datamanager: implement data files splitting

split data files in multiple files of a max size (default 10Mib)
In this way every data snapshot will change only the datafiles that have some
changes instead of the whole single file.

internal/datamanager/data.go

@@ -20,21 +20,35 @@ import (
 	"encoding/json"
 	"fmt"
 	"io"
+	"sort"
 	"strings"
 
 	ostypes "github.com/sorintlab/agola/internal/objectstorage/types"
 	"github.com/sorintlab/agola/internal/sequence"
+
+	uuid "github.com/satori/go.uuid"
 	errors "golang.org/x/xerrors"
 )
 
+const (
+	DefaultMaxDataFileSize = 10 * 1024 * 1024
+)
+
 type DataStatus struct {
 	DataSequence string `json:"data_sequence,omitempty"`
 	WalSequence  string `json:"wal_sequence,omitempty"`
-	Files        map[string][]string `json:"files,omitempty"`
+	// an entry id ordered list of files for a specific data type (map key)
+	Files map[string][]*DataStatusFile `json:"files,omitempty"`
+}
+
+type DataStatusFile struct {
+	ID string `json:"id,omitempty"`
+	// the last entry id in this file
+	LastEntryID string `json:"last_entry_id,omitempty"`
 }
 
 type DataFileIndex struct {
-	Index map[string]int `json:"index,omitempty"`
+	Index map[string]int64 `json:"index,omitempty"`
 }
 
 type DataEntry struct {
@@ -47,105 +61,50 @@ func dataStatusPath(sequence string) string {
 	return fmt.Sprintf("%s/%s.status", storageDataDir, sequence)
 }
 
-func dataFileIndexPath(datatype, sequence string) string {
+func DataFileIndexPath(dataType, id string) string {
-	return fmt.Sprintf("%s/%s/%s.index", storageDataDir, datatype, sequence)
+	return fmt.Sprintf("%s/%s/%s.index", storageDataDir, dataType, id)
 }
 
-func dataFilePath(datatype, sequence string) string {
+func DataFilePath(dataType, id string) string {
-	return fmt.Sprintf("%s/%s/%s.data", storageDataDir, datatype, sequence)
+	return fmt.Sprintf("%s/%s/%s.data", storageDataDir, dataType, id)
 }
 
+// TODO(sgotti) this implementation could be heavily optimized to store less data in memory
+
 // TODO(sgotti)
 // split/merge data files at max N bytes (i.e 16MiB) so we'll rewrite only files
 // with changed data
 
-func (d *DataManager) writeData(ctx context.Context, wals []*WalData) error {
+// walIndex is a map of dataType of id of walEntry
-	dataSequence, err := sequence.IncSequence(ctx, d.e, etcdWalSeqKey)
+// TODO(sgotti) write this index to local disk (a temporary sqlite lite) instead of storing all in memory
-	if err != nil {
+type walIndex map[string]walActions
-		return err
-	}
 
-	for _, dataType := range d.dataTypes {
+// walDataEntries is an order by id list of data entries
-		if err := d.writeDataType(ctx, wals, dataType, dataSequence.String()); err != nil {
+type walActions []*Action
-			return err
-		}
-	}
 
-	var lastWalSequence string
+func (w walActions) Len() int           { return len(w) }
-	for _, walData := range wals {
+func (w walActions) Less(i, j int) bool { return w[i].ID < w[j].ID }
-		lastWalSequence = walData.WalSequence
+func (w walActions) Swap(i, j int)      { w[i], w[j] = w[j], w[i] }
-	}
 
-	dataStatus := &DataStatus{
+func (d *DataManager) walIndex(ctx context.Context, wals []*WalData) (walIndex, error) {
-		DataSequence: dataSequence.String(),
+	wimap := map[string]map[string]*Action{}
-		WalSequence:  lastWalSequence,
-		Files:        make(map[string][]string),
-	}
-	for _, dataType := range d.dataTypes {
-		dataStatus.Files[dataType] = []string{dataFilePath(dataType, dataSequence.String())}
-	}
-
-	dataStatusj, err := json.Marshal(dataStatus)
-	if err != nil {
-		return err
-	}
-	if err := d.ost.WriteObject(dataStatusPath(dataSequence.String()), bytes.NewReader(dataStatusj), int64(len(dataStatusj)), true); err != nil {
-		return err
-	}
-
-	return nil
-}
-
-func (d *DataManager) writeDataType(ctx context.Context, wals []*WalData, datatype, dataSequence string) error {
-	curDataStatus, err := d.GetLastDataStatus()
-	if err != nil && err != ostypes.ErrNotExist {
-		return err
-	}
-
-	dataEntriesMap := map[string]*DataEntry{}
-	if err != ostypes.ErrNotExist {
-		curDataSequence := curDataStatus.DataSequence
-
-		oldDataf, err := d.ost.ReadObject(dataFilePath(datatype, curDataSequence))
-		if err != nil && err != ostypes.ErrNotExist {
-			return err
-		}
-		if err != ostypes.ErrNotExist {
-			dec := json.NewDecoder(oldDataf)
-			for {
-				var de *DataEntry
-
-				err := dec.Decode(&de)
-				if err == io.EOF {
-					// all done
-					break
-				}
-				if err != nil {
-					oldDataf.Close()
-					return err
-				}
-				dataEntriesMap[de.ID] = de
-			}
-			oldDataf.Close()
-		}
-	}
 
