/* * Minio Cloud Storage, (C) 2016 Minio, Inc. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package main import ( "fmt" "os" slashpath "path" "sort" "strings" "sync" "github.com/Sirupsen/logrus" "github.com/klauspost/reedsolomon" ) const ( // Part metadata file. metadataFile = "part.json" // Maximum erasure blocks. maxErasureBlocks = 16 ) // XL layer structure. type XL struct { ReedSolomon reedsolomon.Encoder // Erasure encoder/decoder. DataBlocks int ParityBlocks int storageDisks []StorageAPI nameSpaceLockMap map[nameSpaceParam]*nameSpaceLock nameSpaceLockMapMutex *sync.Mutex readQuorum int writeQuorum int } // lockNS - locks the given resource, using a previously allocated // name space lock or initializing a new one. func (xl XL) lockNS(volume, path string, readLock bool) { xl.nameSpaceLockMapMutex.Lock() defer xl.nameSpaceLockMapMutex.Unlock() param := nameSpaceParam{volume, path} nsLock, found := xl.nameSpaceLockMap[param] if !found { nsLock = newNSLock() } if readLock { nsLock.RLock() } else { nsLock.Lock() } xl.nameSpaceLockMap[param] = nsLock } // unlockNS - unlocks any previously acquired read or write locks. func (xl XL) unlockNS(volume, path string, readLock bool) { xl.nameSpaceLockMapMutex.Lock() defer xl.nameSpaceLockMapMutex.Unlock() param := nameSpaceParam{volume, path} if nsLock, found := xl.nameSpaceLockMap[param]; found { if readLock { nsLock.RUnlock() } else { nsLock.Unlock() } if nsLock.InUse() { xl.nameSpaceLockMap[param] = nsLock } } } // newXL instantiate a new XL. func newXL(disks ...string) (StorageAPI, error) { // Initialize XL. xl := &XL{} // Verify disks. totalDisks := len(disks) if totalDisks > maxErasureBlocks { return nil, errMaxDisks } // isEven function to verify if a given number if even. isEven := func(number int) bool { return number%2 == 0 } // TODO: verify if this makes sense in future. if !isEven(totalDisks) { return nil, errNumDisks } // Calculate data and parity blocks. dataBlocks, parityBlocks := totalDisks/2, totalDisks/2 // Initialize reed solomon encoding. rs, err := reedsolomon.New(dataBlocks, parityBlocks) if err != nil { return nil, err } // Save the reedsolomon. xl.DataBlocks = dataBlocks xl.ParityBlocks = parityBlocks xl.ReedSolomon = rs // Initialize all storage disks. storageDisks := make([]StorageAPI, len(disks)) for index, disk := range disks { var err error storageDisks[index], err = newFS(disk) if err != nil { return nil, err } } // Save all the initialized storage disks. xl.storageDisks = storageDisks // Initialize name space lock map. xl.nameSpaceLockMap = make(map[nameSpaceParam]*nameSpaceLock) xl.nameSpaceLockMapMutex = &sync.Mutex{} // Figure out read and write quorum based on number of storage disks. // Read quorum should be always N/2 + 1 (due to Vandermonde matrix // erasure requirements) xl.readQuorum = len(xl.storageDisks)/2 + 1 // Write quorum is assumed if we have total disks + 3 // parity. (Need to discuss this again) xl.writeQuorum = len(xl.storageDisks)/2 + 3 if xl.writeQuorum > len(xl.storageDisks) { xl.writeQuorum = len(xl.storageDisks) } // Return successfully initialized. return xl, nil } // MakeVol - make a volume. func (xl XL) MakeVol(volume string) error { if !isValidVolname(volume) { return errInvalidArgument } // Collect if all disks report volume exists. var volumeExistsMap = make(map[int]struct{}) // Make a volume entry on all underlying storage disks. for index, disk := range xl.storageDisks { if err := disk.MakeVol(volume); err != nil { log.WithFields(logrus.Fields{ "volume": volume, }).Errorf("MakeVol failed with %s", err) // We ignore error if errVolumeExists and creating a volume again. if err == errVolumeExists { volumeExistsMap[index] = struct{}{} continue } return err } } // Return err if all disks report volume exists. if len(volumeExistsMap) == len(xl.storageDisks) { return errVolumeExists } return nil } // DeleteVol - delete a volume. func (xl XL) DeleteVol(volume string) error { if !isValidVolname(volume) { return