Bump version to 2.7.1

Update README.md

Gopkg.lock

@@ -113,6 +113,18 @@
   packages = ["."]
   revision = "c2b33e84"
 
+[[projects]]
+  name = "github.com/klauspost/cpuid"
+  packages = ["."]
+  revision = "750c0591dbbd50ef88371c665ad49e426a4b830b"
+  version = "v1.3.1"
+
+[[projects]]
+  name = "github.com/klauspost/reedsolomon"
+  packages = ["."]
+  revision = "7daa20bf74337a939c54f892a2eca9d9b578eb7f"
+  version = "v1.9.9"
+
 [[projects]]
   name = "github.com/kr/fs"
   packages = ["."]
@@ -131,6 +143,18 @@
   packages = ["."]
   revision = "3f5f724cb5b182a5c278d6d3d55b40e7f8c2efb4"
 
+[[projects]]
+  name = "github.com/minio/highwayhash"
+  packages = ["."]
+  revision = "86a2a969d04373bf05ca722517d30fb1c9a3e4f9"
+  version = "v1.0.1"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/mmcloughlin/avo"
+  packages = ["attr","build","buildtags","gotypes","internal/prnt","internal/stack","ir","operand","pass","printer","reg","src","x86"]
+  revision = "443f81d771042b019379ae4bfcd0a591cb47c88a"
+
 [[projects]]
   name = "github.com/ncw/swift"
   packages = ["."]
@@ -173,6 +197,12 @@
   packages = ["blowfish","chacha20","curve25519","ed25519","ed25519/internal/edwards25519","internal/subtle","pbkdf2","poly1305","ssh","ssh/agent","ssh/internal/bcrypt_pbkdf","ssh/terminal"]
   revision = "056763e48d71961566155f089ac0f02f1dda9b5a"
 
+[[projects]]
+  name = "golang.org/x/mod"
+  packages = ["semver"]
+  revision = "859b3ef565e237f9f1a0fb6b55385c497545680d"
+  version = "v0.3.0"
+
 [[projects]]
   branch = "master"
   name = "golang.org/x/net"
@@ -196,6 +226,18 @@
   revision = "342b2e1fbaa52c93f31447ad2c6abc048c63e475"
   version = "v0.3.2"
 
+[[projects]]
+  branch = "master"
+  name = "golang.org/x/tools"
+  packages = ["go/ast/astutil","go/gcexportdata","go/internal/gcimporter","go/internal/packagesdriver","go/packages","go/types/typeutil","internal/event","internal/event/core","internal/event/keys","internal/event/label","internal/gocommand","internal/packagesinternal","internal/typesinternal"]
+  revision = "5d1fdd8fa3469142b9369713b23d8413d6d83189"
+
+[[projects]]
+  branch = "master"
+  name = "golang.org/x/xerrors"
+  packages = [".","internal"]
+  revision = "5ec99f83aff198f5fbd629d6c8d8eb38a04218ca"
+
 [[projects]]
   name = "google.golang.org/api"
   packages = ["drive/v3","googleapi","googleapi/transport","internal","internal/gensupport","internal/third_party/uritemplates","iterator","option","option/internaloption","storage/v1","transport/cert","transport/http","transport/http/internal/propagation"]
@@ -223,6 +265,6 @@
 [solve-meta]
   analyzer-name = "dep"
   analyzer-version = 1
-  inputs-digest = "e462352e0b0c726247078462e30a79330ac7a8b9dc62e9ed9d1e097b684224e9"
+  inputs-digest = "0e6ea2be64dedc36cb9192f1d410917ea72896302011e55b6df5e4c00c1c2f1c"
   solver-name = "gps-cdcl"
   solver-version = 1

README.md

@@ -2,7 +2,7 @@
 
 Duplicacy is a new generation cross-platform cloud backup tool based on the idea of [Lock-Free Deduplication](https://github.com/gilbertchen/duplicacy/wiki/Lock-Free-Deduplication).  
 
-This repository hosts source code, design documents, and binary releases of the command line version of Duplicacy.  There is also a Duplicacy GUI frontend built for Windows and Mac OS X available from https://duplicacy.com.
+This repository hosts source code, design documents, and binary releases of the command line version of Duplicacy.  There is also a Web GUI frontend built for Windows, macOS, and Linux, available from https://duplicacy.com.
 
 There is a special edition of Duplicacy developed for VMware vSphere (ESXi) named [Vertical Backup](https://www.verticalbackup.com) that can back up virtual machine files on ESXi to local drives, network or cloud storages.
 
@@ -10,14 +10,15 @@ There is a special edition of Duplicacy developed for VMware vSphere (ESXi) name
 
 There are 3 core advantages of Duplicacy over any other open-source or commercial backup tools:
 
-* Duplicacy is the *only* cloud backup tool that allows multiple computers to back up to the same cloud storage, taking advantage of cross-computer deduplication whenever possible, without direct communication among them.  This feature literally turns any cloud storage server supporting only a basic set of file operations into a sophisticated deduplication-aware server.  
+* Duplicacy is the *only* cloud backup tool that allows multiple computers to back up to the same cloud storage, taking advantage of cross-computer deduplication whenever possible, without direct communication among them.  This feature turns any cloud storage server supporting only a basic set of file operations into a sophisticated deduplication-aware server.  
 
-* Unlike other chunk-based backup tools where chunks are grouped into pack files and a chunk database is used to track which chunks are stored inside each pack file, Duplicacy takes a database-less approach where every chunk is saved independently using its hash as the file name to facilitate quick lookups.  The lack of a centralized chunk database not only makes the implementation less error-prone, but also produces a highly maintainable piece of software with plenty of room for development of new features and usability enhancements.
+* Unlike other chunk-based backup tools where chunks are grouped into pack files and a chunk database is used to track which chunks are stored inside each pack file, Duplicacy takes a database-less approach where every chunk is saved independently using its hash as the file name to facilitate quick lookups.  The avoidance of a centralized chunk database not only produces a simpler and less error-prone implementation, but also makes it easier to develop advanced features, such as [Asymmetric Encryption](https://github.com/gilbertchen/duplicacy/wiki/RSA-encryption) for stronger encryption and [Erasure Coding](https://github.com/gilbertchen/duplicacy/wiki/Erasure-coding) for resilient data protection.
 
 * Duplicacy is fast.  While the performance wasn't the top-priority design goal, Duplicacy has been shown to outperform other backup tools by a considerable margin, as indicated by the following results obtained from a [benchmarking experiment](https://github.com/gilbertchen/benchmarking) backing up the [Linux code base](https://github.com/torvalds/linux) using Duplicacy and 3 other open-source backup tools.
 
 [![Comparison of Duplicacy, restic, Attic, duplicity](https://github.com/gilbertchen/duplicacy/blob/master/images/duplicacy_benchmark_speed.png "Comparison of Duplicacy, restic, Attic, duplicity")](https://github.com/gilbertchen/benchmarking)
 
+
 ## Getting Started
 
 * [A brief introduction](https://github.com/gilbertchen/duplicacy/wiki/Quick-Start)
@@ -44,6 +45,7 @@ Duplicacy currently provides the following storage backends:
 * WebDAV (under beta testing)
 * pcloud (via WebDAV)
 * Box.com (via WebDAV)
+* File Fabric by [Storage Made Easy](https://storagemadeeasy.com/)
 
 Please consult the [wiki page](https://github.com/gilbertchen/duplicacy/wiki/Storage-Backends) on how to set up Duplicacy to work with each cloud storage.
 
@@ -64,9 +66,9 @@ to find the differences from previous backups and only then uploading the differ
 
 [Duplicati](https://duplicati.com) is one of the first backup tools that adopt the chunk-based approach to split files into chunks which are then uploaded to the storage.  The chunk-based approach got the incremental backup model right in the sense that every incremental backup is actually a full snapshot.  As Duplicati splits files into fixed-size chunks,  deletions or insertions of a few bytes will foil the deduplication.  Cloud support is extensive, but multiple clients can't back up to the same storage location.
 
-[Attic](https://attic-backup.org) has been acclaimed by some as the [Holy Grail of backups](https://www.stavros.io/posts/holy-grail-backups).  It follows the same incremental backup model like Duplicati, but embraces the variable-size chunk algorithm for better performance and higher deduplication efficiency (not susceptible to byte insertion and deletion any more).  Deletions of old backup is also supported.  However, no cloud backends are implemented.  Although concurrent backups from multiple clients to the same storage is in theory possible by the use of locking, it is
+[Attic](https://attic-backup.org) has been acclaimed by some as the [Holy Grail of backups](https://www.stavros.io/posts/holy-grail-backups).  It follows the same incremental backup model like Duplicati but embraces the variable-size chunk algorithm for better performance and higher deduplication efficiency (not susceptible to byte insertion and deletion any more).  Deletions of old backup are also supported.  However, no cloud backends are implemented.  Although concurrent backups from multiple clients to the same storage is in theory possible by the use of locking, it is
 [not recommended](http://librelist.com/browser//attic/2014/11/11/backing-up-multiple-servers-into-a-single-repository/#e96345aa5a3469a87786675d65da492b) by the developer due to chunk indices being kept in a local cache.
-Concurrent access is not only a convenience; it is a necessity for better deduplication.  For instance, if multiple machines with the same OS installed can back up their entire drives to the same storage, only one copy of the system files needs to be stored, greatly reducing the storage space regardless of the number of machines.  Attic still adopts the traditional approach of using a centralized indexing database to manage chunks, and relies heavily on caching to improve performance.  The presence of exclusive locking makes it hard to be extended to cloud storages.
+Concurrent access is not only a convenience; it is a necessity for better deduplication.  For instance, if multiple machines with the same OS installed can back up their entire drives to the same storage, only one copy of the system files needs to be stored, greatly reducing the storage space regardless of the number of machines.  Attic still adopts the traditional approach of using a centralized indexing database to manage chunks and relies heavily on caching to improve performance.  The presence of exclusive locking makes it hard to be extended to cloud storages.
 
 [restic](https://restic.github.io) is a more recent addition. It uses a format similar to the git packfile format.  Multiple clients backing up to the same storage are still guarded by
 [locks](https://github.com/restic/restic/blob/master/doc/Design.md#locks), and because a chunk database is used, deduplication isn't real-time (different clients sharing the same files will upload different copies of the same chunks).  A prune operation will completely block all other clients connected to the storage from doing their regular backups.  Moreover, since most cloud storage services do not provide a locking service, the best effort is to use some basic file operations to simulate a lock, but distributed locking is known to be a hard problem and it is unclear how reliable restic's lock implementation is.  A faulty implementation may cause a prune operation to accidentally delete data still in use, resulting in unrecoverable data loss.  This is the exact problem that we avoided by taking the lock-free approach.

duplicacy/duplicacy_main.go

@@ -458,8 +458,26 @@ func configRepository(context *cli.Context, init bool) {
 		if iterations == 0 {
 			iterations = duplicacy.CONFIG_DEFAULT_ITERATIONS
 		}
+
+		dataShards := 0
+		parityShards := 0
+		shards := context.String("erasure-coding")
+		if shards != "" {
+			shardsRegex := regexp.MustCompile(`^([0-9]+):([0-9]+)$`)
+			matched := shardsRegex.FindStringSubmatch(shards)
+			if matched == nil {
+				duplicacy.LOG_ERROR("STORAGE_ERASURECODE", "Invalid erasure coding parameters: %s", shards)
+			} else {
+				dataShards, _ = strconv.Atoi(matched[1])
+				parityShards, _ = strconv.Atoi(matched[2])
+				if dataShards == 0 || dataShards > 256 || parityShards == 0 || parityShards > dataShards {
+					duplicacy.LOG_ERROR("STORAGE_ERASURECODE", "Invalid erasure coding parameters: %s", shards)
+				}
+			}
+		}
+
 		duplicacy.ConfigStorage(storage, iterations, compressionLevel, averageChunkSize, maximumChunkSize,
-			minimumChunkSize, storagePassword, otherConfig, bitCopy, context.String("key"))
+			minimumChunkSize, storagePassword, otherConfig, bitCopy, context.String("key"), dataShards, parityShards)
 	}
 
 	duplicacy.Preferences = append(duplicacy.Preferences, preference)
@@ -565,6 +583,11 @@ func setPreference(context *cli.Context) {
 		newPreference.FiltersFile = context.String("filters")
 	}
 
+	triBool = context.Generic("exclude-by-attribute").(*TriBool)
+	if triBool.IsSet() {
+		newPreference.ExcludeByAttribute = triBool.IsTrue()
+	}
+
 	key := context.String("key")
 	value := context.String("value")
 
@@ -746,7 +769,7 @@ func backupRepository(context *cli.Context) {
 	uploadRateLimit := context.Int("limit-rate")
 	enumOnly := context.Bool("enum-only")
 	storage.SetRateLimits(0, uploadRateLimit)
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, preference.NobackupFile, preference.FiltersFile)
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, preference.NobackupFile, preference.FiltersFile, preference.ExcludeByAttribute)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	backupManager.SetupSnapshotCache(preference.Name)
@@ -799,6 +822,7 @@ func restoreRepository(context *cli.Context) {
 	setOwner := !context.Bool("ignore-owner")
 
 	showStatistics := context.Bool("stats")
+	persist := context.Bool("persist")
 
 	var patterns []string
 	for _, pattern := range context.Args() {
@@ -823,13 +847,17 @@ func restoreRepository(context *cli.Context) {
 	duplicacy.LOG_INFO("SNAPSHOT_FILTER", "Loaded %d include/exclude pattern(s)", len(patterns))
 
 	storage.SetRateLimits(context.Int("limit-rate"), 0)
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, preference.NobackupFile, preference.FiltersFile)
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, preference.NobackupFile, preference.FiltersFile, preference.ExcludeByAttribute)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	loadRSAPrivateKey(context.String("key"), context.String("key-passphrase"), preference, backupManager, false)
 
 	backupManager.SetupSnapshotCache(preference.Name)
-	backupManager.Restore(repository, revision, true, quickMode, threads, overwrite, deleteMode, setOwner, showStatistics, patterns)
+	failed := backupManager.Restore(repository, revision, true, quickMode, threads, overwrite, deleteMode, setOwner, showStatistics, patterns, persist)
+	if failed > 0 {
+		duplicacy.LOG_ERROR("RESTORE_FAIL", "%d file(s) were not restored correctly", failed)
+		return
+	}
 
 	runScript(context, preference.Name, "post")
 }
@@ -865,7 +893,7 @@ func listSnapshots(context *cli.Context) {
 	tag := context.String("t")
 	revisions := getRevisions(context)
 
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", preference.ExcludeByAttribute)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	id := preference.SnapshotID
@@ -921,7 +949,7 @@ func checkSnapshots(context *cli.Context) {
 	tag := context.String("t")
 	revisions := getRevisions(context)
 
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", false)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	loadRSAPrivateKey(context.String("key"), context.String("key-passphrase"), preference, backupManager, false)
@@ -939,9 +967,10 @@ func checkSnapshots(context *cli.Context) {
 	checkChunks := context.Bool("chunks")
 	searchFossils := context.Bool("fossils")
 	resurrect := context.Bool("resurrect")
+	persist := context.Bool("persist")
 
 	backupManager.SetupSnapshotCache(preference.Name)
-	backupManager.SnapshotManager.CheckSnapshots(id, revisions, tag, showStatistics, showTabular, checkFiles, checkChunks, searchFossils, resurrect, threads)
+	backupManager.SnapshotManager.CheckSnapshots(id, revisions, tag, showStatistics, showTabular, checkFiles, checkChunks, searchFossils, resurrect, threads, persist)
 
 	runScript(context, preference.Name, "post")
 }
@@ -979,7 +1008,8 @@ func printFile(context *cli.Context) {
 		snapshotID = context.String("id")
 	}
 
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", false)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	loadRSAPrivateKey(context.String("key"), context.String("key-passphrase"), preference, backupManager, false)
@@ -1037,13 +1067,13 @@ func diff(context *cli.Context) {
 	}
 
 	compareByHash := context.Bool("hash")
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", false)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	loadRSAPrivateKey(context.String("key"), context.String("key-passphrase"), preference, backupManager, false)
 
 	backupManager.SetupSnapshotCache(preference.Name)
-	backupManager.SnapshotManager.Diff(repository, snapshotID, revisions, path, compareByHash, preference.NobackupFile, preference.FiltersFile)
+	backupManager.SnapshotManager.Diff(repository, snapshotID, revisions, path, compareByHash, preference.NobackupFile, preference.FiltersFile, preference.ExcludeByAttribute)
 
 	runScript(context, preference.Name, "post")
 }
@@ -1082,7 +1112,7 @@ func showHistory(context *cli.Context) {
 
 	revisions := getRevisions(context)
 	showLocalHash := context.Bool("hash")
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", false)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	backupManager.SetupSnapshotCache(preference.Name)
@@ -1145,7 +1175,7 @@ func pruneSnapshots(context *cli.Context) {
 		os.Exit(ArgumentExitCode)
 	}
 
-	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "")
+	backupManager := duplicacy.CreateBackupManager(preference.SnapshotID, storage, repository, password, "", "", false)
 	duplicacy.SavePassword(*preference, "password", password)
 
 	backupManager.SetupSnapshotCache(preference.Name)
@@ -1165,9 +1195,14 @@ func copySnapshots(context *cli.Context) {
 		os.Exit(ArgumentExitCode)
 	}
 