 	for _, walData := range wals {
 		walFilef, err := d.ReadWal(walData.WalSequence)
 		if err != nil {
-			return err
+			return nil, err
 		}
 		dec := json.NewDecoder(walFilef)
 		var header *WalHeader
 		if err = dec.Decode(&header); err != nil && err != io.EOF {
 			walFilef.Close()
-			return err
+			return nil, err
 		}
 		walFilef.Close()
 
 		walFile, err := d.ReadWalData(header.WalDataFileID)
 		if err != nil {
-			return errors.Errorf("cannot read wal data file %q: %w", header.WalDataFileID, err)
+			return nil, errors.Errorf("cannot read wal data file %q: %w", header.WalDataFileID, err)
 		}
 		defer walFile.Close()
 
@@ -159,51 +118,92 @@ func (d *DataManager) writeDataType(ctx context.Context, wals []*WalData, dataty
 				break
 			}
 			if err != nil {
-				return errors.Errorf("failed to decode wal file: %w", err)
+				return nil, errors.Errorf("failed to decode wal file: %w", err)
-			}
-			if action.DataType != datatype {
-				continue
 			}
 
-			switch action.ActionType {
+			if _, ok := wimap[action.DataType]; !ok {
-			case ActionTypePut:
+				wimap[action.DataType] = map[string]*Action{}
-				de := &DataEntry{
-					ID:       action.ID,
-					DataType: action.DataType,
-					Data:     action.Data,
-				}
-				dataEntriesMap[de.ID] = de
-			case ActionTypeDelete:
-				delete(dataEntriesMap, action.ID)
 			}
+
+			// only keep the last action for every entry id
+			wimap[action.DataType][action.ID] = action
 		}
 	}
 
-	dataEntries := []*DataEntry{}
+	wi := map[string]walActions{}
-	for _, de := range dataEntriesMap {
+	for dataType, dd := range wimap {
-		dataEntries = append(dataEntries, de)
+		for _, de := range dd {
+			wi[dataType] = append(wi[dataType], de)
+		}
+		sort.Sort(wi[dataType])
 	}
 
-	dataFileIndex := &DataFileIndex{
+	return wi, nil
-		Index: make(map[string]int),
 }
 
-	var buf bytes.Buffer
+// writeDataSnapshot will create a new data snapshot merging the uncheckpointed
-	pos := 0
+// wals. It will split data files at maxDataFileSize bytes so we'll rewrite only
-	for _, de := range dataEntries {
+// files with changed data.
-		dataFileIndex.Index[de.ID] = pos
+// Only new files will be created, previous snapshot data files won't be touched
-
+//
-		dataEntryj, err := json.Marshal(de)
+// TODO(sgotti) add a function to merge small data files (i.e after deletions) to avoid fragmentation
+// TODO(sgotti) add a function to delete old data files keeping only N snapshots
+func (d *DataManager) writeDataSnapshot(ctx context.Context, wals []*WalData) error {
+	dataSequence, err := sequence.IncSequence(ctx, d.e, etcdCheckpointSeqKey)
 	if err != nil {
 		return err
 	}
-		if _, err := buf.Write(dataEntryj); err != nil {
+
+	var lastWalSequence string
+	for _, walData := range wals {
+		lastWalSequence = walData.WalSequence
+	}
+
+	dataStatus := &DataStatus{
+		DataSequence: dataSequence.String(),
+		WalSequence:  lastWalSequence,
+		Files:        make(map[string][]*DataStatusFile),
+	}
+
+	wi, err := d.walIndex(ctx, wals)
+	if err != nil {
 		return err
 	}
 