errInvalidArgument } // Collect if all disks report volume not found. var volumeNotFoundMap = make(map[int]struct{}) // Remove a volume entry on all underlying storage disks. for index, disk := range xl.storageDisks { if err := disk.DeleteVol(volume); err != nil { log.WithFields(logrus.Fields{ "volume": volume, }).Errorf("DeleteVol failed with %s", err) // We ignore error if errVolumeNotFound. if err == errVolumeNotFound { volumeNotFoundMap[index] = struct{}{} continue } return err } } // Return err if all disks report volume not found. if len(volumeNotFoundMap) == len(xl.storageDisks) { return errVolumeNotFound } return nil } // ListVols - list volumes. func (xl XL) ListVols() (volsInfo []VolInfo, err error) { emptyCount := 0 // Success vols map carries successful results of ListVols from // each disks. var successVolsMap = make(map[int][]VolInfo) for index, disk := range xl.storageDisks { var vlsInfo []VolInfo vlsInfo, err = disk.ListVols() if err == nil { if len(vlsInfo) == 0 { emptyCount++ } else { successVolsMap[index] = vlsInfo } } } // If all list operations resulted in an empty count which is same // as your total storage disks, then it is a valid case return // success with empty vols. if emptyCount == len(xl.storageDisks) { return []VolInfo{}, nil } else if len(successVolsMap) < xl.readQuorum { // If there is data and not empty, then we attempt quorum verification. // Verify if we have enough quorum to list vols. return nil, errReadQuorum } // Loop through success vols map and return the first value. for index := range xl.storageDisks { if _, ok := successVolsMap[index]; ok { volsInfo = successVolsMap[index] break } } return volsInfo, nil } // StatVol - get volume stat info. func (xl XL) StatVol(volume string) (volInfo VolInfo, err error) { if !isValidVolname(volume) { return VolInfo{}, errInvalidArgument } var statVols []VolInfo volumeNotFoundErrCnt := 0 for _, disk := range xl.storageDisks { volInfo, err = disk.StatVol(volume) if err == nil { // Collect all the successful attempts to verify quorum // subsequently. statVols = append(statVols, volInfo) } else if err == errVolumeNotFound { // Count total amount of volume not found errors. volumeNotFoundErrCnt++ } else if err != nil { log.WithFields(logrus.Fields{ "volume": volume, }).Errorf("StatVol failed with %s", err) return VolInfo{}, err } } // If volume not found err count is same as total storage disks, we // really don't have the bucket, report a valid error. if volumeNotFoundErrCnt == len(xl.storageDisks) { return VolInfo{}, errVolumeNotFound } else if len(statVols) < xl.readQuorum { // If one of the disks have bucket we need to validate if we // have read quorum, if not fail. return VolInfo{}, errReadQuorum } // If successful remove all the duplicates and keep the latest one. volInfo = removeDuplicateVols(statVols)[0] return volInfo, nil } // isLeafDirectory - check if a given path is leaf directory. i.e // there are no more directories inside it. Erasure code backend // format it means that the parent directory is the actual object name. func isLeafDirectory(disk StorageAPI, volume, leafPath string) (isLeaf bool) { var markerPath string var xlListCount = 1000 // Count page. for { fileInfos, eof, err := disk.ListFiles(volume, leafPath, markerPath, false, xlListCount) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "leafPath": leafPath, "markerPath": markerPath, "recursive": false, "count": xlListCount, }).Errorf("ListFiles failed with %s", err) break } for _, fileInfo := range fileInfos { if fileInfo.Mode.IsDir() { // Directory found, not a leaf directory, return right here. return false } } if eof { break } // MarkerPath to get the next set of files. markerPath = fileInfos[len(fileInfos)-1].Name } // Exhausted all the entries, no directories found must be leaf // return right here. return true } // extractMetadata - extract file metadata. func extractMetadata(disk StorageAPI, volume, path string) (fileMetadata, error) { metadataFilePath := slashpath.Join(path, metadataFile) // We are not going to read partial data from metadata file, // read the whole file always. offset := int64(0) metadataReader, err := disk.ReadFile(volume, metadataFilePath, offset) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": metadataFilePath, "offset": offset, }).Errorf("ReadFile failed with %s", err) return nil, err } // Close metadata reader. defer metadataReader.Close() metadata, err := fileMetadataDecode(metadataReader) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": metadataFilePath, "offset": offset, }).Errorf("fileMetadataDecode failed with %s", err) return nil, err } return metadata, nil } // Extract file info from paths. func extractFileInfo(disk StorageAPI, volume, path string) (FileInfo, error) { fileInfo := FileInfo{} fileInfo.Volume = volume fileInfo.Name = path metadata, err := extractMetadata(disk, volume, path) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("extractMetadata failed with %s", err) return FileInfo{}, err } fileSize, err := metadata.GetSize() if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("GetSize failed with %s", err) return FileInfo{}, err } fileModTime, err := metadata.GetModTime() if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("GetModTime failed with %s", err) return FileInfo{}, err } fileInfo.Size = fileSize fileInfo.Mode = os.FileMode(0644) fileInfo.ModTime = fileModTime return fileInfo, nil } // byFileInfoName is a collection satisfying sort.Interface. type byFileInfoName []FileInfo func (d byFileInfoName) Len() int { return len(d) } func (d byFileInfoName) Swap(i, j int) { d[i], d[j] = d[j], d[i] } func (d byFileInfoName) Less(i, j int) bool { return d[i].Name < d[j].Name } // ListFiles files at prefix. func (xl XL) ListFiles(volume, prefix, marker string, recursive bool, count int) (filesInfo []FileInfo, eof bool, err error) { if !isValidVolname(volume) { return nil, true, errInvalidArgument } // TODO: Fix: If readQuorum is met, its assumed that disks are in consistent file list. // exclude disks those are not in consistent file list and check count of remaining disks // are met readQuorum. // Treat empty file list specially emptyCount := 0 errCount := 0 successCount := 0 var firstFilesInfo []FileInfo var firstEOF bool var firstErr error for _, disk := range xl.storageDisks { if filesInfo, eof, err = listFiles(disk, volume, prefix, marker, recursive, count); err == nil { // we need to return first successful result if firstFilesInfo == nil { firstFilesInfo = filesInfo firstEOF = eof } if len(filesInfo) == 0 { emptyCount++ } else { successCount++ } } else { if firstErr == nil { firstErr = err } errCount++ } } if errCount >= xl.readQuorum { return nil, false, firstErr } else if successCount >= xl.readQuorum { return firstFilesInfo, firstEOF, nil } else if emptyCount >= xl.readQuorum { return []FileInfo{}, true, nil } return nil, false, errReadQuorum } func listFiles(disk StorageAPI, volume, prefix, marker string, recursive bool, count int) (filesInfo []FileInfo, eof bool, err error) { var fsFilesInfo []FileInfo var markerPath = marker if marker != "" { isLeaf := isLeafDirectory(disk, volume, retainSlash(marker)) if isLeaf { // For leaf for now we just point to the first block, make it // dynamic in future based on the availability of storage disks. markerPath = slashpath.Join(marker, metadataFile) } } // Loop and capture the proper fileInfos, requires extraction and // separation of XL related metadata information. for { fsFilesInfo, eof, err = disk.ListFiles(volume, prefix, markerPath, recursive, count) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "prefix": prefix, "marker": markerPath, "recursive": recursive, "count": count, }).Errorf("ListFiles failed with %s", err) return nil, true, err } for _, fsFileInfo := range fsFilesInfo { // Skip metadata files. if strings.HasSuffix(fsFileInfo.Name, metadataFile) { continue } var fileInfo FileInfo var isLeaf bool if