-	threads := context.Int("threads")
-	if threads < 1 {
-		threads = 1
+	uploadingThreads := context.Int("threads")
+	if uploadingThreads < 1 {
+		uploadingThreads = 1
+	}
+
+	downloadingThreads := context.Int("download-threads")
+	if downloadingThreads < 1 {
+		downloadingThreads = 1
 	}
 
 	repository, source := getRepositoryPreference(context, context.String("from"))
@@ -1175,7 +1210,7 @@ func copySnapshots(context *cli.Context) {
 	runScript(context, source.Name, "pre")
 
 	duplicacy.LOG_INFO("STORAGE_SET", "Source storage set to %s", source.StorageURL)
-	sourceStorage := duplicacy.CreateStorage(*source, false, threads)
+	sourceStorage := duplicacy.CreateStorage(*source, false, downloadingThreads)
 	if sourceStorage == nil {
 		return
 	}
@@ -1185,7 +1220,7 @@ func copySnapshots(context *cli.Context) {
 		sourcePassword = duplicacy.GetPassword(*source, "password", "Enter source storage password:", false, false)
 	}
 
-	sourceManager := duplicacy.CreateBackupManager(source.SnapshotID, sourceStorage, repository, sourcePassword, "", "")
+	sourceManager := duplicacy.CreateBackupManager(source.SnapshotID, sourceStorage, repository, sourcePassword, "", "", false)
 	sourceManager.SetupSnapshotCache(source.Name)
 	duplicacy.SavePassword(*source, "password", sourcePassword)
 
@@ -1205,7 +1240,7 @@ func copySnapshots(context *cli.Context) {
 	}
 
 	duplicacy.LOG_INFO("STORAGE_SET", "Destination storage set to %s", destination.StorageURL)
-	destinationStorage := duplicacy.CreateStorage(*destination, false, threads)
+	destinationStorage := duplicacy.CreateStorage(*destination, false, uploadingThreads)
 	if destinationStorage == nil {
 		return
 	}
@@ -1220,7 +1255,7 @@ func copySnapshots(context *cli.Context) {
 	destinationStorage.SetRateLimits(0, context.Int("upload-limit-rate"))
 
 	destinationManager := duplicacy.CreateBackupManager(destination.SnapshotID, destinationStorage, repository,
-		destinationPassword, "", "")
+		                                                  destinationPassword, "", "", false)
 	duplicacy.SavePassword(*destination, "password", destinationPassword)
 	destinationManager.SetupSnapshotCache(destination.Name)
 
@@ -1230,7 +1265,7 @@ func copySnapshots(context *cli.Context) {
 		snapshotID = context.String("id")
 	}
 
-	sourceManager.CopySnapshots(destinationManager, snapshotID, revisions, threads)
+	sourceManager.CopySnapshots(destinationManager, snapshotID, revisions, uploadingThreads, downloadingThreads)
 	runScript(context, source.Name, "post")
 }
 
@@ -1403,6 +1438,11 @@ func main() {
 					Usage:    "the RSA public key to encrypt file chunks",
 					Argument: "<public key>",
 				},
+				cli.StringFlag{
+					Name:     "erasure-coding",
+					Usage:    "enable erasure coding to protect against storage corruption",
+					Argument: "<data shards>:<parity shards>",
+				},
 			},
 			Usage:     "Initialize the storage if necessary and the current directory as the repository",
 			ArgsUsage: "<snapshot id> <storage url>",
@@ -1515,6 +1555,10 @@ func main() {
 					Usage:    "the RSA private key to decrypt file chunks",
 					Argument: "<private key>",
 				},
+				cli.BoolFlag{
+					Name:  "persist",
+					Usage: "continue processing despite chunk errors or existing files (without -overwrite), reporting any affected files",
+        },
 				cli.StringFlag{
 					Name:     "key-passphrase",
 					Usage:    "the passphrase to decrypt the RSA private key",
@@ -1642,6 +1686,10 @@ func main() {
 					Usage:    "number of threads used to verify chunks",
 					Argument: "<n>",
 				},
+				cli.BoolFlag{
+					Name:  "persist",
+					Usage: "continue processing despite chunk errors, reporting any affected (corrupted) files",
+				},
 			},
 			Usage:     "Check the integrity of snapshots",
 			ArgsUsage: " ",
@@ -1882,6 +1930,11 @@ func main() {
 					Usage:    "the RSA public key to encrypt file chunks",
 					Argument: "<public key>",
 				},
+				cli.StringFlag{
+					Name:     "erasure-coding",
+					Usage:    "enable erasure coding to protect against storage corruption",
+					Argument: "<data shards>:<parity shards>",
+				},
 			},
 			Usage:     "Add an additional storage to be used for the existing repository",
 			ArgsUsage: "<storage name> <snapshot id> <storage url>",
@@ -1921,6 +1974,12 @@ func main() {
 					Argument: "<file name>",
 					Value:    "",
 				},
+				cli.GenericFlag{
+					Name:  "exclude-by-attribute",
+					Usage: "Exclude files based on file attributes. (macOS only, com_apple_backup_excludeItem)",
+					Value: &TriBool{},
+					Arg:   "true",
+				},
 				cli.StringFlag{
 					Name:  "key",
 					Usage: "add a key/password whose value is supplied by the -value option",
@@ -1985,6 +2044,12 @@ func main() {
 					Usage:    "number of uploading threads",
 					Argument: "<n>",
 				},
+				cli.IntFlag{
+					Name:     "download-threads",
+					Value:    1,
+					Usage:    "number of downloading threads",
+					Argument: "<n>",
+				},
 				cli.StringFlag{
 					Name:     "key",
 					Usage:    "the RSA private key to decrypt file chunks from the source storage",
@@ -2114,7 +2179,7 @@ func main() {
 	app.Name = "duplicacy"
 	app.HelpName = "duplicacy"
 	app.Usage = "A new generation cloud backup tool based on lock-free deduplication"
-	app.Version = "2.6.0" + " (" + GitCommit + ")"
+	app.Version = "2.7.1" + " (" + GitCommit + ")"
 
 	// If the program is interrupted, call the RunAtError function.
 	c := make(chan os.Signal, 1)

src/duplicacy_backupmanager.go

@@ -35,7 +35,11 @@ type BackupManager struct {
 	config *Config // contains a number of options
 
 	nobackupFile string // don't backup directory when this file name is found
-	filtersFile string  // the path to the filters file
+
+  filtersFile string  // the path to the filters file
+
+  excludeByAttribute bool // don't backup file based on file attribute
+
 }
 
 func (manager *BackupManager) SetDryRun(dryRun bool) {
@@ -45,7 +49,7 @@ func (manager *BackupManager) SetDryRun(dryRun bool) {
 // CreateBackupManager creates a backup manager using the specified 'storage'.  'snapshotID' is a unique id to
 // identify snapshots created for this repository.  'top' is the top directory of the repository.  'password' is the
 // master key which can be nil if encryption is not enabled.
-func CreateBackupManager(snapshotID string, storage Storage, top string, password string, nobackupFile string, filtersFile string) *BackupManager {
+func CreateBackupManager(snapshotID string, storage Storage, top string, password string, nobackupFile string, filtersFile string, excludeByAttribute bool) *BackupManager {
 
 	config, _, err := DownloadConfig(storage, password)
 	if err != nil {
@@ -68,7 +72,10 @@ func CreateBackupManager(snapshotID string, storage Storage, top string, passwor
 		config: config,
 
 		nobackupFile: nobackupFile,
+
 		filtersFile: filtersFile,
+
+		excludeByAttribute: excludeByAttribute,
 	}
 
 	if IsDebugging() {
@@ -184,8 +191,17 @@ func (manager *BackupManager) Backup(top string, quickMode bool, threads int, ta
 
 	LOG_DEBUG("BACKUP_PARAMETERS", "top: %s, quick: %t, tag: %s", top, quickMode, tag)
 
+	if manager.config.DataShards != 0 && manager.config.ParityShards != 0 {
+		LOG_INFO("BACKUP_ERASURECODING", "Erasure coding is enabled with %d data shards and %d parity shards",
+		         manager.config.DataShards, manager.config.ParityShards)
+	}
+
 	if manager.config.rsaPublicKey != nil && len(manager.config.FileKey) > 0 {
-		LOG_INFO("BACKUP_KEY", "RSA encryption is enabled" )
+		LOG_INFO("BACKUP_KEY", "RSA encryption is enabled")
+	}
+
+	if manager.excludeByAttribute {
+		LOG_INFO("BACKUP_EXCLUDE", "Exclude files with no-backup attributes")
 	}
 
 	remoteSnapshot := manager.SnapshotManager.downloadLatestSnapshot(manager.snapshotID)
@@ -201,7 +217,7 @@ func (manager *BackupManager) Backup(top string, quickMode bool, threads int, ta
 
 	LOG_INFO("BACKUP_INDEXING", "Indexing %s", top)
 	localSnapshot, skippedDirectories, skippedFiles, err := CreateSnapshotFromDirectory(manager.snapshotID, shadowTop,
-		                                                                                manager.nobackupFile, manager.filtersFile)
+		                                                                                manager.nobackupFile, manager.filtersFile, manager.excludeByAttribute)
 	if err != nil {
 		LOG_ERROR("SNAPSHOT_LIST", "Failed to list the directory %s: %v", top, err)
 		return false
@@ -746,7 +762,7 @@ func (manager *BackupManager) Backup(top string, quickMode bool, threads int, ta
 // the same as 'top'.  'quickMode' will bypass files with unchanged sizes and timestamps.  'deleteMode' will
 // remove local files that don't exist in the snapshot. 'patterns' is used to include/exclude certain files.
 func (manager *BackupManager) Restore(top string, revision int, inPlace bool, quickMode bool, threads int, overwrite bool,
-	deleteMode bool, setOwner bool, showStatistics bool, patterns []string) bool {
+	deleteMode bool, setOwner bool, showStatistics bool, patterns []string, allowFailures bool) int {
 
 	startTime := time.Now().Unix()
 
@@ -769,7 +785,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 		err = os.Mkdir(top, 0744)
 		if err != nil {
 			LOG_ERROR("RESTORE_MKDIR", "Can't create the directory to be restored: %v", err)
-			return false
+			return 0
 		}
 	}
 
@@ -777,17 +793,17 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 	err = os.Mkdir(path.Join(top, DUPLICACY_DIRECTORY), 0744)
 	if err != nil && !os.IsExist(err) {
 		LOG_ERROR("RESTORE_MKDIR", "Failed to create the preference directory: %v", err)
-		return false
+		return 0
 	}
 
 	remoteSnapshot := manager.SnapshotManager.DownloadSnapshot(manager.snapshotID, revision)
 	manager.SnapshotManager.DownloadSnapshotContents(remoteSnapshot, patterns, true)
 
 	localSnapshot, _, _, err := CreateSnapshotFromDirectory(manager.snapshotID, top, manager.nobackupFile,
-		                                                    manager.filtersFile)
+		                                                    manager.filtersFile, manager.excludeByAttribute)
 	if err != nil {
 		LOG_ERROR("SNAPSHOT_LIST", "Failed to list the repository: %v", err)
-		return false
+		return 0
 	}
 
 	LOG_INFO("RESTORE_START", "Restoring %s to revision %d", top, revision)
@@ -814,6 +830,11 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 
 	var totalFileSize int64
 	var downloadedFileSize int64
+	var failedFiles int
+	var skippedFileSize int64
+	var skippedFiles int64
+	
+	var downloadedFiles []*Entry
 
 	i := 0
 	for _, entry := range remoteSnapshot.Files {
@@ -832,6 +853,8 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 					i++
 					if quickMode && local.IsSameAs(entry) {
 						LOG_TRACE("RESTORE_SKIP", "File %s unchanged (by size and timestamp)", local.Path)
+						skippedFileSize += entry.Size
+						skippedFiles++
 						skipped = true
 					}
 				}
@@ -861,7 +884,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 			err = os.Symlink(entry.Link, fullPath)
 			if err != nil {
 				LOG_ERROR("RESTORE_SYMLINK", "Can't create symlink %s: %v", entry.Path, err)
-				return false
+				return 0
 			}
 			entry.RestoreMetadata(fullPath, nil, setOwner)
 			LOG_TRACE("DOWNLOAD_DONE", "Symlink %s updated", entry.Path)
@@ -870,7 +893,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 
 			if err == nil && !stat.IsDir() {
 				LOG_ERROR("RESTORE_NOTDIR", "The path %s is not a directory", fullPath)
-				return false
+				return 0
 			}
 
 			if os.IsNotExist(err) {
@@ -879,7 +902,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 				err = os.MkdirAll(fullPath, 0700)
 				if err != nil && !os.IsExist(err) {
 					LOG_ERROR("RESTORE_MKDIR", "%v", err)
-					return false
+					return 0
 				}
 			}
 		} else {
@@ -897,14 +920,13 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 	// Sort entries by their starting chunks in order to linearize the access to the chunk chain.
 	sort.Sort(ByChunk(fileEntries))
 
-	chunkDownloader := CreateChunkDownloader(manager.config, manager.storage, nil, showStatistics, threads)
+	chunkDownloader := CreateChunkDownloader(manager.config, manager.storage, nil, showStatistics, threads, allowFailures)
 	chunkDownloader.AddFiles(remoteSnapshot, fileEntries)
 
 	chunkMaker := CreateChunkMaker(manager.config, true)
 
 	startDownloadingTime := time.Now().Unix()
 
-	var downloadedFiles []*Entry
 	// Now download files one by one
 	for _, file := range fileEntries {
 
@@ -914,12 +936,16 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 			if quickMode {
 				if file.IsSameAsFileInfo(stat) {
 					LOG_TRACE("RESTORE_SKIP", "File %s unchanged (by size and timestamp)", file.Path)
+					skippedFileSize += file.Size
+					skippedFiles++
 					continue
 				}
 			}
 
 			if file.Size == 0 && file.IsSameAsFileInfo(stat) {
 				LOG_TRACE("RESTORE_SKIP", "File %s unchanged (size 0)", file.Path)
+				skippedFileSize += file.Size
+				skippedFiles++
 				continue
 			}
 		} else {
@@ -935,22 +961,39 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 			newFile, err := os.OpenFile(fullPath, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, file.GetPermissions())
 			if err != nil {
 				LOG_ERROR("DOWNLOAD_OPEN", "Failed to create empty file: %v", err)
-				return false
+				return 0
 			}
 			newFile.Close()
 
 			file.RestoreMetadata(fullPath, nil, setOwner)
 			if !showStatistics {
 				LOG_INFO("DOWNLOAD_DONE", "Downloaded %s (0)", file.Path)
+				downloadedFileSize += file.Size
+				downloadedFiles = append(downloadedFiles, file)
 			}
 
 			continue
 		}
 
-		if manager.RestoreFile(chunkDownloader, chunkMaker, file, top, inPlace, overwrite, showStatistics,
-			totalFileSize, downloadedFileSize, startDownloadingTime) {
+		downloaded, err := manager.RestoreFile(chunkDownloader, chunkMaker, file, top, inPlace, overwrite, showStatistics,
+			totalFileSize, downloadedFileSize, startDownloadingTime, allowFailures)
+		if err != nil {
+			// RestoreFile returned an error; if allowFailures is false RestoerFile would error out and not return so here
+			// we just need to show a warning
+			failedFiles++
+			LOG_WARN("DOWNLOAD_FAIL", "Failed to restore %s: %v", file.Path, err)
+			continue
+		}
+
+		// No error
+		if downloaded {
+			// No error, file was restored
 			downloadedFileSize += file.Size
 			downloadedFiles = append(downloadedFiles, file)
+		} else {
+			// No error, file was skipped
+			skippedFileSize += file.Size
+			skippedFiles++
 		}
 		file.RestoreMetadata(fullPath, nil, setOwner)
 	}
@@ -978,11 +1021,16 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 		}
 	}
 
+	if failedFiles > 0 {
+		return failedFiles
+	}
+
 	LOG_INFO("RESTORE_END", "Restored %s to revision %d", top, revision)
 	if showStatistics {
 		LOG_INFO("RESTORE_STATS", "Files: %d total, %s bytes", len(fileEntries), PrettySize(totalFileSize))
 		LOG_INFO("RESTORE_STATS", "Downloaded %d file, %s bytes, %d chunks",
 			len(downloadedFiles), PrettySize(downloadedFileSize), chunkDownloader.numberOfDownloadedChunks)
+		LOG_INFO("RESTORE_STATS", "Skipped %d file, %s bytes", skippedFiles, PrettySize(skippedFileSize))
 	}
 
 	runningTime := time.Now().Unix() - startTime
@@ -994,7 +1042,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 
 	chunkDownloader.Stop()
 