-		pos += len(dataEntryj)
+	curDataStatus, err := d.GetLastDataStatus()
+	if err != nil && err != ostypes.ErrNotExist {
+		return err
 	}
-	if err := d.ost.WriteObject(dataFilePath(datatype, dataSequence), &buf, int64(buf.Len()), true); err != nil {
+
+	for _, dataType := range d.dataTypes {
+		var curDataStatusFiles []*DataStatusFile
+		if curDataStatus != nil {
+			curDataStatusFiles = curDataStatus.Files[dataType]
+		}
+		dataStatusFiles, err := d.writeDataType(ctx, wi, dataType, curDataStatusFiles)
+		if err != nil {
+			return err
+		}
+		dataStatus.Files[dataType] = dataStatusFiles
+	}
+
+	dataStatusj, err := json.Marshal(dataStatus)
+	if err != nil {
+		return err
+	}
+	if err := d.ost.WriteObject(dataStatusPath(dataSequence.String()), bytes.NewReader(dataStatusj), int64(len(dataStatusj)), true); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func (d *DataManager) writeDataFile(ctx context.Context, buf *bytes.Buffer, size int64, dataFileIndex *DataFileIndex, dataFileID, dataType string) error {
+	if buf.Len() == 0 {
+		return fmt.Errorf("empty data entries")
+	}
+
+	if err := d.ost.WriteObject(DataFilePath(dataType, dataFileID), buf, size, true); err != nil {
 		return err
 	}
 
@@ -211,21 +211,293 @@ func (d *DataManager) writeDataType(ctx context.Context, wals []*WalData, dataty
 	if err != nil {
 		return err
 	}
-	if err := d.ost.WriteObject(dataFileIndexPath(datatype, dataSequence), bytes.NewReader(dataFileIndexj), int64(len(dataFileIndexj)), true); err != nil {
+	if err := d.ost.WriteObject(DataFileIndexPath(dataType, dataFileID), bytes.NewReader(dataFileIndexj), int64(len(dataFileIndexj)), true); err != nil {
 		return err
 	}
 