fsFileInfo.Mode.IsDir() { isLeaf = isLeafDirectory(disk, volume, fsFileInfo.Name) } if isLeaf || !fsFileInfo.Mode.IsDir() { // Extract the parent of leaf directory or file to get the // actual name. path := slashpath.Dir(fsFileInfo.Name) fileInfo, err = extractFileInfo(disk, volume, path) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("extractFileInfo failed with %s", err) // For a leaf directory, if err is FileNotFound then // perhaps has a missing metadata. Ignore it and let // healing finish its job it will become available soon. if err == errFileNotFound { continue } // For any other errors return to the caller. return nil, true, err } } else { fileInfo = fsFileInfo } filesInfo = append(filesInfo, fileInfo) count-- if count == 0 { break } } if len(fsFilesInfo) > 0 { // markerPath for the next disk.ListFiles() iteration. markerPath = fsFilesInfo[len(fsFilesInfo)-1].Name } if count == 0 && recursive && !strings.HasSuffix(markerPath, metadataFile) { // If last entry is not part.json then loop once more to check if we // have reached eof. fsFilesInfo, eof, err = disk.ListFiles(volume, prefix, markerPath, recursive, 1) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "prefix": prefix, "marker": markerPath, "recursive": recursive, "count": 1, }).Errorf("ListFiles failed with %s", err) return nil, true, err } if !eof { // part.N and part.json are always in pairs and hence this // entry has to be part.json. If not better to manually investigate // and fix it. // For the next ListFiles() call we can safely assume that the // marker is "object/part.json" if !strings.HasSuffix(fsFilesInfo[0].Name, metadataFile) { log.WithFields(logrus.Fields{ "volume": volume, "prefix": prefix, "fsFileInfo.Name": fsFilesInfo[0].Name, }).Errorf("ListFiles failed with %s, expected %s to be a part.json file.", err, fsFilesInfo[0].Name) return nil, true, errUnexpected } } } if count == 0 || eof { break } } // Sort to make sure we sort entries back properly. sort.Sort(byFileInfoName(filesInfo)) return filesInfo, eof, nil } // Object API. // StatFile - stat a file func (xl XL) StatFile(volume, path string) (FileInfo, error) { if !isValidVolname(volume) { return FileInfo{}, errInvalidArgument } if !isValidPath(path) { return FileInfo{}, errInvalidArgument } // Acquire read lock. readLock := true xl.lockNS(volume, path, readLock) _, metadata, heal, err := xl.listOnlineDisks(volume, path) xl.unlockNS(volume, path, readLock) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("listOnlineDisks failed with %s", err) return FileInfo{}, err } if heal { // Heal in background safely, since we already have read quorum disks. go func() { if err = xl.healFile(volume, path); err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("healFile failed with %s", err) return } }() } // Extract metadata. size, err := metadata.GetSize() if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("GetSize failed with %s", err) return FileInfo{}, err } modTime, err := metadata.GetModTime() if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("GetModTime failed with %s", err) return FileInfo{}, err } // Return file info. return FileInfo{ Volume: volume, Name: path, Size: size, ModTime: modTime, Mode: os.FileMode(0644), }, nil } // DeleteFile - delete a file func (xl XL) DeleteFile(volume, path string) error { if !isValidVolname(volume) { return errInvalidArgument } if !isValidPath(path) { return errInvalidArgument } // Loop through and delete each chunks. for index, disk := range xl.storageDisks { erasureFilePart := slashpath.Join(path, fmt.Sprintf("part.%d", index)) err := disk.DeleteFile(volume, erasureFilePart) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("DeleteFile failed with %s", err) return err } metadataFilePath := slashpath.Join(path, metadataFile) err = disk.DeleteFile(volume, metadataFilePath) if err != nil { log.WithFields(logrus.Fields{ "volume": volume, "path": path, }).Errorf("DeleteFile failed with %s", err) return err } } return nil }