-	return true
+	return 0
 }
 
 // fileEncoder encodes one file at a time to avoid loading the full json description of the entire file tree
@@ -1154,8 +1202,11 @@ func (manager *BackupManager) UploadSnapshot(chunkMaker *ChunkMaker, uploader *C
 // Restore downloads a file from the storage.  If 'inPlace' is false, the download file is saved first to a temporary
 // file under the .duplicacy directory and then replaces the existing one.  Otherwise, the existing file will be
 // overwritten directly.
+// Return: true, nil:    Restored file; 
+//         false, nil:   Skipped file; 
+//         false, error: Failure to restore file (only if allowFailures == true)
 func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chunkMaker *ChunkMaker, entry *Entry, top string, inPlace bool, overwrite bool,
-	showStatistics bool, totalFileSize int64, downloadedFileSize int64, startTime int64) bool {
+	showStatistics bool, totalFileSize int64, downloadedFileSize int64, startTime int64, allowFailures bool) (bool, error) {
 
 	LOG_TRACE("DOWNLOAD_START", "Downloading %s", entry.Path)
 
@@ -1200,7 +1251,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 				existingFile, err = os.OpenFile(fullPath, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, 0600)
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_CREATE", "Failed to create the file %s for in-place writing: %v", fullPath, err)
-					return false
+					return false, nil
 				}
 
 				n := int64(1)
@@ -1212,30 +1263,24 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 				_, err = existingFile.Seek(entry.Size-n, 0)
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_CREATE", "Failed to resize the initial file %s for in-place writing: %v", fullPath, err)
-					return false
+					return false, nil
 				}
 				_, err = existingFile.Write([]byte("\x00\x00")[:n])
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_CREATE", "Failed to initialize the sparse file %s for in-place writing: %v", fullPath, err)
-					return false
+					return false, nil
 				}
 				existingFile.Close()
 				existingFile, err = os.Open(fullPath)
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_OPEN", "Can't reopen the initial file just created: %v", err)
-					return false
+					return false, nil
 				}
 				isNewFile = true
 			}
 		} else {
 			LOG_TRACE("DOWNLOAD_OPEN", "Can't open the existing file: %v", err)
 		}
-	} else {
-		if !overwrite {
-			LOG_ERROR("DOWNLOAD_OVERWRITE",
-				"File %s already exists.  Please specify the -overwrite option to continue", entry.Path)
-			return false
-		}
 	}
 
 	// The key in this map is the number of zeroes.  The value is the corresponding hash.
@@ -1304,7 +1349,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 					}
 					if err != nil {
 						LOG_ERROR("DOWNLOAD_SPLIT", "Failed to read existing file: %v", err)
-						return false
+						return false, nil
 					}
 				}
 				if count > 0 {
@@ -1325,6 +1370,19 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 			}
 
 			fileHash = hex.EncodeToString(fileHasher.Sum(nil))
+
+			if fileHash == entry.Hash && fileHash != "" {
+				LOG_TRACE("DOWNLOAD_SKIP", "File %s unchanged (by hash)", entry.Path)
+				return false, nil
+			}
+
+			// fileHash != entry.Hash, warn/error depending on -overwrite option
+			if !overwrite {
+				LOG_WERROR(allowFailures, "DOWNLOAD_OVERWRITE",
+							"File %s already exists.  Please specify the -overwrite option to overwrite", entry.Path)
+				return false, fmt.Errorf("file exists")
+			}
+
 		} else {
 			// If it is not inplace, we want to reuse any chunks in the existing file regardless their offets, so
 			// we run the chunk maker to split the original file.
@@ -1344,9 +1402,11 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 					return nil, false
 				})
 		}
+
+		// This is an additional check comparing fileHash to entry.Hash above, so this should no longer occur
 		if fileHash == entry.Hash && fileHash != "" {
 			LOG_TRACE("DOWNLOAD_SKIP", "File %s unchanged (by hash)", entry.Path)
-			return false
+			return false, nil
 		}
 	}
 
@@ -1374,7 +1434,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 			existingFile, err = os.OpenFile(fullPath, os.O_RDWR, 0)
 			if err != nil {
 				LOG_ERROR("DOWNLOAD_OPEN", "Failed to open the file %s for in-place writing", fullPath)
-				return false
+				return false, nil
 			}
 		}
 
@@ -1406,7 +1466,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 			_, err = existingFile.Seek(offset, 0)
 			if err != nil {
 				LOG_ERROR("DOWNLOAD_SEEK", "Failed to set the offset to %d for file %s: %v", offset, fullPath, err)
-				return false
+				return false, nil
 			}
 
 			// Check if the chunk is available in the existing file
@@ -1416,17 +1476,20 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 				_, err := io.CopyN(hasher, existingFile, int64(existingLengths[j]))
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_READ", "Failed to read the existing chunk %s: %v", hash, err)
-					return false
+					return false, nil
 				}
 				if IsDebugging() {
 					LOG_DEBUG("DOWNLOAD_UNCHANGED", "Chunk %s is unchanged", manager.config.GetChunkIDFromHash(hash))
 				}
 			} else {
 				chunk := chunkDownloader.WaitForChunk(i)
+				if chunk.isBroken {
+					return false, fmt.Errorf("chunk %s is corrupted", manager.config.GetChunkIDFromHash(hash))
+				}
 				_, err = existingFile.Write(chunk.GetBytes()[start:end])
 				if err != nil {
 					LOG_ERROR("DOWNLOAD_WRITE", "Failed to write to the file: %v", err)
-					return false
+					return false, nil
 				}
 				hasher.Write(chunk.GetBytes()[start:end])
 			}
@@ -1437,15 +1500,15 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 		// Must truncate the file if the new size is smaller
 		if err = existingFile.Truncate(offset); err != nil {
 			LOG_ERROR("DOWNLOAD_TRUNCATE", "Failed to truncate the file at %d: %v", offset, err)
-			return false
+			return false, nil
 		}
 
 		// Verify the download by hash
 		hash := hex.EncodeToString(hasher.Sum(nil))
 		if hash != entry.Hash && hash != "" && entry.Hash != "" && !strings.HasPrefix(entry.Hash, "#") {
-			LOG_ERROR("DOWNLOAD_HASH", "File %s has a mismatched hash: %s instead of %s (in-place)",
+			LOG_WERROR(allowFailures, "DOWNLOAD_HASH", "File %s has a mismatched hash: %s instead of %s (in-place)",
 				fullPath, "", entry.Hash)
-			return false
+			return false, fmt.Errorf("file corrupt (hash mismatch)")
 		}
 
 	} else {
@@ -1454,7 +1517,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 		newFile, err = os.OpenFile(temporaryPath, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, 0600)
 		if err != nil {
 			LOG_ERROR("DOWNLOAD_OPEN", "Failed to open file for writing: %v", err)
-			return false
+			return false, nil
 		}
 
 		hasher := manager.config.NewFileHasher()
@@ -1492,6 +1555,9 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 
 			if !hasLocalCopy {
 				chunk := chunkDownloader.WaitForChunk(i)
+				if chunk.isBroken {
+					return false, fmt.Errorf("chunk %s is corrupted", manager.config.GetChunkIDFromHash(hash))
+				}
 				// If the chunk was downloaded from the storage, we may still need a portion of it.
 				start := 0
 				if i == entry.StartChunk {
@@ -1507,7 +1573,7 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 			_, err = newFile.Write(data)
 			if err != nil {
 				LOG_ERROR("DOWNLOAD_WRITE", "Failed to write file: %v", err)
-				return false
+				return false, nil
 			}
 
 			hasher.Write(data)
@@ -1516,9 +1582,9 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 
 		hash := hex.EncodeToString(hasher.Sum(nil))
 		if hash != entry.Hash && hash != "" && entry.Hash != "" && !strings.HasPrefix(entry.Hash, "#") {
-			LOG_ERROR("DOWNLOAD_HASH", "File %s has a mismatched hash: %s instead of %s",
+			LOG_WERROR(allowFailures, "DOWNLOAD_HASH", "File %s has a mismatched hash: %s instead of %s",
 				entry.Path, hash, entry.Hash)
-			return false
+			return false, fmt.Errorf("file corrupt (hash mismatch)")
 		}
 
 		if existingFile != nil {
@@ -1532,31 +1598,40 @@ func (manager *BackupManager) RestoreFile(chunkDownloader *ChunkDownloader, chun
 		err = os.Remove(fullPath)
 		if err != nil && !os.IsNotExist(err) {
 			LOG_ERROR("DOWNLOAD_REMOVE", "Failed to remove the old file: %v", err)
-			return false
+			return false, nil
 		}
 
 		err = os.Rename(temporaryPath, fullPath)
 		if err != nil {
 			LOG_ERROR("DOWNLOAD_RENAME", "Failed to rename the file %s to %s: %v", temporaryPath, fullPath, err)
-			return false
+			return false, nil
 		}
 	}
 
 	if !showStatistics {
 		LOG_INFO("DOWNLOAD_DONE", "Downloaded %s (%d)", entry.Path, entry.Size)
 	}
-	return true
+	return true, nil
 }
 
 // CopySnapshots copies the specified snapshots from one storage to the other.
 func (manager *BackupManager) CopySnapshots(otherManager *BackupManager, snapshotID string,
-	revisionsToBeCopied []int, threads int) bool {
+	revisionsToBeCopied []int, uploadingThreads int, downloadingThreads int) bool {
 
 	if !manager.config.IsCompatiableWith(otherManager.config) {
 		LOG_ERROR("CONFIG_INCOMPATIBLE", "Two storages are not compatible for the copy operation")
 		return false
 	}
 
+	if otherManager.config.DataShards != 0 && otherManager.config.ParityShards != 0 {
+		LOG_INFO("BACKUP_ERASURECODING", "Erasure coding is enabled for the destination storage with %d data shards and %d parity shards",
+		         otherManager.config.DataShards, otherManager.config.ParityShards)
+	}
+
+	if otherManager.config.rsaPublicKey != nil && len(otherManager.config.FileKey) > 0 {
+		LOG_INFO("BACKUP_KEY", "RSA encryption is enabled for the destination")
+	}
+
 	if snapshotID == "" && len(revisionsToBeCopied) > 0 {
 		LOG_ERROR("SNAPSHOT_ERROR", "You must specify the snapshot id when one or more revisions are specified.")
 		return false
@@ -1695,63 +1770,64 @@ func (manager *BackupManager) CopySnapshots(otherManager *BackupManager, snapsho
 
 	LOG_DEBUG("SNAPSHOT_COPY", "Found %d chunks on destination storage", len(otherChunks))
 
-	chunksToCopy := 0
-	chunksToSkip := 0
+	var chunksToCopy []string
 
 	for chunkHash := range chunks {
 		otherChunkID := otherManager.config.GetChunkIDFromHash(chunkHash)
-		if _, found := otherChunks[otherChunkID]; found {
-			chunksToSkip++
-		} else {
-			chunksToCopy++
+		if _, found := otherChunks[otherChunkID]; !found {
+			chunksToCopy = append(chunksToCopy, chunkHash)
 		}
 	}
 
-	LOG_DEBUG("SNAPSHOT_COPY", "Chunks to copy = %d, to skip = %d, total = %d", chunksToCopy, chunksToSkip, chunksToCopy+chunksToSkip)
-	LOG_DEBUG("SNAPSHOT_COPY", "Total chunks in source snapshot revisions = %d\n", len(chunks))
+	LOG_INFO("SNAPSHOT_COPY", "Chunks to copy: %d, to skip: %d, total: %d", len(chunksToCopy), len(chunks) - len(chunksToCopy), len(chunks))
 
-	chunkDownloader := CreateChunkDownloader(manager.config, manager.storage, nil, false, threads)
+	chunkDownloader := CreateChunkDownloader(manager.config, manager.storage, nil, false, downloadingThreads, false)
 
-	chunkUploader := CreateChunkUploader(otherManager.config, otherManager.storage, nil, threads,
+	var uploadedBytes int64
+	startTime := time.Now()
+
+	copiedChunks := 0
+	chunkUploader := CreateChunkUploader(otherManager.config, otherManager.storage, nil, uploadingThreads,
 		func(chunk *Chunk, chunkIndex int, skipped bool, chunkSize int, uploadSize int) {
-			if skipped {
-				LOG_INFO("SNAPSHOT_COPY", "Chunk %s (%d/%d) exists at the destination", chunk.GetID(), chunkIndex, len(chunks))
-			} else {
-				LOG_INFO("SNAPSHOT_COPY", "Chunk %s (%d/%d) copied to the destination", chunk.GetID(), chunkIndex, len(chunks))
+			action := "Skipped"
+			if !skipped {
+				copiedChunks++
+				action = "Copied"
 			}
+
+			atomic.AddInt64(&uploadedBytes, int64(chunkSize))
+
+			elapsedTime := time.Now().Sub(startTime).Seconds()
+			speed := int64(float64(atomic.LoadInt64(&uploadedBytes)) / elapsedTime)
+			remainingTime := int64(float64(len(chunksToCopy) - chunkIndex - 1) / float64(chunkIndex + 1) * elapsedTime)
+			percentage := float64(chunkIndex + 1) / float64(len(chunksToCopy)) * 100.0
+			LOG_INFO("COPY_PROGRESS", "%s chunk %s (%d/%d) %sB/s %s %.1f%%",
+					action, chunk.GetID(), chunkIndex + 1, len(chunksToCopy),
+					PrettySize(speed), PrettyTime(remainingTime), percentage)
 			otherManager.config.PutChunk(chunk)
 		})
 
 	chunkUploader.Start()
 
-	totalCopied := 0
-	totalSkipped := 0
-	chunkIndex := 0
-
-	for chunkHash, isSnapshot := range chunks {
-		chunkIndex++
+	for _, chunkHash := range chunksToCopy {
+		chunkDownloader.AddChunk(chunkHash)
+	}
+	for i, chunkHash := range chunksToCopy {
 		chunkID := manager.config.GetChunkIDFromHash(chunkHash)
 		newChunkID := otherManager.config.GetChunkIDFromHash(chunkHash)
-		if _, found := otherChunks[newChunkID]; !found {
-			LOG_DEBUG("SNAPSHOT_COPY", "Copying chunk %s to %s", chunkID, newChunkID)
-			i := chunkDownloader.AddChunk(chunkHash)
-			chunk := chunkDownloader.WaitForChunk(i)
-			newChunk := otherManager.config.GetChunk()
-			newChunk.Reset(true)
-			newChunk.Write(chunk.GetBytes())
-			newChunk.isSnapshot = isSnapshot
-			chunkUploader.StartChunk(newChunk, chunkIndex)
-			totalCopied++
-		} else {
-			LOG_INFO("SNAPSHOT_COPY", "Chunk %s (%d/%d) skipped at the destination", chunkID, chunkIndex, len(chunks))
-			totalSkipped++
-		}
+		LOG_DEBUG("SNAPSHOT_COPY", "Copying chunk %s to %s", chunkID, newChunkID)
+		chunk := chunkDownloader.WaitForChunk(i)
+		newChunk := otherManager.config.GetChunk()
+		newChunk.Reset(true)
+		newChunk.Write(chunk.GetBytes())
+		newChunk.isSnapshot = chunks[chunkHash]
+		chunkUploader.StartChunk(newChunk, i)
 	}
 
 	chunkDownloader.Stop()
 	chunkUploader.Stop()
 
-	LOG_INFO("SNAPSHOT_COPY", "Copy complete, %d total chunks, %d chunks copied, %d skipped", totalCopied+totalSkipped, totalCopied, totalSkipped)
+	LOG_INFO("SNAPSHOT_COPY", "Copied %d new chunks and skipped %d existing chunks", copiedChunks, len(chunks) - copiedChunks)
 
 	for _, snapshot := range snapshots {
 		if revisionMap[snapshot.ID][snapshot.Revision] == false {

src/duplicacy_backupmanager_test.go

@@ -169,6 +169,12 @@ func getFileHash(path string) (hash string) {
 	return hex.EncodeToString(hasher.Sum(nil))
 }
 
+func assertRestoreFailures(t *testing.T, failedFiles int, expectedFailedFiles int) {
+	if failedFiles != expectedFailedFiles {
+		t.Errorf("Failed to restore %d instead of %d file(s)", failedFiles, expectedFailedFiles)
+	}
+}
+
 func TestBackupManager(t *testing.T) {
 
 	rand.Seed(time.Now().UnixNano())
@@ -226,12 +232,20 @@ func TestBackupManager(t *testing.T) {
 	cleanStorage(storage)
 
 	time.Sleep(time.Duration(delay) * time.Second)
+
+	dataShards := 0
+	parityShards := 0
+	if testErasureCoding {
+		dataShards = 5
+		parityShards = 2
+	}
+
 	if testFixedChunkSize {
-		if !ConfigStorage(storage, 16384, 100, 64*1024, 64*1024, 64*1024, password, nil, false, "") {
+		if !ConfigStorage(storage, 16384, 100, 64*1024, 64*1024, 64*1024, password, nil, false, "", dataShards, parityShards) {
 			t.Errorf("Failed to initialize the storage")
 		}
 	} else {
-		if !ConfigStorage(storage, 16384, 100, 64*1024, 256*1024, 16*1024, password, nil, false, "") {
+		if !ConfigStorage(storage, 16384, 100, 64*1024, 256*1024, 16*1024, password, nil, false, "", dataShards, parityShards) {
 			t.Errorf("Failed to initialize the storage")
 		}
 	}
@@ -239,15 +253,16 @@ func TestBackupManager(t *testing.T) {
 	time.Sleep(time.Duration(delay) * time.Second)
 
 	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
-	backupManager := CreateBackupManager("host1", storage, testDir, password, "", "")
+	backupManager := CreateBackupManager("host1", storage, testDir, password, "", "", false)
 	backupManager.SetupSnapshotCache("default")
 