 	return nil
 }
 
+type ActionGroup struct {
+	DataStatusFile          *DataStatusFile
+	StartActionIndex        int
+	ActionsSize             int
+	PreviousDataStatusFiles []*DataStatusFile
+}
+
+func (d *DataManager) actionGroups(ctx context.Context, wi walIndex, dataType string, curDataStatusFiles []*DataStatusFile) ([]*ActionGroup, []*DataStatusFile) {
+	dataStatusFiles := []*DataStatusFile{}
+	remainingDataStatusFiles := []*DataStatusFile{}
+
+	actionGroups := []*ActionGroup{}
+
+	var startActionIndex int
+	var actionsSize int
+
+	var actionIndex int
+	var curDataStatusFileIndex int
+	for {
+		var action *Action
+		if actionIndex <= len(wi[dataType])-1 {
+			action = wi[dataType][actionIndex]
+		}
+
+		var curDataStatusFile *DataStatusFile
+		if curDataStatusFileIndex <= len(curDataStatusFiles)-1 {
+			curDataStatusFile = curDataStatusFiles[curDataStatusFileIndex]
+		}
+
+		if action == nil {
+			if actionsSize > 0 {
+				actionGroup := &ActionGroup{
+					DataStatusFile:          curDataStatusFile,
+					StartActionIndex:        startActionIndex,
+					ActionsSize:             actionsSize,
+					PreviousDataStatusFiles: dataStatusFiles,
+				}
+				actionGroups = append(actionGroups, actionGroup)
+				curDataStatusFileIndex++
+				if curDataStatusFileIndex <= len(curDataStatusFiles)-1 {
+					remainingDataStatusFiles = curDataStatusFiles[curDataStatusFileIndex:]
+				}
+			}
+			break
+		}
+
+		if curDataStatusFile != nil {
+			if curDataStatusFile.LastEntryID >= action.ID || curDataStatusFileIndex == len(curDataStatusFiles)-1 {
+				// continue using this status file
+				actionIndex++
+				actionsSize++
+			} else {
+				// find new status file
+				if actionsSize > 0 {
+					actionGroup := &ActionGroup{
+						DataStatusFile:          curDataStatusFile,
+						StartActionIndex:        startActionIndex,
+						ActionsSize:             actionsSize,
+						PreviousDataStatusFiles: dataStatusFiles,
+					}
+					actionGroups = append(actionGroups, actionGroup)
+
+					startActionIndex = actionIndex
+					actionsSize = 0
+					dataStatusFiles = []*DataStatusFile{}
+				} else {
+					dataStatusFiles = append(dataStatusFiles, curDataStatusFile)
+				}
+				curDataStatusFileIndex++
+			}
+		} else {
+			actionIndex++
+			actionsSize++
+		}
+	}
+
+	return actionGroups, remainingDataStatusFiles
+}
+
+func (d *DataManager) writeDataType(ctx context.Context, wi walIndex, dataType string, curDataStatusFiles []*DataStatusFile) ([]*DataStatusFile, error) {
+	type SplitPoint struct {
+		pos         int64
+		lastEntryID string
+	}
+
+	if len(wi[dataType]) == 0 {
+		// no actions
+		return curDataStatusFiles, nil
+	}
+	actionGroups, remainingDataStatusFiles := d.actionGroups(ctx, wi, dataType, curDataStatusFiles)
+
+	dataStatusFiles := []*DataStatusFile{}
+
+	for _, actionGroup := range actionGroups {
+		dataStatusFiles = append(dataStatusFiles, actionGroup.PreviousDataStatusFiles...)
+
+		splitPoints := []SplitPoint{}
+		dataFileIndexes := []*DataFileIndex{}
+		dataFileIndex := &DataFileIndex{
+			Index: make(map[string]int64),
+		}
+		dataEntries := []*DataEntry{}
+		var buf bytes.Buffer
+		var pos int64
+		var lastEntryID string
+
+		if actionGroup.DataStatusFile != nil {
+			// TODO(sgotti) instead of reading all entries in memory decode it's contents one by one when needed
+			oldDataf, err := d.ost.ReadObject(DataFilePath(dataType, actionGroup.DataStatusFile.ID))
+			if err != nil && err != ostypes.ErrNotExist {
+				return nil, err
+			}
+			if err != ostypes.ErrNotExist {
+				dec := json.NewDecoder(oldDataf)
+				for {
+					var de *DataEntry
+
+					err := dec.Decode(&de)
+					if err == io.EOF {
+						// all done
+						break
+					}
+					if err != nil {
+						oldDataf.Close()
+						return nil, err
+					}
+
+					dataEntries = append(dataEntries, de)
+				}
+				oldDataf.Close()
+			}
+		}
+
+		dataEntryIndex := 0
+		actionIndex := actionGroup.StartActionIndex
+
+		// iterate over data entries and action in order
+		for {
+			exists := false
+			useAction := false
+
+			var action *Action
+			if actionIndex < actionGroup.StartActionIndex+actionGroup.ActionsSize {
+				action = wi[dataType][actionIndex]
+			}
+
+			var de *DataEntry
+			if dataEntryIndex <= len(dataEntries)-1 {
+				de = dataEntries[dataEntryIndex]
+			}
+
+			if de == nil && action == nil {
+				break
+			}
+
+			if action != nil {
+				if de != nil {
+					if de.ID == action.ID {
+						exists = true
+						useAction = true
+					}
+					if de.ID > action.ID {
+						useAction = true
+					}
+				} else {
+					useAction = true
+				}
+
+				if useAction {
+					de = nil
+					switch action.ActionType {
+					case ActionTypePut:
+						de = &DataEntry{
+							ID:       action.ID,
+							DataType: action.DataType,
+							Data:     action.Data,
+						}
+						if exists {
+							// replace current data entry with the action data
+							dataEntryIndex++
+						}
+					case ActionTypeDelete:
+						if exists {
+							// skip current data entry
+							dataEntryIndex++
+						}
+					}
+					actionIndex++
+				} else {
+					dataEntryIndex++
+				}
+			} else {
+				dataEntryIndex++
+			}
+
+			if de != nil {
+				lastEntryID = de.ID
+				dataEntryj, err := json.Marshal(de)
+				if err != nil {
+					return nil, err
+				}
+				if _, err := buf.Write(dataEntryj); err != nil {
+					return nil, err
+				}
+				dataFileIndex.Index[de.ID] = pos
+				prevPos := pos
+				pos += int64(len(dataEntryj))
+				var lastSplitPos int64
+				if len(splitPoints) > 0 {
+					lastSplitPos = splitPoints[len(splitPoints)-1].pos
+				}
+				if pos-lastSplitPos > d.maxDataFileSize {
+					// add split point only if it's different (less) than the previous one
+					if lastSplitPos < prevPos {
+						splitPoints = append(splitPoints, SplitPoint{pos: int64(buf.Len()), lastEntryID: lastEntryID})
+						dataFileIndexes = append(dataFileIndexes, dataFileIndex)
+						dataFileIndex = &DataFileIndex{
+							Index: make(map[string]int64),
+						}
+					}
+				}
+			}
+		}
+
+		// save remaining data
+		if buf.Len() != 0 {
+			var curPos int64
+			var lastSplitPos int64
+			if len(splitPoints) > 0 {
+				lastSplitPos = splitPoints[len(splitPoints)-1].pos
+			}
+			// add final split point if there's something left in the buffer
+			if lastSplitPos != int64(buf.Len()) {
+				splitPoints = append(splitPoints, SplitPoint{pos: int64(buf.Len()), lastEntryID: lastEntryID})
+			}
+			dataFileIndexes = append(dataFileIndexes, dataFileIndex)
+			for i, sp := range splitPoints {
+				curDataFileID := uuid.NewV4().String()
+				if err := d.writeDataFile(ctx, &buf, sp.pos-curPos, dataFileIndexes[i], curDataFileID, dataType); err != nil {
+					return nil, err
+				}
+				// insert new dataStatusFile
+				dataStatusFiles = append(dataStatusFiles, &DataStatusFile{
+					ID:          curDataFileID,
+					LastEntryID: sp.lastEntryID,
+				})
+
+				curPos = sp.pos
+			}
+		}
+
+	}
+
+	dataStatusFiles = append(dataStatusFiles, remainingDataStatusFiles...)
+
+	return dataStatusFiles, nil
+}
+
 func (d *DataManager) Read(dataType, id string) (io.Reader, error) {
 	curDataStatus, err := d.GetLastDataStatus()
 	if err != nil {
 		return nil, err
 	}
-	dataSequence := curDataStatus.DataSequence
+	curFiles := curDataStatus.Files
 
