the number of blocks to be connected is large,
 	// use parallel routine to load majority of blocks
-	if *syncWorkers > 1 {
+	if bulk && *syncWorkers > 1 {
 		chainBestHeight, err := chain.GetBestBlockHeight()
 		if err != nil {
 			return err
@@ -376,7 +376,7 @@ func resyncIndex() error {
 			}
 			// after parallel load finish the sync using standard way,
 			// new blocks may have been created in the meantime
-			return resyncIndex()
+			return resyncIndex(false)
 		}
 	}
 
@@ -418,27 +418,31 @@ func connectBlocksParallel(
 ) error {
 	var wg sync.WaitGroup
 	hch := make(chan string, numWorkers)
-
-	work := func() {
+	running := make([]bool, numWorkers)
+	work := func(i int) {
 		defer wg.Done()
 		for hash := range hch {
+			running[i] = true
 			block, err := chain.GetBlock(hash)
 			if err != nil {
 				glog.Error("Connect block ", hash, " error ", err)
+				running[i] = false
 				continue
 			}
 			if *dryRun {
+				running[i] = false
 				continue
 			}
 			err = index.ConnectBlock(block)
 			if err != nil {
 				glog.Error("Connect block ", hash, " error ", err)
 			}
+			running[i] = false
 		}
 	}
 	for i := 0; i < numWorkers; i++ {
 		wg.Add(1)
-		go work()
+		go work(i)
 	}
 	var (
 		err  error
@@ -450,8 +454,25 @@ func connectBlocksParallel(
 			break
 		}
 		hch <- hash
-		if h%1000 == 0 {
+		if h > 0 && h%1000 == 0 {
 			glog.Info("connecting block ", h, " ", hash)
+			if h%50000 == 0 {
+				// wait for the workers to finish block
+			WaitAgain:
+				for {
+					for _, r := range running {
+						if r {
+							glog.Info("Waiting ", running)
+							time.Sleep(time.Millisecond * 500)
+							continue WaitAgain
+						}
+					}
+					break
+				}
+				if err = index.CompactDatabase(); err != nil {
+					break
+				}
+			}
 		}
 	}
 	close(hch)

db/rocksdb.go

@@ -6,6 +6,8 @@ import (
 	"encoding/binary"
 	"encoding/hex"
 	"errors"
+	"os"
+	"path/filepath"
 
 	"github.com/bsm/go-vlq"
 	"github.com/golang/glog"
@@ -21,10 +23,11 @@ func RepairRocksDB(name string) error {
 
 // RocksDB handle
 type RocksDB struct {
-	db  *gorocksdb.DB
-	wo  *gorocksdb.WriteOptions
-	ro  *gorocksdb.ReadOptions
-	cfh []*gorocksdb.ColumnFamilyHandle
+	path string
+	db   *gorocksdb.DB
+	wo   *gorocksdb.WriteOptions
+	ro   *gorocksdb.ReadOptions
+	cfh  []*gorocksdb.ColumnFamilyHandle
 }
 
 const (
@@ -36,11 +39,7 @@ const (
 
 var cfNames = []string{"default", "height", "outputs", "inputs"}
 
-// NewRocksDB opens an internal handle to RocksDB environment.  Close
-// needs to be called to release it.
-func NewRocksDB(path string) (d *RocksDB, err error) {
-	glog.Infof("rocksdb: open %s", path)
-
+func openDB(path string) (*gorocksdb.DB, []*gorocksdb.ColumnFamilyHandle, error) {
 	fp := gorocksdb.NewBloomFilter(10)
 	bbto := gorocksdb.NewDefaultBlockBasedTableOptions()
 	bbto.SetBlockSize(16 << 10)                        // 16kb
@@ -61,25 +60,37 @@ func NewRocksDB(path string) (d *RocksDB, err error) {
 
 	db, cfh, err := gorocksdb.OpenDbColumnFamilies(opts, path, cfNames, fcOptions)
 	if err != nil {
-		return
+		return nil, nil, err
 	}
+	return db, cfh, nil
 
+}
+
+// NewRocksDB opens an internal handle to RocksDB environment.  Close
+// needs to be called to release it.
+func NewRocksDB(path string) (d *RocksDB, err error) {
+	glog.Infof("rocksdb: open %s", path)
+	db, cfh, err := openDB(path)
 	wo := gorocksdb.NewDefaultWriteOptions()
 	ro := gorocksdb.NewDefaultReadOptions()
 	ro.SetFillCache(false)
+	return &RocksDB{path, db, wo, ro, cfh}, nil
+}
 
-	return &RocksDB{db, wo, ro, cfh}, nil
+func (d *RocksDB) closeDB() error {
+	for _, h := range d.cfh {
+		h.Destroy()
+	}
+	d.db.Close()
+	return nil
 }
 
 // Close releases the RocksDB environment opened in NewRocksDB.
 func (d *RocksDB) Close() error {
 	glog.Infof("rocksdb: close")
-	for _, h := range d.cfh {
-		h.Destroy()
-	}
+	d.closeDB()
 	d.wo.Destroy()
 	d.ro.Destroy()
-	d.db.Close()
 	return nil
 }
 
@@ -448,6 +459,39 @@ func (d *RocksDB) DisconnectBlocks(
 	return err
 }
 
+func dirSize(path string) (int64, error) {
+	var size int64
+	err := filepath.Walk(path, func(_ string, info os.FileInfo, err error) error {
+		if !info.IsDir() {
+			size += info.Size()
+		}
+		return err
+	})
+	return size, err
+}
+
+// CompactDatabase compacts the database
+// After unsuccessful experiment with CompactRange method (slow and actually fragmenting the db without compacting)
+// the method now closes the db instance and opens it again.
+// This means that during compact nobody can access the dababase!
+func (d *RocksDB) CompactDatabase() error {
+	size, _ := dirSize(d.path)
+	glog.Info("Compacting database, db size ", size)
+	err := d.closeDB()
+	if err != nil {
+		return err
+	}
+	d.db = nil
+	db, cfh, err := openDB(d.path)
+	if err != nil {
+		return err
+	}
+	d.db, d.cfh = db, cfh
+	size, _ = dirSize(d.path)
+	glog.Info("Compacting database finished, db size ", size)
+	return nil
+}
+
 // Helpers
 
 const txIdUnpackedLen = 32