 	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
 	backupManager.Backup(testDir+"/repository1" /*quickMode=*/, true, threads, "first", false, false, 0, false)
 	time.Sleep(time.Duration(delay) * time.Second)
 	SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
-	backupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, false /*quickMode=*/, false, threads /*overwrite=*/, true,
-		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil)
+	failedFiles := backupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, false /*quickMode=*/, false, threads /*overwrite=*/, true,
+		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, false)
+	assertRestoreFailures(t, failedFiles, 0)
 
 	for _, f := range []string{"file1", "file2", "dir1/file3"} {
 		if _, err := os.Stat(testDir + "/repository2/" + f); os.IsNotExist(err) {
@@ -270,8 +285,9 @@ func TestBackupManager(t *testing.T) {
 	backupManager.Backup(testDir+"/repository1" /*quickMode=*/, true, threads, "second", false, false, 0, false)
 	time.Sleep(time.Duration(delay) * time.Second)
 	SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
-	backupManager.Restore(testDir+"/repository2", 2 /*inPlace=*/, true /*quickMode=*/, true, threads /*overwrite=*/, true,
-		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil)
+	failedFiles = backupManager.Restore(testDir+"/repository2", 2 /*inPlace=*/, true /*quickMode=*/, true, threads /*overwrite=*/, true,
+		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, false)
+	assertRestoreFailures(t, failedFiles, 0)
 
 	for _, f := range []string{"file1", "file2", "dir1/file3"} {
 		hash1 := getFileHash(testDir + "/repository1/" + f)
@@ -298,8 +314,9 @@ func TestBackupManager(t *testing.T) {
 	createRandomFile(testDir+"/repository2/dir5/file5", 100)
 
 	SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
-	backupManager.Restore(testDir+"/repository2", 3 /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
-		/*deleteMode=*/ true /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil)
+	failedFiles = backupManager.Restore(testDir+"/repository2", 3 /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
+		/*deleteMode=*/ true /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, false)
+	assertRestoreFailures(t, failedFiles, 0)
 
 	for _, f := range []string{"file1", "file2", "dir1/file3"} {
 		hash1 := getFileHash(testDir + "/repository1/" + f)
@@ -325,8 +342,9 @@ func TestBackupManager(t *testing.T) {
 	os.Remove(testDir + "/repository1/file2")
 	os.Remove(testDir + "/repository1/dir1/file3")
 	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
-	backupManager.Restore(testDir+"/repository1", 3 /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
-		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, []string{"+file2", "+dir1/file3", "-*"})
+	failedFiles = backupManager.Restore(testDir+"/repository1", 3 /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
+		/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, []string{"+file2", "+dir1/file3", "-*"} /*allowFailures=*/, false)
+	assertRestoreFailures(t, failedFiles, 0)
 
 	for _, f := range []string{"file1", "file2", "dir1/file3"} {
 		hash1 := getFileHash(testDir + "/repository1/" + f)
@@ -341,7 +359,7 @@ func TestBackupManager(t *testing.T) {
 		t.Errorf("Expected 3 snapshots but got %d", numberOfSnapshots)
 	}
 	backupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{1, 2, 3} /*tag*/, "",
-		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1)
+		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1 /*allowFailures*/, false)
 	backupManager.SnapshotManager.PruneSnapshots("host1", "host1" /*revisions*/, []int{1} /*tags*/, nil /*retentions*/, nil,
 		/*exhaustive*/ false /*exclusive=*/, false /*ignoredIDs*/, nil /*dryRun*/, false /*deleteOnly*/, false /*collectOnly*/, false, 1)
 	numberOfSnapshots = backupManager.SnapshotManager.ListSnapshots( /*snapshotID*/ "host1" /*revisionsToList*/, nil /*tag*/, "" /*showFiles*/, false /*showChunks*/, false)
@@ -349,7 +367,7 @@ func TestBackupManager(t *testing.T) {
 		t.Errorf("Expected 2 snapshots but got %d", numberOfSnapshots)
 	}
 	backupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{2, 3} /*tag*/, "",
-		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1)
+		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1 /*allowFailures*/, false)
 	backupManager.Backup(testDir+"/repository1" /*quickMode=*/, false, threads, "fourth", false, false, 0, false)
 	backupManager.SnapshotManager.PruneSnapshots("host1", "host1" /*revisions*/, nil /*tags*/, nil /*retentions*/, nil,
 		/*exhaustive*/ false /*exclusive=*/, true /*ignoredIDs*/, nil /*dryRun*/, false /*deleteOnly*/, false /*collectOnly*/, false, 1)
@@ -358,9 +376,348 @@ func TestBackupManager(t *testing.T) {
 		t.Errorf("Expected 3 snapshots but got %d", numberOfSnapshots)
 	}
 	backupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{2, 3, 4} /*tag*/, "",
-		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1)
+		/*showStatistics*/ false /*showTabular*/, false /*checkFiles*/, false /*checkChunks*/, false /*searchFossils*/, false /*resurrect*/, false, 1 /*allowFailures*/, false)
 
 	/*buf := make([]byte, 1<<16)
 	  runtime.Stack(buf, true)
 	  fmt.Printf("%s", buf)*/
 }
+
+// Create file with random file with certain seed
+func createRandomFileSeeded(path string, maxSize int, seed int64) {
+	rand.Seed(seed)
+	file, err := os.OpenFile(path, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, 0644)
+	if err != nil {
+		LOG_ERROR("RANDOM_FILE", "Can't open %s for writing: %v", path, err)
+		return
+	}
+
+	defer file.Close()
+
+	size := maxSize/2 + rand.Int()%(maxSize/2)
+
+	buffer := make([]byte, 32*1024)
+	for size > 0 {
+		bytes := size
+		if bytes > cap(buffer) {
+			bytes = cap(buffer)
+		}
+		rand.Read(buffer[:bytes])
+		bytes, err = file.Write(buffer[:bytes])
+		if err != nil {
+			LOG_ERROR("RANDOM_FILE", "Failed to write to %s: %v", path, err)
+			return
+		}
+		size -= bytes
+	}
+}
+
+func corruptFile(path string, start int, length int, seed int64) {
+	rand.Seed(seed)
+
+	file, err := os.OpenFile(path, os.O_WRONLY, 0644)
+	if err != nil {
+		LOG_ERROR("CORRUPT_FILE", "Can't open %s for writing: %v", path, err)
+		return
+	}
+
+	defer func() {
+		if file != nil {
+			file.Close()
+		}
+	}()
+
+	_, err = file.Seek(int64(start), 0)
+	if err != nil {
+		LOG_ERROR("CORRUPT_FILE", "Can't seek to the offset %d: %v", start, err)
+		return
+	}
+
+	buffer := make([]byte, length)
+	rand.Read(buffer)
+
+	_, err = file.Write(buffer)
+	if err != nil {
+		LOG_ERROR("CORRUPT_FILE", "Failed to write to %s: %v", path, err)
+		return
+	}
+}
+
+func TestPersistRestore(t *testing.T) {
+	// We want deterministic output here so we can test the expected files are corrupted by missing or corrupt chunks
+	// There use rand functions with fixed seed, and known keys
+
+	setTestingT(t)
+	SetLoggingLevel(INFO)
+
+	defer func() {
+		if r := recover(); r != nil {
+			switch e := r.(type) {
+			case Exception:
+				t.Errorf("%s %s", e.LogID, e.Message)
+				debug.PrintStack()
+			default:
+				t.Errorf("%v", e)
+				debug.PrintStack()
+			}
+		}
+	}()
+
+	testDir := path.Join(os.TempDir(), "duplicacy_test")
+	os.RemoveAll(testDir)
+	os.MkdirAll(testDir, 0700)
+	os.Mkdir(testDir+"/repository1", 0700)
+	os.Mkdir(testDir+"/repository1/dir1", 0700)
+	os.Mkdir(testDir+"/repository1/.duplicacy", 0700)
+	os.Mkdir(testDir+"/repository2", 0700)
+	os.Mkdir(testDir+"/repository2/.duplicacy", 0700)
+	os.Mkdir(testDir+"/repository3", 0700)
+	os.Mkdir(testDir+"/repository3/.duplicacy", 0700)
+
+	maxFileSize := 1000000
+	//maxFileSize := 200000
+
+	createRandomFileSeeded(testDir+"/repository1/file1", maxFileSize,1)
+	createRandomFileSeeded(testDir+"/repository1/file2", maxFileSize,2)
+	createRandomFileSeeded(testDir+"/repository1/dir1/file3", maxFileSize,3)
+
+	threads := 1
+
+	password := "duplicacy"
+
+	// We want deterministic output, plus ability to test encrypted storage
+	// So make unencrypted storage with default keys, and encrypted as bit-identical copy of this but with password
+	unencStorage, err := loadStorage(testDir+"/unenc_storage", threads)
+	if err != nil {
+		t.Errorf("Failed to create storage: %v", err)
+		return
+	}
+	delay := 0
+	if _, ok := unencStorage.(*ACDStorage); ok {
+		delay = 1
+	}
+	if _, ok := unencStorage.(*OneDriveStorage); ok {
+		delay = 5
+	}
+
+	time.Sleep(time.Duration(delay) * time.Second)
+	cleanStorage(unencStorage)
+
+	if !ConfigStorage(unencStorage, 16384, 100, 64*1024, 256*1024, 16*1024, "", nil, false, "", 0, 0) {
+		t.Errorf("Failed to initialize the unencrypted storage")
+	}
+	time.Sleep(time.Duration(delay) * time.Second)
+	unencConfig, _, err := DownloadConfig(unencStorage, "")
+	if err != nil {
+		t.Errorf("Failed to download storage config: %v", err)
+		return
+	}
+
+	// Make encrypted storage
+	storage, err := loadStorage(testDir+"/enc_storage", threads)
+	if err != nil {
+		t.Errorf("Failed to create encrypted storage: %v", err)
+		return
+	}
+	time.Sleep(time.Duration(delay) * time.Second)
+	cleanStorage(storage)
+
+	if !ConfigStorage(storage, 16384, 100, 64*1024, 256*1024, 16*1024, password, unencConfig, true, "", 0, 0) {
+		t.Errorf("Failed to initialize the encrypted storage")
+	}
+	time.Sleep(time.Duration(delay) * time.Second)
+
+	// do unencrypted backup
+	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
+	unencBackupManager := CreateBackupManager("host1", unencStorage, testDir, "", "", "", false)
+	unencBackupManager.SetupSnapshotCache("default")
+
+	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
+	unencBackupManager.Backup(testDir+"/repository1" /*quickMode=*/, true, threads, "first", false, false, 0, false)
+	time.Sleep(time.Duration(delay) * time.Second)
+
+
+	// do encrypted backup
+	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
+	encBackupManager := CreateBackupManager("host1", storage, testDir, password, "", "", false)
+	encBackupManager.SetupSnapshotCache("default")
+
+	SetDuplicacyPreferencePath(testDir + "/repository1/.duplicacy")
+	encBackupManager.Backup(testDir+"/repository1" /*quickMode=*/, true, threads, "first", false, false, 0, false)
+	time.Sleep(time.Duration(delay) * time.Second)
+
+
+	// check snapshots
+	unencBackupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{1} /*tag*/, "",
+		/*showStatistics*/ true /*showTabular*/, false /*checkFiles*/, true /*checkChunks*/, false,
+		/*searchFossils*/ false /*resurrect*/, false, 1 /*allowFailures*/, false)
+
+	encBackupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{1} /*tag*/, "",
+		/*showStatistics*/ true /*showTabular*/, false /*checkFiles*/, true /*checkChunks*/, false,
+		 /*searchFossils*/ false /*resurrect*/, false, 1 /*allowFailures*/, false)
+		
+	// check functions
+	checkAllUncorrupted := func(cmpRepository string) {
+			for _, f := range []string{"file1", "file2", "dir1/file3"} {
+				if _, err := os.Stat(testDir + cmpRepository + "/" + f); os.IsNotExist(err) {
+					t.Errorf("File %s does not exist", f)
+					continue
+				}
+
+				hash1 := getFileHash(testDir + "/repository1/" + f)
+				hash2 := getFileHash(testDir + cmpRepository + "/" + f)
+				if hash1 != hash2 {
+					t.Errorf("File %s has different hashes: %s vs %s", f, hash1, hash2)
+				}
+		}
+	}
+	checkMissingFile := func(cmpRepository string, expectMissing string) {
+			for _, f := range []string{"file1", "file2", "dir1/file3"} {
+				_, err := os.Stat(testDir + cmpRepository + "/" + f)
+				if err==nil {
+					if f==expectMissing {
+						t.Errorf("File %s exists, expected to be missing", f)
+					}
+					continue
+				}
+				if os.IsNotExist(err) {
+					if f!=expectMissing {
+						t.Errorf("File %s does not exist", f)
+					}
+					continue
+				}
+
+				hash1 := getFileHash(testDir + "/repository1/" + f)
+				hash2 := getFileHash(testDir + cmpRepository + "/" + f)
+				if hash1 != hash2 {
+					t.Errorf("File %s has different hashes: %s vs %s", f, hash1, hash2)
+				}
+		}
+	}
+	checkCorruptedFile  := func(cmpRepository string, expectCorrupted string) {
+		for _, f := range []string{"file1", "file2", "dir1/file3"} {
+			if _, err := os.Stat(testDir + cmpRepository + "/" + f); os.IsNotExist(err) {
+				t.Errorf("File %s does not exist", f)
+				continue
+			}
+	
+			hash1 := getFileHash(testDir + "/repository1/" + f)
+			hash2 := getFileHash(testDir + cmpRepository + "/" + f)
+			if (f==expectCorrupted) {
+				if hash1 == hash2 {
+					t.Errorf("File %s has same hashes, expected to be corrupted: %s vs %s", f, hash1, hash2)
+				}
+	
+			} else {
+				if hash1 != hash2 {
+					t.Errorf("File %s has different hashes: %s vs %s", f, hash1, hash2)
+				}
+			}
+		}
+	}
+
+	// test restore all uncorrupted to repository3
+	SetDuplicacyPreferencePath(testDir + "/repository3/.duplicacy")
+	failedFiles := unencBackupManager.Restore(testDir+"/repository3", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, false,
+		 /*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, false)
+	assertRestoreFailures(t, failedFiles, 0)
+	checkAllUncorrupted("/repository3")
+
+	// test for corrupt files and -persist
+	// corrupt a chunk 
+	chunkToCorrupt1 := "/4d/538e5dfd2b08e782bfeb56d1360fb5d7eb9d8c4b2531cc2fca79efbaec910c"
+		// this should affect file1
+	chunkToCorrupt2 := "/2b/f953a766d0196ce026ae259e76e3c186a0e4bcd3ce10f1571d17f86f0a5497"
+		// this should affect dir1/file3
+	
+	for i := 0; i < 2; i++ {
+		if i==0 {
+			// test corrupt chunks
+			corruptFile(testDir+"/unenc_storage"+"/chunks"+chunkToCorrupt1, 128, 128, 4)
+			corruptFile(testDir+"/enc_storage"+"/chunks"+chunkToCorrupt2, 128, 128, 4)
+		} else {
+			// test missing chunks
+			os.Remove(testDir+"/unenc_storage"+"/chunks"+chunkToCorrupt1)
+			os.Remove(testDir+"/enc_storage"+"/chunks"+chunkToCorrupt2)
+		}
+		
+		// check snapshots with --persist (allowFailures == true)
+		// this would cause a panic and os.Exit from duplicacy_log if allowFailures == false
+		unencBackupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{1} /*tag*/, "",
+			/*showStatistics*/ true /*showTabular*/, false /*checkFiles*/, true /*checkChunks*/, false,
+			/*searchFossils*/ false /*resurrect*/, false, 1 /*allowFailures*/, true)
+
+		encBackupManager.SnapshotManager.CheckSnapshots( /*snapshotID*/ "host1" /*revisions*/, []int{1} /*tag*/, "",
+			/*showStatistics*/ true /*showTabular*/, false /*checkFiles*/, true /*checkChunks*/, false,
+			/*searchFossils*/ false /*resurrect*/, false, 1 /*allowFailures*/, true)
+
+		
+		// test restore corrupted, inPlace = true, corrupted files will have hash failures
+		os.RemoveAll(testDir+"/repository2")
+		SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
+		failedFiles = unencBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, false,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 1)
+
+		// check restore, expect file1 to be corrupted
+		checkCorruptedFile("/repository2", "file1")
+
+		
+		os.RemoveAll(testDir+"/repository2")
+		SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
+		failedFiles = encBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, false,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 1)
+
+		// check restore, expect file3 to be corrupted
+		checkCorruptedFile("/repository2", "dir1/file3")
+
+		//SetLoggingLevel(DEBUG)
+		// test restore corrupted, inPlace = false, corrupted files will be missing
+		os.RemoveAll(testDir+"/repository2")
+		SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
+		failedFiles = unencBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, false /*quickMode=*/, false, threads /*overwrite=*/, false,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 1)
+
+		// check restore, expect file1 to be corrupted
+		checkMissingFile("/repository2", "file1")
+
+		
+		os.RemoveAll(testDir+"/repository2")
+		SetDuplicacyPreferencePath(testDir + "/repository2/.duplicacy")
+		failedFiles = encBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, false /*quickMode=*/, false, threads /*overwrite=*/, false,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 1)
+
+		// check restore, expect file3 to be corrupted
+		checkMissingFile("/repository2", "dir1/file3")
+
+		// test restore corrupted files from different backups, inPlace = true
+		// with overwrite=true, corrupted file1 from unenc will be restored correctly from enc
+		// the latter will not touch the existing file3 with correct hash
+		os.RemoveAll(testDir+"/repository2")
+		failedFiles = unencBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, false,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 1)
+
+		failedFiles = encBackupManager.Restore(testDir+"/repository2", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 0)
+		checkAllUncorrupted("/repository2")
+
+		// restore to repository3, with overwrite and allowFailures (true/false), quickMode = false (use hashes)
+		// should always succeed as uncorrupted files already exist with correct hash, so these will be ignored
+		SetDuplicacyPreferencePath(testDir + "/repository3/.duplicacy")
+		failedFiles = unencBackupManager.Restore(testDir+"/repository3", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, false)
+		assertRestoreFailures(t, failedFiles, 0)
+		checkAllUncorrupted("/repository3")
+
+		failedFiles = unencBackupManager.Restore(testDir+"/repository3", threads /*inPlace=*/, true /*quickMode=*/, false, threads /*overwrite=*/, true,
+			/*deleteMode=*/ false /*setowner=*/, false /*showStatistics=*/, false /*patterns=*/, nil /*allowFailures=*/, true)
+		assertRestoreFailures(t, failedFiles, 0)
+		checkAllUncorrupted("/repository3")
+	}
+
+}

src/duplicacy_chunk.go

@@ -22,6 +22,8 @@ import (
 	"runtime"
 