-	dataFileIndexf, err := d.ost.ReadObject(dataFileIndexPath(dataType, dataSequence))
+	var matchingDataFileID string
+	// get the matching data file for the action entry ID
+	if len(curFiles[dataType]) == 0 {
+		return nil, ostypes.ErrNotExist
+	}
+
+	matchingDataFileID = curFiles[dataType][0].ID
+	for _, dataStatusFile := range curFiles[dataType] {
+		if dataStatusFile.LastEntryID > id {
+			matchingDataFileID = dataStatusFile.ID
+			break
+		}
+	}
+
+	dataFileIndexf, err := d.ost.ReadObject(DataFileIndexPath(dataType, matchingDataFileID))
 	if err != nil {
 		return nil, err
 	}
@@ -243,7 +515,7 @@ func (d *DataManager) Read(dataType, id string) (io.Reader, error) {
 		return nil, ostypes.ErrNotExist
 	}
 
-	dataf, err := d.ost.ReadObject(dataFilePath(dataType, dataSequence))
+	dataf, err := d.ost.ReadObject(DataFilePath(dataType, matchingDataFileID))
 	if err != nil {
 		return nil, err
 	}

internal/datamanager/datamanager.go

@@ -38,7 +38,7 @@ import (
 // wals/{walSeq}
 
 const (
-	DefaultCheckpointInterval   = 1 * time.Minute
+	DefaultCheckpointInterval   = 10 * time.Second
 	DefaultEtcdWalsKeepNum      = 100
 	DefaultMinCheckpointWalsNum = 100
 )
@@ -63,6 +63,8 @@ var (
 	etcdWalSeqKey                     = path.Join(etcdWalBaseDir, "walseq")
 	etcdLastCommittedStorageWalSeqKey = path.Join(etcdWalBaseDir, "lastcommittedstoragewalseq")
 
+	etcdCheckpointSeqKey = path.Join(etcdWalBaseDir, "checkpointseq")
+
 	etcdSyncLockKey       = path.Join(etcdWalBaseDir, "synclock")
 	etcdCheckpointLockKey = path.Join(etcdWalBaseDir, "checkpointlock")
 	etcdWalCleanerLockKey = path.Join(etcdWalBaseDir, "walcleanerlock")
@@ -88,6 +90,7 @@ type DataManagerConfig struct {
 	CheckpointInterval time.Duration
 	// MinCheckpointWalsNum is the minimum number of wals required before doing a checkpoint
 	MinCheckpointWalsNum int
+	MaxDataFileSize      int64
 }
 
 type DataManager struct {
@@ -100,6 +103,7 @@ type DataManager struct {
 	etcdWalsKeepNum      int
 	checkpointInterval   time.Duration
 	minCheckpointWalsNum int
+	maxDataFileSize      int64
 }
 
 func NewDataManager(ctx context.Context, logger *zap.Logger, conf *DataManagerConfig) (*DataManager, error) {
@@ -118,6 +122,9 @@ func NewDataManager(ctx context.Context, logger *zap.Logger, conf *DataManagerCo
 	if conf.MinCheckpointWalsNum < 1 {
 		return nil, errors.New("minCheckpointWalsNum must be greater than 0")
 	}
+	if conf.MaxDataFileSize == 0 {
+		conf.MaxDataFileSize = DefaultMaxDataFileSize
+	}
 
 	d := &DataManager{
 		basePath:             conf.BasePath,
@@ -129,6 +136,7 @@ func NewDataManager(ctx context.Context, logger *zap.Logger, conf *DataManagerCo
 		etcdWalsKeepNum:      conf.EtcdWalsKeepNum,
 		checkpointInterval:   conf.CheckpointInterval,
 		minCheckpointWalsNum: conf.MinCheckpointWalsNum,
+		maxDataFileSize:      conf.MaxDataFileSize,
 	}
 