 	"github.com/bkaradzic/go-lz4"
+	"github.com/minio/highwayhash"
+	"github.com/klauspost/reedsolomon"
 )
 
 // A chunk needs to acquire a new buffer and return the old one for every encrypt/decrypt operation, therefore
@@ -65,14 +67,18 @@ type Chunk struct {
 
 	isSnapshot bool // Indicates if the chunk is a snapshot chunk (instead of a file chunk).  This is only used by RSA
 	                // encryption, where a snapshot chunk is not encrypted by RSA
+	
+	isBroken bool // Indicates the chunk did not download correctly. This is only used for -persist (allowFailures) mode
 }
 
 // Magic word to identify a duplicacy format encrypted file, plus a version number.
-var ENCRYPTION_HEADER = "duplicacy\000"
+var ENCRYPTION_BANNER = "duplicacy\000"
 
 // RSA encrypted chunks start with "duplicacy\002"
 var ENCRYPTION_VERSION_RSA byte = 2
 
+var ERASURE_CODING_BANNER = "duplicacy\003"
+
 // CreateChunk creates a new chunk.
 func CreateChunk(config *Config, bufferNeeded bool) *Chunk {
 
@@ -122,6 +128,7 @@ func (chunk *Chunk) Reset(hashNeeded bool) {
 	chunk.id = ""
 	chunk.size = 0
 	chunk.isSnapshot = false
+	chunk.isBroken = false
 }
 
 // Write implements the Writer interface.
@@ -224,7 +231,7 @@ func (chunk *Chunk) Encrypt(encryptionKey []byte, derivationKey string, isSnapsh
 		// Start with the magic number and the version number.
 		if usingRSA {
 			// RSA encryption starts "duplicacy\002"
-			encryptedBuffer.Write([]byte(ENCRYPTION_HEADER)[:len(ENCRYPTION_HEADER) - 1])
+			encryptedBuffer.Write([]byte(ENCRYPTION_BANNER)[:len(ENCRYPTION_BANNER) - 1])
 			encryptedBuffer.Write([]byte{ENCRYPTION_VERSION_RSA})
 
 			// Then the encrypted key
@@ -235,7 +242,7 @@ func (chunk *Chunk) Encrypt(encryptionKey []byte, derivationKey string, isSnapsh
 			binary.Write(encryptedBuffer, binary.LittleEndian, uint16(len(encryptedKey)))
 			encryptedBuffer.Write(encryptedKey)
 		} else {
-			encryptedBuffer.Write([]byte(ENCRYPTION_HEADER))
+			encryptedBuffer.Write([]byte(ENCRYPTION_BANNER))
 		}
 
 		// Followed by the nonce
@@ -248,7 +255,7 @@ func (chunk *Chunk) Encrypt(encryptionKey []byte, derivationKey string, isSnapsh
 		offset = encryptedBuffer.Len()
 	}
 
-	// offset is either 0 or the length of header + nonce
+	// offset is either 0 or the length of banner + nonce
 
 	if chunk.config.CompressionLevel >= -1 && chunk.config.CompressionLevel <= 9 {
 		deflater, _ := zlib.NewWriterLevel(encryptedBuffer, chunk.config.CompressionLevel)
@@ -273,26 +280,79 @@ func (chunk *Chunk) Encrypt(encryptionKey []byte, derivationKey string, isSnapsh
 		return fmt.Errorf("Invalid compression level: %d", chunk.config.CompressionLevel)
 	}
 
-	if len(encryptionKey) == 0 {
-		chunk.buffer, encryptedBuffer = encryptedBuffer, chunk.buffer
-		return nil
+	if len(encryptionKey) > 0 {
+
+		// PKCS7 is used.  The sizes of compressed chunks leak information about the original chunks so we want the padding sizes
+		// to be the maximum allowed by PKCS7
+		dataLength := encryptedBuffer.Len() - offset
+		paddingLength := 256 - dataLength%256
+
+		encryptedBuffer.Write(bytes.Repeat([]byte{byte(paddingLength)}, paddingLength))
+		encryptedBuffer.Write(bytes.Repeat([]byte{0}, gcm.Overhead()))
+
+		// The encrypted data will be appended to the duplicacy banner and the once.
+		encryptedBytes := gcm.Seal(encryptedBuffer.Bytes()[:offset], nonce,
+			encryptedBuffer.Bytes()[offset:offset+dataLength+paddingLength], nil)
+
+		encryptedBuffer.Truncate(len(encryptedBytes))
 	}
 
-	// PKCS7 is used.  Compressed chunk sizes leaks information about the original chunks so we want the padding sizes
-	// to be the maximum allowed by PKCS7
-	dataLength := encryptedBuffer.Len() - offset
-	paddingLength := 256 - dataLength%256
+	if chunk.config.DataShards == 0 || chunk.config.ParityShards == 0 {
+		chunk.buffer, encryptedBuffer = encryptedBuffer, chunk.buffer
+		return
+	}
 
-	encryptedBuffer.Write(bytes.Repeat([]byte{byte(paddingLength)}, paddingLength))
-	encryptedBuffer.Write(bytes.Repeat([]byte{0}, gcm.Overhead()))
+	// Start erasure coding
+	encoder, err := reedsolomon.New(chunk.config.DataShards, chunk.config.ParityShards)
+	if err != nil {
+		return err
+	}
+	chunkSize := len(encryptedBuffer.Bytes())
+	shardSize := (chunkSize + chunk.config.DataShards - 1) / chunk.config.DataShards
+	// Append zeros to make the last shard to have the same size as other
+	encryptedBuffer.Write(make([]byte, shardSize * chunk.config.DataShards - chunkSize))
+	// Grow the buffer for parity shards
+	encryptedBuffer.Grow(shardSize * chunk.config.ParityShards)
+	// Now create one slice for each shard, reusing the data in the buffer
+	data := make([][]byte, chunk.config.DataShards + chunk.config.ParityShards)
+	for i := 0; i < chunk.config.DataShards + chunk.config.ParityShards; i++ {
+		data[i] = encryptedBuffer.Bytes()[i * shardSize: (i + 1) * shardSize]
+	}
+	// This populates the parity shard
+	encoder.Encode(data)
 
-	// The encrypted data will be appended to the duplicacy header and the once.
-	encryptedBytes := gcm.Seal(encryptedBuffer.Bytes()[:offset], nonce,
-		encryptedBuffer.Bytes()[offset:offset+dataLength+paddingLength], nil)
+	// Prepare the chunk to be uploaded
+	chunk.buffer.Reset()
+	// First the banner
+	chunk.buffer.Write([]byte(ERASURE_CODING_BANNER))
+	// Then the header which includes the chunk size, data/parity and a 2-byte checksum
+	header := make([]byte, 14)
+	binary.LittleEndian.PutUint64(header[0:], uint64(chunkSize))
+	binary.LittleEndian.PutUint16(header[8:], uint16(chunk.config.DataShards))
+	binary.LittleEndian.PutUint16(header[10:], uint16(chunk.config.ParityShards))
+	header[12] = header[0] ^ header[2] ^ header[4] ^ header[6] ^ header[8] ^ header[10]
+	header[13] = header[1] ^ header[3] ^ header[5] ^ header[7] ^ header[9] ^ header[11]
+	chunk.buffer.Write(header)
+	// Calculate the highway hash for each shard
+	hashKey := make([]byte, 32)
+	for _, part := range data {
+		hasher, err := highwayhash.New(hashKey)
+		if err != nil {
+			return err
+		}
+		_, err = hasher.Write(part)
+		if err != nil {
+			return err
+		}
+		chunk.buffer.Write(hasher.Sum(nil))
+	}
 
-	encryptedBuffer.Truncate(len(encryptedBytes))
-
-	chunk.buffer, encryptedBuffer = encryptedBuffer, chunk.buffer
+	// Copy the data
+	for _, part := range data {
+		chunk.buffer.Write(part)
+	}
+	// Append the header again for redundancy
+	chunk.buffer.Write(header)
 
 	return nil
 
@@ -322,7 +382,122 @@ func (chunk *Chunk) Decrypt(encryptionKey []byte, derivationKey string) (err err
 	}()
 
 	chunk.buffer, encryptedBuffer = encryptedBuffer, chunk.buffer
-	headerLength := len(ENCRYPTION_HEADER)
+	bannerLength := len(ENCRYPTION_BANNER)
+
+	if len(encryptedBuffer.Bytes()) > bannerLength && string(encryptedBuffer.Bytes()[:bannerLength]) == ERASURE_CODING_BANNER {
+
+		// The chunk was encoded with erasure coding
+		if len(encryptedBuffer.Bytes()) < bannerLength + 14 {
+			return fmt.Errorf("Erasure coding header truncated (%d bytes)", len(encryptedBuffer.Bytes()))
+		}
+		// Check the header checksum
+		header := encryptedBuffer.Bytes()[bannerLength: bannerLength + 14]
+		if header[12] != header[0] ^ header[2] ^ header[4] ^ header[6] ^ header[8] ^ header[10] ||
+		   header[13] != header[1] ^ header[3] ^ header[5] ^ header[7] ^ header[9] ^ header[11] {
+			return fmt.Errorf("Erasure coding header corrupted (%x)", header)
+		}
+
+		// Read the parameters
+		chunkSize := int(binary.LittleEndian.Uint64(header[0:8]))
+		dataShards := int(binary.LittleEndian.Uint16(header[8:10]))
+		parityShards := int(binary.LittleEndian.Uint16(header[10:12]))
+		shardSize := (chunkSize + chunk.config.DataShards - 1) / chunk.config.DataShards
+		// This is the length the chunk file should have
+		expectedLength := bannerLength + 2 * len(header) + (dataShards + parityShards) * (shardSize + 32)
+		// The minimum length that can be recovered from
+		minimumLength := bannerLength + len(header) + (dataShards + parityShards) * 32 + dataShards * shardSize
+		LOG_DEBUG("CHUNK_ERASURECODE", "Chunk size: %d bytes, data size: %d, parity: %d/%d", chunkSize, len(encryptedBuffer.Bytes()), dataShards, parityShards)
+		if len(encryptedBuffer.Bytes()) > expectedLength {
+			LOG_WARN("CHUNK_ERASURECODE", "Chunk has %d bytes (instead of %d)", len(encryptedBuffer.Bytes()), expectedLength)
+		} else if len(encryptedBuffer.Bytes()) == expectedLength {
+			// Correct size; fall through
+		} else if len(encryptedBuffer.Bytes()) > minimumLength {
+			LOG_WARN("CHUNK_ERASURECODE", "Chunk is truncated (%d out of %d bytes)", len(encryptedBuffer.Bytes()), expectedLength)
+		} else {
+			return fmt.Errorf("Not enough chunk data for recovery; chunk size: %d bytes, data size: %d, parity: %d/%d", chunkSize, len(encryptedBuffer.Bytes()), dataShards, parityShards)
+		}
+
+		// Where the hashes start
+		hashOffset := bannerLength + len(header)
+		// Where the data start
+		dataOffset := hashOffset + (dataShards + parityShards) * 32
+
+		data := make([][]byte, dataShards + parityShards)
+		recoveryNeeded := false
+		hashKey := make([]byte, 32)
+		availableShards := 0
+		for i := 0; i < dataShards + parityShards; i++ {
+			start := dataOffset + i * shardSize
+			if start + shardSize > len(encryptedBuffer.Bytes()) {
+				// the current shard is incomplete
+				break
+			}
+			// Now verify the hash
+			hasher, err := highwayhash.New(hashKey)
+			if err != nil {
+				return err
+			}
+			_, err = hasher.Write(encryptedBuffer.Bytes()[start: start + shardSize])
+			if err != nil {
+				return err
+			}
+			if bytes.Compare(hasher.Sum(nil), encryptedBuffer.Bytes()[hashOffset + i * 32: hashOffset + (i + 1) * 32]) != 0 {
+				if i < dataShards {
+					recoveryNeeded = true
+				}
+			} else {
+				// The shard is good
+				data[i] = encryptedBuffer.Bytes()[start: start + shardSize]
+				availableShards++
+				if availableShards >= dataShards {
+					// We have enough shards to recover; skip the remaining shards
+					break
+				}
+			}
+		}
+
+		if !recoveryNeeded {
+			// Remove the padding zeros from the last shard
+			encryptedBuffer.Truncate(dataOffset + chunkSize)
+			// Skip the header and hashes
+			encryptedBuffer.Read(encryptedBuffer.Bytes()[:dataOffset])
+		} else {
+			if availableShards < dataShards {
+				return fmt.Errorf("Not enough chunk data for recover; only %d out of %d shards are complete", availableShards, dataShards + parityShards)
+			}
+
+			// Show the validity of shards using a string of * and -
+			slots := ""
+			for _, part := range data {
+				if len(part) != 0 {
+					slots += "*"
+				} else {
+					slots += "-"
+				}
+			}
+
+			LOG_WARN("CHUNK_ERASURECODE", "Recovering a %d byte chunk from %d byte shards: %s", chunkSize, shardSize, slots)
+			encoder, err := reedsolomon.New(dataShards, parityShards)
+			if err != nil {
+				return err
+			}
+			err = encoder.Reconstruct(data)
+			if err != nil {
+				return err
+			}
+			LOG_DEBUG("CHUNK_ERASURECODE", "Chunk data successfully recovered")
+			buffer := AllocateChunkBuffer()
+			buffer.Reset()
+			for i := 0; i < dataShards; i++ {
+				buffer.Write(data[i])
+			}
+			buffer.Truncate(chunkSize)
+
+			ReleaseChunkBuffer(encryptedBuffer)
+			encryptedBuffer = buffer
+		}
+
+	}
 
 	if len(encryptionKey) > 0 {
 
@@ -340,15 +515,15 @@ func (chunk *Chunk) Decrypt(encryptionKey []byte, derivationKey string) (err err
 			key = hasher.Sum(nil)
 		}
 
-		if len(encryptedBuffer.Bytes()) < headerLength + 12 {
+		if len(encryptedBuffer.Bytes()) < bannerLength + 12 {
 			return fmt.Errorf("No enough encrypted data (%d bytes) provided", len(encryptedBuffer.Bytes()))
 		}
 
-		if string(encryptedBuffer.Bytes()[:headerLength-1]) != ENCRYPTION_HEADER[:headerLength-1] {
+		if string(encryptedBuffer.Bytes()[:bannerLength-1]) != ENCRYPTION_BANNER[:bannerLength-1] {
 			return fmt.Errorf("The storage doesn't seem to be encrypted")
 		}
 
-		encryptionVersion := encryptedBuffer.Bytes()[headerLength-1]
+		encryptionVersion := encryptedBuffer.Bytes()[bannerLength-1]
 		if encryptionVersion != 0 && encryptionVersion != ENCRYPTION_VERSION_RSA {
 			return fmt.Errorf("Unsupported encryption version %d", encryptionVersion)
 		}
@@ -359,14 +534,14 @@ func (chunk *Chunk) Decrypt(encryptionKey []byte, derivationKey string) (err err
 				return fmt.Errorf("An RSA private key is required to decrypt the chunk")
 			}
 
-			encryptedKeyLength := binary.LittleEndian.Uint16(encryptedBuffer.Bytes()[headerLength:headerLength+2])
+			encryptedKeyLength := binary.LittleEndian.Uint16(encryptedBuffer.Bytes()[bannerLength:bannerLength+2])
 
-			if len(encryptedBuffer.Bytes()) < headerLength + 14 + int(encryptedKeyLength) {
+			if len(encryptedBuffer.Bytes()) < bannerLength + 14 + int(encryptedKeyLength) {
 				return fmt.Errorf("No enough encrypted data (%d bytes) provided", len(encryptedBuffer.Bytes()))
 			}
 
-			encryptedKey := encryptedBuffer.Bytes()[headerLength + 2:headerLength + 2 + int(encryptedKeyLength)]
-			headerLength += 2 + int(encryptedKeyLength)
+			encryptedKey := encryptedBuffer.Bytes()[bannerLength + 2:bannerLength + 2 + int(encryptedKeyLength)]
+			bannerLength += 2 + int(encryptedKeyLength)
 
 			decryptedKey, err := rsa.DecryptOAEP(sha256.New(), rand.Reader, chunk.config.rsaPrivateKey, encryptedKey, nil)
 			if err != nil {
@@ -385,8 +560,8 @@ func (chunk *Chunk) Decrypt(encryptionKey []byte, derivationKey string) (err err
 			return err
 		}
 
-		offset = headerLength + gcm.NonceSize()
-		nonce := encryptedBuffer.Bytes()[headerLength:offset]
+		offset = bannerLength + gcm.NonceSize()
+		nonce := encryptedBuffer.Bytes()[bannerLength:offset]
 
 		decryptedBytes, err := gcm.Open(encryptedBuffer.Bytes()[:offset], nonce,
 			encryptedBuffer.Bytes()[offset:], nil)

src/duplicacy_chunk_test.go

@@ -12,7 +12,46 @@ import (
 	"testing"
 )
 
-func TestChunk(t *testing.T) {
+func TestErasureCoding(t *testing.T) {
+	key := []byte("duplicacydefault")
+
+	config := CreateConfig()
+	config.HashKey = key
+	config.IDKey = key
+	config.MinimumChunkSize = 100
+	config.CompressionLevel = DEFAULT_COMPRESSION_LEVEL
+	config.DataShards = 5
+    config.ParityShards = 2
+
+	chunk := CreateChunk(config, true)
+	chunk.Reset(true)
+	data := make([]byte, 100)
+	for i := 0; i < len(data); i++ {
+		data[i] = byte(i)
+	}
+	chunk.Write(data)
+	err := chunk.Encrypt([]byte(""), "", false)
+	if err != nil {
+		t.Errorf("Failed to encrypt the test data: %v", err)
+		return
+	}
+
+	encryptedData := make([]byte, chunk.GetLength())
+	copy(encryptedData, chunk.GetBytes())
+
+	crypto_rand.Read(encryptedData[280:300])
+
+	chunk.Reset(false)
+	chunk.Write(encryptedData)
+	err = chunk.Decrypt([]byte(""), "")
+	if err != nil {
+		t.Errorf("Failed to decrypt the data: %v", err)
+		return
+	}
+	return
+}
+
+func TestChunkBasic(t *testing.T) {
 
 	key := []byte("duplicacydefault")
 
@@ -32,7 +71,10 @@ func TestChunk(t *testing.T) {
 		config.rsaPublicKey = privateKey.Public().(*rsa.PublicKey)
 	}
 
-	remainderLength := -1
+	if testErasureCoding {
+		config.DataShards = 5
+		config.ParityShards = 2
+	}
 
 	for i := 0; i < 500; i++ {
 
@@ -56,10 +98,14 @@ func TestChunk(t *testing.T) {
 		encryptedData := make([]byte, chunk.GetLength())
 		copy(encryptedData, chunk.GetBytes())
 
-		if remainderLength == -1 {
-			remainderLength = len(encryptedData) % 256
-		} else if len(encryptedData)%256 != remainderLength {
-			t.Errorf("Incorrect padding size")
+		if testErasureCoding {
+			offset := 24 + 32 * 7
+			start := rand.Int() % (len(encryptedData) - offset) + offset
+			length := (len(encryptedData) - offset) / 7
+			if start + length > len(encryptedData) {
+				length = len(encryptedData) - start
+			}
+			crypto_rand.Read(encryptedData[start: start+length])
 		}
 
 		chunk.Reset(false)

src/duplicacy_chunkdownloader.go

@@ -31,11 +31,15 @@ type ChunkDownloadCompletion struct {
 // corresponding ChunkDownloadTask is sent to the dowloading goroutine.  Once a chunk is downloaded, it will be
 // inserted in the completed task list.
 type ChunkDownloader struct {
+	totalChunkSize            int64 // Total chunk size
+	downloadedChunkSize       int64 // Downloaded chunk size
+
 	config         *Config      // Associated config
 	storage        Storage      // Download from this storage
 	snapshotCache  *FileStorage // Used as cache if not nil; usually for downloading snapshot chunks
 	showStatistics bool         // Show a stats log for each chunk if true
 	threads        int          // Number of threads
+	allowFailures  bool         // Whether to failfast on download error, or continue
 
 	taskList       []ChunkDownloadTask // The list of chunks to be downloaded
 	completedTasks map[int]bool        // Store downloaded chunks
@@ -46,20 +50,21 @@ type ChunkDownloader struct {
 	completionChannel chan ChunkDownloadCompletion // A downloading goroutine sends back the chunk via this channel after downloading
 
 	startTime                 int64 // The time it starts downloading
-	totalChunkSize            int64 // Total chunk size
-	downloadedChunkSize       int64 // Downloaded chunk size
 	numberOfDownloadedChunks  int   // The number of chunks that have been downloaded
 	numberOfDownloadingChunks int   // The number of chunks still being downloaded
 	numberOfActiveChunks      int   // The number of chunks that is being downloaded or has been downloaded but not reclaimed
+
+	NumberOfFailedChunks      int   // The number of chunks that can't be downloaded
 }
 
-func CreateChunkDownloader(config *Config, storage Storage, snapshotCache *FileStorage, showStatistics bool, threads int) *ChunkDownloader {
+func CreateChunkDownloader(config *Config, storage Storage, snapshotCache *FileStorage, showStatistics bool, threads int, allowFailures bool) *ChunkDownloader {
 	downloader := &ChunkDownloader{
 		config:         config,
 		storage:        storage,
 		snapshotCache:  snapshotCache,
 		showStatistics: showStatistics,
 		threads:        threads,
+		allowFailures:  allowFailures,
 
 		taskList:       nil,
 		completedTasks: make(map[int]bool),
@@ -250,6 +255,9 @@ func (downloader *ChunkDownloader) WaitForChunk(chunkIndex int) (chunk *Chunk) {
 		downloader.taskList[completion.chunkIndex].chunk = completion.chunk
 		downloader.numberOfDownloadedChunks++
 		downloader.numberOfDownloadingChunks--
+		if completion.chunk.isBroken {
+			downloader.NumberOfFailedChunks++
+		}
 	}
 	return downloader.taskList[chunkIndex].chunk
 }
@@ -277,6 +285,9 @@ func (downloader *ChunkDownloader) WaitForCompletion() {
 			downloader.numberOfActiveChunks--
 			downloader.numberOfDownloadedChunks++
 			downloader.numberOfDownloadingChunks--
+			if completion.chunk.isBroken {
+				downloader.NumberOfFailedChunks++
+			}
 		}
 
 		// Pass the tasks one by one to the download queue
@@ -303,7 +314,10 @@ func (downloader *ChunkDownloader) Stop() {
 		downloader.taskList[completion.chunkIndex].chunk = completion.chunk
 		downloader.numberOfDownloadedChunks++
 		downloader.numberOfDownloadingChunks--
-	}
+		if completion.chunk.isBroken {
+			downloader.NumberOfFailedChunks++
+		}
+}
 
 	for i := range downloader.completedTasks {
 		downloader.config.PutChunk(downloader.taskList[i].chunk)
@@ -357,13 +371,22 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 	// will be set up before the encryption
 	chunk.Reset(false)
 
+	// If failures are allowed, complete the task properly
+	completeFailedChunk := func(chunk *Chunk) {
+		if downloader.allowFailures {
+			chunk.isBroken = true
+			downloader.completionChannel <- ChunkDownloadCompletion{chunk: chunk, chunkIndex: task.chunkIndex}
+		}
+	}
+
 	const MaxDownloadAttempts = 3
 	for downloadAttempt := 0; ; downloadAttempt++ {
 
 		// Find the chunk by ID first.
 		chunkPath, exist, _, err := downloader.storage.FindChunk(threadIndex, chunkID, false)
 		if err != nil {
-			LOG_ERROR("DOWNLOAD_CHUNK", "Failed to find the chunk %s: %v", chunkID, err)
+			completeFailedChunk(chunk)
+			LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CHUNK", "Failed to find the chunk %s: %v", chunkID, err)
 			return false
 		}
 
@@ -371,7 +394,8 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 			// No chunk is found.  Have to find it in the fossil pool again.
 			fossilPath, exist, _, err := downloader.storage.FindChunk(threadIndex, chunkID, true)
 			if err != nil {
-				LOG_ERROR("DOWNLOAD_CHUNK", "Failed to find the chunk %s: %v", chunkID, err)
+				completeFailedChunk(chunk)
+				LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CHUNK", "Failed to find the chunk %s: %v", chunkID, err)
 				return false
 			}
 
@@ -393,11 +417,12 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 					continue
 				}
 
+				completeFailedChunk(chunk)
 				// A chunk is not found.  This is a serious error and hopefully it will never happen.
 				if err != nil {
-					LOG_FATAL("DOWNLOAD_CHUNK", "Chunk %s can't be found: %v", chunkID, err)
+					LOG_WERROR(downloader.allowFailures,  "DOWNLOAD_CHUNK", "Chunk %s can't be found: %v", chunkID, err)
 				} else {
-					LOG_FATAL("DOWNLOAD_CHUNK", "Chunk %s can't be found", chunkID)
+					LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CHUNK", "Chunk %s can't be found", chunkID)
 				}
 				return false
 			}
@@ -406,7 +431,8 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 			// downloading again.
 			err = downloader.storage.MoveFile(threadIndex, fossilPath, chunkPath)
 			if err != nil {
-				LOG_FATAL("DOWNLOAD_CHUNK", "Failed to resurrect chunk %s: %v", chunkID, err)
+				completeFailedChunk(chunk)
+				LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CHUNK", "Failed to resurrect chunk %s: %v", chunkID, err)
 				return false
 			}
 
@@ -423,7 +449,8 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 				chunk.Reset(false)
 				continue
 			} else {
-				LOG_ERROR("DOWNLOAD_CHUNK", "Failed to download the chunk %s: %v", chunkID, err)
+				completeFailedChunk(chunk)
+				LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CHUNK", "Failed to download the chunk %s: %v", chunkID, err)
 				return false
 			}
 		}
@@ -435,7 +462,8 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 				chunk.Reset(false)
 				continue
 			} else {
-				LOG_ERROR("DOWNLOAD_DECRYPT", "Failed to decrypt the chunk %s: %v", chunkID, err)
+				completeFailedChunk(chunk)
+				LOG_WERROR(downloader.allowFailures, "DOWNLOAD_DECRYPT", "Failed to decrypt the chunk %s: %v", chunkID, err)
 				return false
 			}
 		}
@@ -447,7 +475,8 @@ func (downloader *ChunkDownloader) Download(threadIndex int, task ChunkDownloadT
 				chunk.Reset(false)
 				continue
 			} else {
-				LOG_FATAL("DOWNLOAD_CORRUPTED", "The chunk %s has a hash id of %s", chunkID, actualChunkID)
+				completeFailedChunk(chunk)
+				LOG_WERROR(downloader.allowFailures, "DOWNLOAD_CORRUPTED", "The chunk %s has a hash id of %s", chunkID, actualChunkID)
 				return false
 			}
 		}

src/duplicacy_chunkuploader_test.go

@@ -101,7 +101,7 @@ func TestUploaderAndDownloader(t *testing.T) {
 
 	chunkUploader.Stop()
 
-	chunkDownloader := CreateChunkDownloader(config, storage, nil, true, testThreads)
+	chunkDownloader := CreateChunkDownloader(config, storage, nil, true, testThreads, false)
 	chunkDownloader.totalChunkSize = int64(totalFileSize)
 
 	for _, chunk := range chunks {

src/duplicacy_config.go

@@ -18,6 +18,7 @@ import (
 	"fmt"
 	"hash"
 	"os"
+	"strings"
 	"runtime"
 	"runtime/debug"
 	"sync/atomic"
@@ -34,8 +35,8 @@ var DEFAULT_KEY = []byte("duplicacy")
 // standard zlib levels of -1 to 9.
 var DEFAULT_COMPRESSION_LEVEL = 100
 
-// The new header of the config file (to differentiate from the old format where the salt and iterations are fixed)
-var CONFIG_HEADER = "duplicacy\001"
+// The new banner of the config file (to differentiate from the old format where the salt and iterations are fixed)
+var CONFIG_BANNER = "duplicacy\001"
 
 // The length of the salt used in the new format
 var CONFIG_SALT_LENGTH = 32
@@ -70,6 +71,10 @@ type Config struct {
 	// for encrypting a non-chunk file
 	FileKey []byte `json:"-"`
 
+	// for erasure coding
+	DataShards int `json:'data-shards'`
+	ParityShards int `json:'parity-shards'`
+
 	// for RSA encryption
 	rsaPrivateKey *rsa.PrivateKey
 	rsaPublicKey *rsa.PublicKey
@@ -180,6 +185,10 @@ func (config *Config) Print() {
 		LOG_TRACE("CONFIG_INFO", "Metadata chunks are encrypted")
 	}
 
+	if config.DataShards != 0 && config.ParityShards != 0 {
+		LOG_TRACE("CONFIG_INFO", "Data shards: %d, parity shards: %d", config.DataShards, config.ParityShards)
+	}
+
 	if config.rsaPublicKey != nil {
 		pkisPublicKey, _ := x509.MarshalPKIXPublicKey(config.rsaPublicKey)
 
@@ -386,11 +395,11 @@ func DownloadConfig(storage Storage, password string) (config *Config, isEncrypt
 		return nil, false, err
 	}
 
-	if len(configFile.GetBytes()) < len(ENCRYPTION_HEADER) {
+	if len(configFile.GetBytes()) < len(ENCRYPTION_BANNER) {
 		return nil, false, fmt.Errorf("The storage has an invalid config file")
 	}
 
-	if string(configFile.GetBytes()[:len(ENCRYPTION_HEADER)-1]) == ENCRYPTION_HEADER[:len(ENCRYPTION_HEADER)-1] && len(password) == 0 {
+	if string(configFile.GetBytes()[:len(ENCRYPTION_BANNER)-1]) == ENCRYPTION_BANNER[:len(ENCRYPTION_BANNER)-1] && len(password) == 0 {
 		return nil, true, fmt.Errorf("The storage is likely to have been initialized with a password before")
 	}
 
@@ -398,23 +407,23 @@ func DownloadConfig(storage Storage, password string) (config *Config, isEncrypt
 
 	if len(password) > 0 {
 
-		if string(configFile.GetBytes()[:len(ENCRYPTION_HEADER)]) == ENCRYPTION_HEADER {
+		if string(configFile.GetBytes()[:len(ENCRYPTION_BANNER)]) == ENCRYPTION_BANNER {
 			// This is the old config format with a static salt and a fixed number of iterations
 			masterKey = GenerateKeyFromPassword(password, DEFAULT_KEY, CONFIG_DEFAULT_ITERATIONS)
 			LOG_TRACE("CONFIG_FORMAT", "Using a static salt and %d iterations for key derivation", CONFIG_DEFAULT_ITERATIONS)
-		} else if string(configFile.GetBytes()[:len(CONFIG_HEADER)]) == CONFIG_HEADER {
+		} else if string(configFile.GetBytes()[:len(CONFIG_BANNER)]) == CONFIG_BANNER {
 			// This is the new config format with a random salt and a configurable number of iterations
 			encryptedLength := len(configFile.GetBytes()) - CONFIG_SALT_LENGTH - 4
 
 			// Extract the salt and the number of iterations
-			saltStart := configFile.GetBytes()[len(CONFIG_HEADER):]
+			saltStart := configFile.GetBytes()[len(CONFIG_BANNER):]
 			iterations := binary.LittleEndian.Uint32(saltStart[CONFIG_SALT_LENGTH : CONFIG_SALT_LENGTH+4])
 			LOG_TRACE("CONFIG_ITERATIONS", "Using %d iterations for key derivation", iterations)
 			masterKey = GenerateKeyFromPassword(password, saltStart[:CONFIG_SALT_LENGTH], int(iterations))
 
-			// Copy to a temporary buffer to replace the header and remove the salt and the number of riterations
+			// Copy to a temporary buffer to replace the banner and remove the salt and the number of riterations
 			var encrypted bytes.Buffer
-			encrypted.Write([]byte(ENCRYPTION_HEADER))
+			encrypted.Write([]byte(ENCRYPTION_BANNER))
 			encrypted.Write(saltStart[CONFIG_SALT_LENGTH+4:])
 
 			configFile.Reset(false)
@@ -423,7 +432,7 @@ func DownloadConfig(storage Storage, password string) (config *Config, isEncrypt
 				LOG_ERROR("CONFIG_DOWNLOAD", "Encrypted config has %d bytes instead of expected %d bytes", len(configFile.GetBytes()), encryptedLength)
 			}
 		} else {
-			return nil, true, fmt.Errorf("The config file has an invalid header")
+			return nil, true, fmt.Errorf("The config file has an invalid banner")
 		}
 