 	// add trailing slash the basepath

internal/datamanager/datamanager_test.go

@@ -16,9 +16,13 @@ package datamanager
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
+	"io"
 	"io/ioutil"
 	"os"
+	"reflect"
+	"sort"
 	"testing"
 	"time"
 
@@ -147,7 +151,7 @@ func TestEtcdReset(t *testing.T) {
 	dm, err = NewDataManager(ctx, logger, dmConfig)
 	dmReadyCh = make(chan struct{})
 
-	t.Logf("starting wal")
+	t.Logf("starting datamanager")
 	go dm.Run(ctx, dmReadyCh)
 	<-dmReadyCh
 
@@ -293,8 +297,12 @@ func TestWalCleaner(t *testing.T) {
 		}
 	}
 
-	dm.checkpoint(ctx)
+	if err := dm.checkpoint(ctx); err != nil {
-	dm.walCleaner(ctx)
+		t.Fatalf("unexpected err: %v", err)
+	}
+	if err := dm.walCleaner(ctx); err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
 
 	walsCount := 0
 	for range dm.ListEtcdWals(ctx, 0) {
@@ -333,6 +341,12 @@ func TestReadObject(t *testing.T) {
 	}
 	dm, err := NewDataManager(ctx, logger, dmConfig)
 
+	dmReadyCh := make(chan struct{})
+	go dm.Run(ctx, dmReadyCh)
+	<-dmReadyCh
+
+	time.Sleep(5 * time.Second)
+
 	actions := []*Action{}
 	for i := 0; i < 20; i++ {
 		actions = append(actions, &Action{
@@ -343,12 +357,6 @@ func TestReadObject(t *testing.T) {
 		})
 	}
 
-	dmReadyCh := make(chan struct{})
-	go dm.Run(ctx, dmReadyCh)
-	<-dmReadyCh
-
-	time.Sleep(5 * time.Second)
-
 	// populate with a wal
 	_, err = dm.WriteWal(ctx, actions, nil)
 	if err != nil {
@@ -398,8 +406,12 @@ func TestReadObject(t *testing.T) {
 	}
 
 	// do a checkpoint and wal clean
-	dm.checkpoint(ctx)
+	if err := dm.checkpoint(ctx); err != nil {
-	dm.walCleaner(ctx)
+		t.Fatalf("unexpected err: %v", err)
+	}
+	if err := dm.walCleaner(ctx); err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
 