 		// Decrypt the config file.  masterKey == nil means no encryption.
@@ -487,15 +496,15 @@ func UploadConfig(storage Storage, config *Config, password string, iterations i
 			return false
 		}
 
-		// The new encrypted format for config is CONFIG_HEADER + salt + #iterations + encrypted content
+		// The new encrypted format for config is CONFIG_BANNER + salt + #iterations + encrypted content
 		encryptedLength := len(chunk.GetBytes()) + CONFIG_SALT_LENGTH + 4
 
-		// Copy to a temporary buffer to replace the header and add the salt and the number of iterations
+		// Copy to a temporary buffer to replace the banner and add the salt and the number of iterations
 		var encrypted bytes.Buffer
-		encrypted.Write([]byte(CONFIG_HEADER))
+		encrypted.Write([]byte(CONFIG_BANNER))
 		encrypted.Write(salt)
 		binary.Write(&encrypted, binary.LittleEndian, uint32(iterations))
-		encrypted.Write(chunk.GetBytes()[len(ENCRYPTION_HEADER):])
+		encrypted.Write(chunk.GetBytes()[len(ENCRYPTION_BANNER):])
 
 		chunk.Reset(false)
 		chunk.Write(encrypted.Bytes())
@@ -528,7 +537,7 @@ func UploadConfig(storage Storage, config *Config, password string, iterations i
 // it simply creates a file named 'config' that stores various parameters as well as a set of keys if encryption
 // is enabled.
 func ConfigStorage(storage Storage, iterations int, compressionLevel int, averageChunkSize int, maximumChunkSize int,
-	minimumChunkSize int, password string, copyFrom *Config, bitCopy bool, keyFile string) bool {
+	minimumChunkSize int, password string, copyFrom *Config, bitCopy bool, keyFile string, dataShards int, parityShards int) bool {
 
 	exist, _, _, err := storage.GetFileInfo(0, "config")
 	if err != nil {
@@ -550,14 +559,24 @@ func ConfigStorage(storage Storage, iterations int, compressionLevel int, averag
 	if keyFile != "" {
 		config.loadRSAPublicKey(keyFile)
 	}
+
+	config.DataShards = dataShards
+	config.ParityShards = parityShards
+
 	return UploadConfig(storage, config, password, iterations)
 }
 
 func (config *Config) loadRSAPublicKey(keyFile string) {
-	encodedKey, err := ioutil.ReadFile(keyFile)
-	if err != nil {
-		LOG_ERROR("BACKUP_KEY", "Failed to read the public key file: %v", err)
-		return
+	encodedKey := []byte(keyFile)
+	var err error
+
+	// keyFile may be the actually key, in which case we don't need to read from a file
+	if !strings.Contains(keyFile, "-----BEGIN") {
+		encodedKey, err = ioutil.ReadFile(keyFile)
+		if err != nil {
+			LOG_ERROR("BACKUP_KEY", "Failed to read the public key file: %v", err)
+			return
+		}
 	}
 
 	decodedKey, _ := pem.Decode(encodedKey)
@@ -593,10 +612,16 @@ func (config *Config) loadRSAPrivateKey(keyFile string, passphrase string) {
 		return
 	}
 
-	encodedKey, err := ioutil.ReadFile(keyFile)
-	if err != nil {
-		LOG_ERROR("RSA_PRIVATE", "Failed to read the private key file: %v", err)
-		return
+	encodedKey := []byte(keyFile)
+	var err error
+
+	// keyFile may be the actually key, in which case we don't need to read from a file
+	if !strings.Contains(keyFile, "-----BEGIN") {
+		encodedKey, err = ioutil.ReadFile(keyFile)
+		if err != nil {
+			LOG_ERROR("RSA_PRIVATE", "Failed to read the private key file: %v", err)
+			return
+		}
 	}
 
 	decodedKey, _ := pem.Decode(encodedKey)

src/duplicacy_entry.go

@@ -443,7 +443,7 @@ func (files FileInfoCompare) Less(i, j int) bool {
 
 // ListEntries returns a list of entries representing file and subdirectories under the directory 'path'.  Entry paths
 // are normalized as relative to 'top'.  'patterns' are used to exclude or include certain files.
-func ListEntries(top string, path string, fileList *[]*Entry, patterns []string, nobackupFile string, discardAttributes bool) (directoryList []*Entry,
+func ListEntries(top string, path string, fileList *[]*Entry, patterns []string, nobackupFile string, discardAttributes bool, excludeByAttribute bool) (directoryList []*Entry,
 	skippedFiles []string, err error) {
 
 	LOG_DEBUG("LIST_ENTRIES", "Listing %s", path)
@@ -524,6 +524,11 @@ func ListEntries(top string, path string, fileList *[]*Entry, patterns []string,
 			entry.ReadAttributes(top)
 		}
 
+		if excludeByAttribute && excludedByAttribute(entry.Attributes) {
+			LOG_DEBUG("LIST_EXCLUDE", "%s is excluded by attribute", entry.Path)
+			continue
+		}
+
 		if f.Mode()&(os.ModeNamedPipe|os.ModeSocket|os.ModeDevice) != 0 {
 			LOG_WARN("LIST_SKIP", "Skipped non-regular file %s", entry.Path)
 			skippedFiles = append(skippedFiles, entry.Path)

src/duplicacy_entry_test.go

@@ -9,8 +9,12 @@ import (
 	"math/rand"
 	"os"
 	"path/filepath"
+	"runtime"
 	"sort"
+	"strings"
 	"testing"
+
+	"github.com/gilbertchen/xattr"
 )
 
 func TestEntrySort(t *testing.T) {
@@ -173,7 +177,7 @@ func TestEntryList(t *testing.T) {
 		directory := directories[len(directories)-1]
 		directories = directories[:len(directories)-1]
 		entries = append(entries, directory)
-		subdirectories, _, err := ListEntries(testDir, directory.Path, &entries, nil, "", false)
+		subdirectories, _, err := ListEntries(testDir, directory.Path, &entries, nil, "", false, false)
 		if err != nil {
 			t.Errorf("ListEntries(%s, %s) returned an error: %s", testDir, directory.Path, err)
 		}
@@ -216,3 +220,110 @@ func TestEntryList(t *testing.T) {
 	}
 
 }
+
+// TestEntryExcludeByAttribute tests the excludeByAttribute parameter to the ListEntries function
+func TestEntryExcludeByAttribute(t *testing.T) {
+
+	if !(runtime.GOOS == "darwin" || runtime.GOOS == "linux") {
+		t.Skip("skipping test not darwin or linux")
+	}
+
+	testDir := filepath.Join(os.TempDir(), "duplicacy_test")
+
+	os.RemoveAll(testDir)
+	os.MkdirAll(testDir, 0700)
+
+	// Files or folders named with "exclude" below will have the exclusion attribute set on them
+	// When ListEntries is called with excludeByAttribute true, they should be excluded.
+	DATA := [...]string{
+		"excludefile",
+		"includefile",
+		"excludedir/",
+		"excludedir/file",
+		"includedir/",
+		"includedir/includefile",
+		"includedir/excludefile",
+	}
+
+	for _, file := range DATA {
+		fullPath := filepath.Join(testDir, file)
+		if file[len(file)-1] == '/' {
+			err := os.Mkdir(fullPath, 0700)
+			if err != nil {
+				t.Errorf("Mkdir(%s) returned an error: %s", fullPath, err)
+			}
+			continue
+		}
+
+		err := ioutil.WriteFile(fullPath, []byte(file), 0700)
+		if err != nil {
+			t.Errorf("WriteFile(%s) returned an error: %s", fullPath, err)
+		}
+	}
+
+	for _, file := range DATA {
+		fullPath := filepath.Join(testDir, file)
+		if strings.Contains(file, "exclude") {
+			xattr.Setxattr(fullPath, "com.apple.metadata:com_apple_backup_excludeItem", []byte("com.apple.backupd"))
+		}
+	}
+
+	for _, excludeByAttribute := range [2]bool{true, false} {
+		t.Logf("testing excludeByAttribute: %t", excludeByAttribute)
+		directories := make([]*Entry, 0, 4)
+		directories = append(directories, CreateEntry("", 0, 0, 0))
+
+		entries := make([]*Entry, 0, 4)
+
+		for len(directories) > 0 {
+			directory := directories[len(directories)-1]
+			directories = directories[:len(directories)-1]
+			entries = append(entries, directory)
+			subdirectories, _, err := ListEntries(testDir, directory.Path, &entries, nil, "", false, excludeByAttribute)
+			if err != nil {
+				t.Errorf("ListEntries(%s, %s) returned an error: %s", testDir, directory.Path, err)
+			}
+			directories = append(directories, subdirectories...)
+		}
+
+		entries = entries[1:]
+
+		for _, entry := range entries {
+			t.Logf("entry: %s", entry.Path)
+		}
+
+		i := 0
+		for _, file := range DATA {
+			entryFound := false
+			var entry *Entry
+			for _, entry = range entries {
+				if entry.Path == file {
+					entryFound = true
+					break
+				}
+			}
+
+			if excludeByAttribute && strings.Contains(file, "exclude") {
+				if entryFound {
+					t.Errorf("file: %s, expected to be excluded but wasn't. attributes: %v", file, entry.Attributes)
+					i++
+				} else {
+					t.Logf("file: %s, excluded", file)
+				}
+			} else {
+				if entryFound {
+					t.Logf("file: %s, included. attributes: %v", file, entry.Attributes)
+					i++
+				} else {
+					t.Errorf("file: %s, expected to be included but wasn't", file)
+				}
+			}
+		}
+
+	}
+
+	if !t.Failed() {
+		os.RemoveAll(testDir)
+	}
+
+}

src/duplicacy_filestorage.go

@@ -79,7 +79,7 @@ func (storage *FileStorage) ListFiles(threadIndex int, dir string) (files []stri
 
 	for _, f := range list {
 		name := f.Name()
-		if f.IsDir() && name[len(name)-1] != '/' {
+		if (f.IsDir() || f.Mode() & os.ModeSymlink != 0) && name[len(name)-1] != '/' {
 			name += "/"
 		}
 		files = append(files, name)
@@ -165,8 +165,8 @@ func (storage *FileStorage) UploadFile(threadIndex int, filePath string, content
 				return err
 			}
 		} else {
-			if !stat.IsDir() {
-				return fmt.Errorf("The path %s is not a directory", dir)
+			if !stat.IsDir() && stat.Mode() & os.ModeSymlink == 0 {
+				return fmt.Errorf("The path %s is not a directory or symlink", dir)
 			}
 		}
 	}

src/duplicacy_log.go

@@ -107,6 +107,15 @@ func LOG_ERROR(logID string, format string, v ...interface{}) {
 	logf(ERROR, logID, format, v...)
 }
 
+func LOG_WERROR(isWarning bool, logID string, format string, v ...interface{}) {
+	if isWarning {
+		logf(WARN, logID, format, v...)
+	} else {
+		logf(ERROR, logID, format, v...)
+	}
+}
+
+
 func LOG_FATAL(logID string, format string, v ...interface{}) {
 	logf(FATAL, logID, format, v...)
 }

src/duplicacy_oneclient.go

@@ -13,6 +13,7 @@ import (
 	"math/rand"
 	"net/http"
 	"strings"
+	"strconv"
 	"sync"
 	"time"
 	"path/filepath"
@@ -86,7 +87,7 @@ func (client *OneDriveClient) call(url string, method string, input interface{},
 	var response *http.Response
 
 	backoff := 1
-	for i := 0; i < 8; i++ {
+	for i := 0; i < 12; i++ {
 
 		LOG_DEBUG("ONEDRIVE_CALL", "%s %s", method, url)
 
@@ -129,6 +130,8 @@ func (client *OneDriveClient) call(url string, method string, input interface{},
 			request.Header.Set("Content-Type", contentType)
 		}
 
+		request.Header.Set("User-Agent", "ISV|Acrosync|Duplicacy/2.0")
+
 		response, err = client.HTTPClient.Do(request)
 		if err != nil {
 			if client.IsConnected {
@@ -145,6 +148,9 @@ func (client *OneDriveClient) call(url string, method string, input interface{},
 					time.Sleep(retryAfter * time.Millisecond)
 				}
 				backoff *= 2
+				if backoff > 256 {
+					backoff = 256
+				}
 				continue
 			}
 			return nil, 0, err
@@ -176,10 +182,20 @@ func (client *OneDriveClient) call(url string, method string, input interface{},
 		} else if response.StatusCode == 409 {
 			return nil, 0, OneDriveError{Status: response.StatusCode, Message: "Conflict"}
 		} else if response.StatusCode > 401 && response.StatusCode != 404 {
-			retryAfter := time.Duration(rand.Float32() * 1000.0 * float32(backoff))
-			LOG_INFO("ONEDRIVE_RETRY", "Response code: %d; retry after %d milliseconds", response.StatusCode, retryAfter)
-			time.Sleep(retryAfter * time.Millisecond)
+			delay := int((rand.Float32() * 0.5 + 0.5) * 1000.0 * float32(backoff))
+			if backoffList, found := response.Header["Retry-After"]; found && len(backoffList) > 0 {
+				retryAfter, _ := strconv.Atoi(backoffList[0])
+				if retryAfter * 1000 > delay {
+					delay = retryAfter * 1000
+				}
+			}
+
+			LOG_INFO("ONEDRIVE_RETRY", "Response code: %d; retry after %d milliseconds", response.StatusCode, delay)
+			time.Sleep(time.Duration(delay) * time.Millisecond)
 			backoff *= 2
+			if backoff > 256 {
+				backoff = 256
+			}
 			continue
 		} else {
 			if err := json.NewDecoder(response.Body).Decode(errorResponse); err != nil {
@@ -314,7 +330,7 @@ func (client *OneDriveClient) UploadFile(path string, content []byte, rateLimit
 
 	// Upload file using the simple method; this is only possible for OneDrive Personal or if the file
 	// is smaller than 4MB for OneDrive Business
-	if !client.IsBusiness || len(content) < 4 * 1024 * 1024 || (client.TestMode && rand.Int() % 2 == 0) {
+	if !client.IsBusiness || (client.TestMode && rand.Int() % 2 == 0) {
 		url := client.APIURL + "/drive/root:/" + path + ":/content"
 
 		readCloser, _, err := client.call(url, "PUT", CreateRateLimitedReader(content, rateLimit), "application/octet-stream")

src/duplicacy_preference.go

@@ -25,7 +25,8 @@ type Preference struct {
 	DoNotSavePassword bool              `json:"no_save_password"`
 	NobackupFile      string            `json:"nobackup_file"`
 	Keys              map[string]string `json:"keys"`
-	FiltersFile           string        `json:"filters"`
+	FiltersFile       string            `json:"filters"`
+	ExcludeByAttribute bool             `json:"exclude_by_attribute"`
 }
 
 var preferencePath string

src/duplicacy_sftpstorage.go

@@ -318,7 +318,11 @@ func (storage *SFTPStorage) UploadFile(threadIndex int, filePath string, content
 			file.Close()
 			return err
 		}
-		file.Close()
+
+		err = file.Close()
+		if err != nil {
+			return err
+		}
 
 		err = storage.getSFTPClient().Rename(temporaryFile, fullPath)
 		if err != nil {

src/duplicacy_snapshot.go

@@ -58,7 +58,7 @@ func CreateEmptySnapshot(id string) (snapshto *Snapshot) {
 
 // CreateSnapshotFromDirectory creates a snapshot from the local directory 'top'.  Only 'Files'
 // will be constructed, while 'ChunkHashes' and 'ChunkLengths' can only be populated after uploading.
-func CreateSnapshotFromDirectory(id string, top string, nobackupFile string, filtersFile string) (snapshot *Snapshot, skippedDirectories []string,
+func CreateSnapshotFromDirectory(id string, top string, nobackupFile string, filtersFile string, excludeByAttribute bool) (snapshot *Snapshot, skippedDirectories []string,
 	skippedFiles []string, err error) {
 
 	snapshot = &Snapshot{
@@ -89,7 +89,7 @@ func CreateSnapshotFromDirectory(id string, top string, nobackupFile string, fil
 		directory := directories[len(directories)-1]
 		directories = directories[:len(directories)-1]
 		snapshot.Files = append(snapshot.Files, directory)
-		subdirectories, skipped, err := ListEntries(top, directory.Path, &snapshot.Files, patterns, nobackupFile, snapshot.discardAttributes)
+		subdirectories, skipped, err := ListEntries(top, directory.Path, &snapshot.Files, patterns, nobackupFile, snapshot.discardAttributes, excludeByAttribute)
 		if err != nil {
 			if directory.Path == "" {
 				LOG_ERROR("LIST_FAILURE", "Failed to list the repository root: %v", err)

src/duplicacy_snapshotmanager.go

@@ -270,7 +270,7 @@ func (reader *sequenceReader) Read(data []byte) (n int, err error) {
 
 func (manager *SnapshotManager) CreateChunkDownloader() {
 	if manager.chunkDownloader == nil {
-		manager.chunkDownloader = CreateChunkDownloader(manager.config, manager.storage, manager.snapshotCache, false, 1)
+		manager.chunkDownloader = CreateChunkDownloader(manager.config, manager.storage, manager.snapshotCache, false, 1, false)
 	}
 }
 
@@ -809,9 +809,9 @@ func (manager *SnapshotManager) ListSnapshots(snapshotID string, revisionsToList
 