 	// wait for the event to be read
 	time.Sleep(500 * time.Millisecond)
@@ -417,3 +429,348 @@ func TestReadObject(t *testing.T) {
 		t.Fatalf("unexpected err: %v", err)
 	}
 }
+
+func testCheckpoint(t *testing.T, ctx context.Context, dm *DataManager, actionGroups [][]*Action, currentEntries map[string]*DataEntry) (map[string]*DataEntry, error) {
+	expectedEntries := map[string]*DataEntry{}
+	for _, e := range currentEntries {
+		expectedEntries[e.ID] = e
+	}
+
+	for _, actionGroup := range actionGroups {
+		for _, action := range actionGroup {
+			switch action.ActionType {
+			case ActionTypePut:
+				expectedEntries[action.ID] = &DataEntry{ID: action.ID, DataType: action.DataType, Data: action.Data}
+			case ActionTypeDelete:
+				delete(expectedEntries, action.ID)
+			}
+		}
+	}
+
+	for _, actionGroup := range actionGroups {
+		// populate with a wal
+		_, err := dm.WriteWal(ctx, actionGroup, nil)
+		if err != nil {
+			return nil, err
+		}
+	}
+
+	// wait for the event to be read
+	time.Sleep(500 * time.Millisecond)
+
+	// do a checkpoint
+	if err := dm.checkpoint(ctx); err != nil {
+		return nil, err
+	}
+
+	if err := checkDataFiles(ctx, t, dm, expectedEntries); err != nil {
+		return nil, err
+	}
+
+	return expectedEntries, nil
+}
+
+// TODO(sgotti) some fuzzy testing will be really good
+func TestCheckpoint(t *testing.T) {
+	dir, err := ioutil.TempDir("", "agola")
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+	defer os.RemoveAll(dir)
+
+	etcdDir, err := ioutil.TempDir(dir, "etcd")
+	tetcd := setupEtcd(t, etcdDir)
+	defer shutdownEtcd(tetcd)
+
+	ctx := context.Background()
+
+	ostDir, err := ioutil.TempDir(dir, "ost")
+	ost, err := posix.New(ostDir)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	dmConfig := &DataManagerConfig{
+		E:   tetcd.TestEtcd.Store,
+		OST: objectstorage.NewObjStorage(ost, "/"),
+		// remove almost all wals to see that they are removed also from changes
+		EtcdWalsKeepNum: 1,
+		DataTypes:       []string{"datatype01"},
+		// checkpoint also with only one wal
+		MinCheckpointWalsNum: 1,
+		// use a small maxDataFileSize
+		MaxDataFileSize: 10 * 1024,
+	}
+	dm, err := NewDataManager(ctx, logger, dmConfig)
+	dmReadyCh := make(chan struct{})
+	go dm.Run(ctx, dmReadyCh)
+	<-dmReadyCh
+
+	time.Sleep(5 * time.Second)
+
+	contents := "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
+
+	// test insert from scratch (no current entries)
+	actions := []*Action{}
+	for i := 200; i < 400; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err := testCheckpoint(t, ctx, dm, [][]*Action{actions}, nil)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test delete of all existing entries
+	actions = []*Action{}
+	for i := 200; i < 400; i++ {
+		actions = append(actions, &Action{
+			ActionType: ActionTypeDelete,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+		})
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test insert from scratch again (no current entries)
+	actions = []*Action{}
+	for i := 200; i < 400; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test delete some existing entries in the middle
+	actions = []*Action{}
+	for i := 250; i < 350; i++ {
+		action := &Action{
+			ActionType: ActionTypeDelete,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test delete of unexisting entries
+	actions = []*Action{}
+	for i := 1000; i < 1010; i++ {
+		action := &Action{
+			ActionType: ActionTypeDelete,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test update and insert at the end
+	actions = []*Action{}
+	for i := 300; i < 500; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test update and insert at the start
+	actions = []*Action{}
+	for i := 0; i < 300; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, [][]*Action{actions}, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+
+	// test multiple wals with different insert, updated, deletes
+	actionGroups := [][]*Action{}
+	for i := 0; i < 150; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+	actionGroups = append(actionGroups, actions)
+	for i := 50; i < 100; i++ {
+		action := &Action{
+			ActionType: ActionTypeDelete,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+	actionGroups = append(actionGroups, actions)
+	for i := 250; i < 300; i++ {
+		action := &Action{
+			ActionType: ActionTypeDelete,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+	for i := 70; i < 80; i++ {
+		action := &Action{
+			ActionType: ActionTypePut,
+			ID:         fmt.Sprintf("object%04d", i),
+			DataType:   "datatype01",
+			Data:       []byte(fmt.Sprintf(`{ "ID": "%d", "Contents": %s }`, i, contents)),
+		}
+		actions = append(actions, action)
+	}
+	actionGroups = append(actionGroups, actions)
+
+	currentEntries, err = testCheckpoint(t, ctx, dm, actionGroups, currentEntries)
+	if err != nil {
+		t.Fatalf("unexpected err: %v", err)
+	}
+}
+
+func checkDataFiles(ctx context.Context, t *testing.T, dm *DataManager, expectedEntriesMap map[string]*DataEntry) error {
+	// read the data file
+	curDataStatus, err := dm.GetLastDataStatus()
+	if err != nil {
+		return err
+	}
+
+	allEntriesMap := map[string]*DataEntry{}
+	var prevLastEntryID string
+
+	for i, file := range curDataStatus.Files["datatype01"] {
+		dataFileIndexf, err := dm.ost.ReadObject(DataFileIndexPath("datatype01", file.ID))
+		if err != nil {
+			return err
+		}
+		var dataFileIndex *DataFileIndex
+		dec := json.NewDecoder(dataFileIndexf)
+		err = dec.Decode(&dataFileIndex)
+		if err != nil {
+			dataFileIndexf.Close()
+			return err
+		}
+
+		dataFileIndexf.Close()
+		dataEntriesMap := map[string]*DataEntry{}
+		dataEntries := []*DataEntry{}
+		dataf, err := dm.ost.ReadObject(DataFilePath("datatype01", file.ID))
+		if err != nil {
+			return err
+		}
+		dec = json.NewDecoder(dataf)
+		var prevEntryID string
+		for {
+			var de *DataEntry
+
+			err := dec.Decode(&de)
+			if err == io.EOF {
+				// all done
+				break
+			}
+			if err != nil {
+				dataf.Close()
+				return err
+			}
+			// check that there are no duplicate entries
+			if _, ok := allEntriesMap[de.ID]; ok {
+				return fmt.Errorf("duplicate entry id: %s", de.ID)
+			}
+			// check that the entries are in order
+			if de.ID < prevEntryID {
+				return fmt.Errorf("previous entry id: %s greater than entry id: %s", prevEntryID, de.ID)
+			}
+
+			dataEntriesMap[de.ID] = de
+			dataEntries = append(dataEntries, de)
+			allEntriesMap[de.ID] = de
+		}
+		dataf.Close()
+
+		// check that the index matches the entries
+		if len(dataFileIndex.Index) != len(dataEntriesMap) {
+			return fmt.Errorf("index entries: %d different than data entries: %d", len(dataFileIndex.Index), len(dataEntriesMap))
+		}
+		indexIDs := make([]string, len(dataFileIndex.Index))
+		entriesIDs := make([]string, len(dataEntriesMap))
+		for id := range dataFileIndex.Index {
+			indexIDs = append(indexIDs, id)
+		}
+		for id := range dataEntriesMap {
+			entriesIDs = append(entriesIDs, id)
+		}
+		sort.Strings(indexIDs)
+		sort.Strings(entriesIDs)
+		if !reflect.DeepEqual(indexIDs, entriesIDs) {
+			return fmt.Errorf("index entries ids don't match data entries ids: index: %v, data: %v", indexIDs, entriesIDs)
+		}
+
+		if file.LastEntryID != dataEntries[len(dataEntries)-1].ID {
+			return fmt.Errorf("lastEntryID for datafile %d: %s is different than real last entry id: %s", i, file.LastEntryID, dataEntries[len(dataEntries)-1].ID)
+		}
+
+		// check that all the files are in order
+		if file.LastEntryID == prevLastEntryID {
+			return fmt.Errorf("lastEntryID for datafile %d is equal than previous file lastEntryID: %s == %s", i, file.LastEntryID, prevLastEntryID)
+		}
+		if file.LastEntryID < prevLastEntryID {
+			return fmt.Errorf("lastEntryID for datafile %d is less than previous file lastEntryID: %s < %s", i, file.LastEntryID, prevLastEntryID)
+		}
+		prevLastEntryID = file.LastEntryID
+	}
+
+	// check that the number of entries is right
+	if len(allEntriesMap) != len(expectedEntriesMap) {
+		return fmt.Errorf("expected %d total entries, got %d", len(expectedEntriesMap), len(allEntriesMap))
+	}
+	if !reflect.DeepEqual(expectedEntriesMap, allEntriesMap) {
+		return fmt.Errorf("expected entries don't match current entries")
+	}
+
+	return nil
+}

internal/datamanager/wal.go

@@ -723,7 +723,7 @@ func (d *DataManager) checkpoint(ctx context.Context) error {
 		return nil
 	}
 
-	if err := d.writeData(ctx, walsData); err != nil {
+	if err := d.writeDataSnapshot(ctx, walsData); err != nil {
 		return errors.Errorf("checkpoint function error: %w", err)
 	}
 

internal/services/configstore/readdb/readdb.go

@@ -133,7 +133,8 @@ func (r *ReadDB) SyncFromDump() (string, error) {
 		return "", nil
 	}
 	for dataType, files := range dumpIndex.Files {
-		dumpf, err := r.ost.ReadObject(files[0])
+		for _, file := range files {
+			dumpf, err := r.ost.ReadObject(datamanager.DataFilePath(dataType, file.ID))
 			if err != nil {
 				return "", err
 			}
@@ -173,6 +174,7 @@ func (r *ReadDB) SyncFromDump() (string, error) {
 				return "", err
 			}
 		}
+	}
 
 	return dumpIndex.WalSequence, nil
 }

internal/services/runservice/readdb/readdb.go

@@ -660,7 +660,8 @@ func (r *ReadDB) SyncFromDump() (string, error) {
 		return "", nil
 	}
 	for dataType, files := range dumpIndex.Files {
-		dumpf, err := r.ost.ReadObject(files[0])
+		for _, file := range files {
+			dumpf, err := r.ost.ReadObject(datamanager.DataFilePath(dataType, file.ID))
 			if err != nil {
 				return "", err
 			}
@@ -700,6 +701,7 @@ func (r *ReadDB) SyncFromDump() (string, error) {
 				return "", err
 			}
 		}
+	}
 
 	return dumpIndex.WalSequence, nil
 }