 // ListSnapshots shows the information about a snapshot.
 func (manager *SnapshotManager) CheckSnapshots(snapshotID string, revisionsToCheck []int, tag string, showStatistics bool, showTabular bool,
-	checkFiles bool, checkChunks, searchFossils bool, resurrect bool, threads int) bool {
+	checkFiles bool, checkChunks, searchFossils bool, resurrect bool, threads int, allowFailures bool) bool {
 
-	manager.chunkDownloader = CreateChunkDownloader(manager.config, manager.storage, manager.snapshotCache, false, threads)
+	manager.chunkDownloader = CreateChunkDownloader(manager.config, manager.storage, manager.snapshotCache, false, threads, allowFailures)
 
 	LOG_DEBUG("LIST_PARAMETERS", "id: %s, revisions: %v, tag: %s, showStatistics: %t, showTabular: %t, checkFiles: %t, searchFossils: %t, resurrect: %t",
 		snapshotID, revisionsToCheck, tag, showStatistics, showTabular, checkFiles, searchFossils, resurrect)
@@ -845,7 +845,7 @@ func (manager *SnapshotManager) CheckSnapshots(snapshotID string, revisionsToChe
 		chunk = strings.Replace(chunk, "/", "", -1)
 		chunkSizeMap[chunk] = allSizes[i]
 
-		if allSizes[i] == 0 {
+		if allSizes[i] == 0 && !strings.HasSuffix(chunk, ".tmp") {
 			LOG_WARN("SNAPSHOT_CHECK", "Chunk %s has a size of 0", chunk)
 			emptyChunks++
 		}
@@ -1020,11 +1020,46 @@ func (manager *SnapshotManager) CheckSnapshots(snapshotID string, revisionsToChe
 	if checkChunks && !checkFiles {
 		manager.chunkDownloader.snapshotCache = nil
 		LOG_INFO("SNAPSHOT_VERIFY", "Verifying %d chunks", len(*allChunkHashes))
+
+		startTime := time.Now()
+		var chunkHashes []string
+
+		// The index of the first chunk to add to the downloader, which may have already downloaded
+		// some metadata chunks so the index doesn't start with 0.
+		chunkIndex := -1
+
 		for chunkHash := range *allChunkHashes {
-			manager.chunkDownloader.AddChunk(chunkHash)
+			chunkHashes = append(chunkHashes, chunkHash)
+			if chunkIndex == -1 {
+				chunkIndex = manager.chunkDownloader.AddChunk(chunkHash)
+			} else {
+				manager.chunkDownloader.AddChunk(chunkHash)
+			}
+		}
+
+		var downloadedChunkSize int64
+		totalChunks := len(*allChunkHashes)
+		for i := 0; i < totalChunks; i++ {
+			chunk := manager.chunkDownloader.WaitForChunk(i + chunkIndex)
+			if chunk.isBroken {
+				continue
+			}
+			downloadedChunkSize += int64(chunk.GetLength())
+
+			elapsedTime := time.Now().Sub(startTime).Seconds()
+			speed := int64(float64(downloadedChunkSize) / elapsedTime)
+			remainingTime := int64(float64(totalChunks - i - 1) / float64(i + 1) * elapsedTime)
+			percentage := float64(i + 1) / float64(totalChunks) * 100.0
+			LOG_INFO("VERIFY_PROGRESS", "Verified chunk %s (%d/%d), %sB/s %s %.1f%%",
+					 manager.config.GetChunkIDFromHash(chunkHashes[i]), i + 1, totalChunks,
+					 PrettySize(speed), PrettyTime(remainingTime), percentage)
+		}
+
+		if manager.chunkDownloader.NumberOfFailedChunks > 0 {
+			LOG_ERROR("SNAPSHOT_VERIFY", "%d out of %d chunks are corrupted", manager.chunkDownloader.NumberOfFailedChunks, totalChunks)
+		} else {
+			LOG_INFO("SNAPSHOT_VERIFY", "All %d chunks have been successfully verified", totalChunks)
 		}
-		manager.chunkDownloader.WaitForCompletion()
-		LOG_INFO("SNAPSHOT_VERIFY", "All %d chunks have been successfully verified", len(*allChunkHashes))
 	}
 	return true
 }
@@ -1379,7 +1414,7 @@ func (manager *SnapshotManager) PrintFile(snapshotID string, revision int, path
 
 // Diff compares two snapshots, or two revision of a file if the file argument is given.
 func (manager *SnapshotManager) Diff(top string, snapshotID string, revisions []int,
-	filePath string, compareByHash bool, nobackupFile string, filtersFile string) bool {
+	filePath string, compareByHash bool, nobackupFile string, filtersFile string, excludeByAttribute bool) bool {
 
 	LOG_DEBUG("DIFF_PARAMETERS", "top: %s, id: %s, revision: %v, path: %s, compareByHash: %t",
 		top, snapshotID, revisions, filePath, compareByHash)
@@ -1392,7 +1427,7 @@ func (manager *SnapshotManager) Diff(top string, snapshotID string, revisions []
 	if len(revisions) <= 1 {
 		// Only scan the repository if filePath is not provided
 		if len(filePath) == 0 {
-			rightSnapshot, _, _, err = CreateSnapshotFromDirectory(snapshotID, top, nobackupFile, filtersFile)
+			rightSnapshot, _, _, err = CreateSnapshotFromDirectory(snapshotID, top, nobackupFile, filtersFile, excludeByAttribute)
 			if err != nil {
 				LOG_ERROR("SNAPSHOT_LIST", "Failed to list the directory %s: %v", top, err)
 				return false

src/duplicacy_snapshotmanager_test.go

@@ -620,7 +620,7 @@ func TestPruneNewSnapshots(t *testing.T) {
 	// Now chunkHash1 wil be resurrected
 	snapshotManager.PruneSnapshots("vm1@host1", "vm1@host1", []int{}, []string{}, []string{}, false, false, []string{}, false, false, false, 1)
 	checkTestSnapshots(snapshotManager, 4, 0)
-	snapshotManager.CheckSnapshots("vm1@host1", []int{2, 3}, "", false, false, false, false, false, false, 1)
+	snapshotManager.CheckSnapshots("vm1@host1", []int{2, 3}, "", false, false, false, false, false, false, 1, false)
 }
 
 // A fossil collection left by an aborted prune should be ignored if any supposedly deleted snapshot exists
@@ -669,7 +669,7 @@ func TestPruneGhostSnapshots(t *testing.T) {
 	// Run the prune again but the fossil collection should be igored, since revision 1 still exists
 	snapshotManager.PruneSnapshots("vm1@host1", "vm1@host1", []int{}, []string{}, []string{}, false, false, []string{}, false, false, false, 1)
 	checkTestSnapshots(snapshotManager, 3, 2)
-	snapshotManager.CheckSnapshots("vm1@host1", []int{1, 2, 3}, "", false, false, false, false, true /*searchFossils*/, false, 1)
+	snapshotManager.CheckSnapshots("vm1@host1", []int{1, 2, 3}, "", false, false, false, false, true /*searchFossils*/, false, 1, false)
 
 	// Prune snapshot 1 again
 	snapshotManager.PruneSnapshots("vm1@host1", "vm1@host1", []int{1}, []string{}, []string{}, false, false, []string{}, false, false, false, 1)
@@ -683,5 +683,5 @@ func TestPruneGhostSnapshots(t *testing.T) {
 	// Run the prune again and this time the fossil collection will be processed and the fossils removed
 	snapshotManager.PruneSnapshots("vm1@host1", "vm1@host1", []int{}, []string{}, []string{}, false, false, []string{}, false, false, false, 1)
 	checkTestSnapshots(snapshotManager, 3, 0)
-	snapshotManager.CheckSnapshots("vm1@host1", []int{2, 3, 4}, "", false, false, false, false, false, false, 1)
+	snapshotManager.CheckSnapshots("vm1@host1", []int{2, 3, 4}, "", false, false, false, false, false, false, 1, false)
 }

src/duplicacy_storage_test.go

@@ -28,6 +28,7 @@ var testQuickMode bool
 var testThreads int
 var testFixedChunkSize bool
 var testRSAEncryption bool
+var testErasureCoding bool
 
 func init() {
 	flag.StringVar(&testStorageName, "storage", "", "the test storage to use")
@@ -36,6 +37,7 @@ func init() {
 	flag.IntVar(&testThreads, "threads", 1, "number of downloading/uploading threads")
 	flag.BoolVar(&testFixedChunkSize, "fixed-chunk-size", false, "fixed chunk size")
 	flag.BoolVar(&testRSAEncryption, "rsa", false, "enable RSA encryption")
+	flag.BoolVar(&testErasureCoding, "erasure-coding", false, "enable Erasure Coding")
 	flag.Parse()
 }
 

src/duplicacy_swiftstorage.go

@@ -129,6 +129,11 @@ func CreateSwiftStorage(storageURL string, key string, threads int) (storage *Sw
 		TrustId:        arguments["trust_id"],
 	}
 
+	err = connection.Authenticate()
+	if err != nil {
+		return nil, err
+	}
+
 	_, _, err = connection.Container(container)
 	if err != nil {
 		return nil, err

src/duplicacy_utils.go

@@ -434,7 +434,7 @@ func PrettyTime(seconds int64) string {
 			seconds/day, (seconds%day)/3600, (seconds%3600)/60, seconds%60)
 	} else if seconds > day {
 		return fmt.Sprintf("1 day %02d:%02d:%02d", (seconds%day)/3600, (seconds%3600)/60, seconds%60)
-	} else if seconds > 0 {
+	} else if seconds >= 0 {
 		return fmt.Sprintf("%02d:%02d:%02d", seconds/3600, (seconds%3600)/60, seconds%60)
 	} else {
 		return "n/a"

src/duplicacy_utils_darwin.go

@@ -0,0 +1,14 @@
+// Copyright (c) Acrosync LLC. All rights reserved.
+// Free for personal use and commercial trial
+// Commercial use requires per-user licenses available from https://duplicacy.com
+
+package duplicacy
+
+import (
+	"strings"
+)
+
+func excludedByAttribute(attirbutes map[string][]byte) bool {
+	value, ok := attirbutes["com.apple.metadata:com_apple_backup_excludeItem"]
+	return ok && strings.Contains(string(value), "com.apple.backupd")
+}

src/duplicacy_utils_freebsd.go

@@ -0,0 +1,13 @@
+// Copyright (c) Acrosync LLC. All rights reserved.
+// Free for personal use and commercial trial
+// Commercial use requires per-user licenses available from https://duplicacy.com
+
+package duplicacy
+
+import (
+)
+
+func excludedByAttribute(attirbutes map[string][]byte) bool {
+	_, ok := attirbutes["duplicacy_exclude"]
+	return ok
+}

src/duplicacy_utils_linux.go

@@ -0,0 +1,13 @@
+// Copyright (c) Acrosync LLC. All rights reserved.
+// Free for personal use and commercial trial
+// Commercial use requires per-user licenses available from https://duplicacy.com
+
+package duplicacy
+
+import (
+)
+
+func excludedByAttribute(attirbutes map[string][]byte) bool {
+	_, ok := attirbutes["duplicacy_exclude"]
+	return ok
+}

src/duplicacy_utils_windows.go

@@ -131,3 +131,7 @@ func SplitDir(fullPath string) (dir string, file string) {
 	i := strings.LastIndex(fullPath, "\\")
 	return fullPath[:i+1], fullPath[i+1:]
 }
+
+func excludedByAttribute(attirbutes map[string][]byte) bool {
+	return false
+}

src/duplicacy_webdavstorage.go

@@ -21,6 +21,7 @@ import (
 	"strings"
 	"sync"
 	"time"
+	"io/ioutil"
 )
 
 type WebDAVStorage struct {
@@ -127,7 +128,12 @@ func (storage *WebDAVStorage) sendRequest(method string, uri string, depth int,
 			dataReader = bytes.NewReader(data)
 		} else if method == "PUT" {
 			headers["Content-Type"] = "application/octet-stream"
-			dataReader = CreateRateLimitedReader(data, storage.UploadRateLimit/storage.threads)
+			headers["Content-Length"] = fmt.Sprintf("%d", len(data))
+			if storage.UploadRateLimit <= 0 {
+				dataReader = bytes.NewReader(data)
+			} else {
+				dataReader = CreateRateLimitedReader(data, storage.UploadRateLimit/storage.threads)
+			}
 		} else if method == "MOVE" {
 			headers["Destination"] = storage.createConnectionString(string(data))
 			headers["Content-Type"] = "application/octet-stream"
@@ -159,7 +165,7 @@ func (storage *WebDAVStorage) sendRequest(method string, uri string, depth int,
 
 		response, err := storage.client.Do(request)
 		if err != nil {
-			LOG_TRACE("WEBDAV_RETRY", "URL request '%s %s' returned an error (%v)", method, uri, err)
+			LOG_TRACE("WEBDAV_ERROR", "URL request '%s %s' returned an error (%v)", method, uri, err)
 			backoff = storage.retry(backoff)
 			continue
 		}
@@ -168,11 +174,13 @@ func (storage *WebDAVStorage) sendRequest(method string, uri string, depth int,
 			return response.Body, response.Header, nil
 		}
 
+		io.Copy(ioutil.Discard, response.Body)
+		response.Body.Close()
+
 		if response.StatusCode == 301 {
 			return nil, nil, errWebDAVMovedPermanently
 		}
 
-		response.Body.Close()
 		if response.StatusCode == 404 {
 			// Retry if it is UPLOAD, otherwise return immediately
 			if method != "PUT" {
@@ -228,6 +236,7 @@ func (storage *WebDAVStorage) getProperties(uri string, depth int, properties ..
 			return nil, err
 		}
 		defer readCloser.Close()
+		defer io.Copy(ioutil.Discard, readCloser)
 
 		object := WebDAVMultiStatus{}
 		err = xml.NewDecoder(readCloser).Decode(&object)
@@ -311,6 +320,12 @@ func (storage *WebDAVStorage) ListFiles(threadIndex int, dir string) (files []st
 			}
 			files = append(files, file)
 			sizes = append(sizes, int64(0))
+
+			// Add the directory to the directory cache
+			storage.directoryCacheLock.Lock()
+			storage.directoryCache[dir + file] = 1
+			storage.directoryCacheLock.Unlock()
+
 		}
 	}
 
@@ -357,6 +372,7 @@ func (storage *WebDAVStorage) DeleteFile(threadIndex int, filePath string) (err
 	if err != nil {
 		return err
 	}
+	io.Copy(ioutil.Discard, readCloser)
 	readCloser.Close()
 	return nil
 }
@@ -367,6 +383,7 @@ func (storage *WebDAVStorage) MoveFile(threadIndex int, from string, to string)
 	if err != nil {
 		return err
 	}
+	io.Copy(ioutil.Discard, readCloser)
 	readCloser.Close()
 	return nil
 }
@@ -380,21 +397,7 @@ func (storage *WebDAVStorage) createParentDirectory(threadIndex int, dir string)
 	}
 	parent := dir[:found]
 
-	storage.directoryCacheLock.Lock()
-	_, exist := storage.directoryCache[parent]
-	storage.directoryCacheLock.Unlock()
-
-	if exist {
-		return nil
-	}
-
-	err = storage.CreateDirectory(threadIndex, parent)
-	if err == nil {
-		storage.directoryCacheLock.Lock()
-		storage.directoryCache[parent] = 1
-		storage.directoryCacheLock.Unlock()
-	}
-	return err
+	return storage.CreateDirectory(threadIndex, parent)
 }
 
 // CreateDirectory creates a new directory.
@@ -407,18 +410,35 @@ func (storage *WebDAVStorage) CreateDirectory(threadIndex int, dir string) (err
 		return nil
 	}
 
+	storage.directoryCacheLock.Lock()
+	_, exist := storage.directoryCache[dir]
+	storage.directoryCacheLock.Unlock()
+
+	if exist {
+		return nil
+	}
+
 	// If there is an error in creating the parent directory, proceed anyway
 	storage.createParentDirectory(threadIndex, dir)
 
 	readCloser, _, err := storage.sendRequest("MKCOL", dir, 0, []byte(""))
 	if err != nil {
-		if err == errWebDAVMethodNotAllowed || err == errWebDAVMovedPermanently {
+		if err == errWebDAVMethodNotAllowed || err == errWebDAVMovedPermanently || err == io.EOF {
 			// We simply ignore these errors and assume that the directory already exists
+			LOG_TRACE("WEBDAV_MKDIR", "Can't create directory %s: %v; error ignored", dir, err)
+			storage.directoryCacheLock.Lock()
+			storage.directoryCache[dir] = 1
+			storage.directoryCacheLock.Unlock()
 			return nil
 		}
 		return err
 	}
+	io.Copy(ioutil.Discard, readCloser)
 	readCloser.Close()
+
+	storage.directoryCacheLock.Lock()
+	storage.directoryCache[dir] = 1
+	storage.directoryCacheLock.Unlock()
 	return nil
 }
 
@@ -443,6 +463,7 @@ func (storage *WebDAVStorage) UploadFile(threadIndex int, filePath string, conte
 	if err != nil {
 		return err
 	}
+	io.Copy(ioutil.Discard, readCloser)
 	readCloser.Close()
 	return nil
 }