1BORG(1) Borg - Deduplicating Archiver BORG(1)
2
6 borg - BorgBackup is a deduplicating backup program with optional com‐
7 pression and authenticated encryption.
8
10 Borg consists of a number of commands. Each command accepts a number of
11 arguments and options and interprets various environment variables.
12 The following sections will describe each command in detail.
13 Commands, options, parameters, paths and such are set in fixed-width.
15 Option values are underlined. Borg has few options accepting a fixed
16 set of values (e.g. --encryption of borg_init). Experimental features
17 are marked with red stripes on the sides, like this paragraph.
18 Experimental features are not stable, which means that they may be
20 changed in incompatible ways or even removed entirely without prior
21 notice in following releases.
22 Positional Arguments and Options: Order matters
24 Borg only supports taking options (-s and --progress in the example) to
25 the left or right of all positional arguments (repo::archive and path
26 in the example), but not in between them:
27 borg create -s --progress repo::archive path # good and preferred
29 borg create repo::archive path -s --progress # also works
30 borg create -s repo::archive path --progress # works, but ugly
31 borg create repo::archive -s --progress path # BAD
32 This is due to a problem in the argparse module:
34 http://bugs.python.org/issue15112
35 Repository URLs
37 Local filesystem (or locally mounted network filesystem):
38 /path/to/repo - filesystem path to repo directory, absolute path
40 path/to/repo - filesystem path to repo directory, relative path
42 Also, stuff like ~/path/to/repo or ~other/path/to/repo works (this is
44 expanded by your shell).
45 Note: you may also prepend a file:// to a filesystem path to get URL
47 style.
48 Remote repositories accessed via ssh user@host:
50 user@host:/path/to/repo - remote repo, absolute path
52 ssh://user@host:port/path/to/repo - same, alternative syntax, port can
54 be given
55 Remote repositories with relative paths can be given using this syntax:
57 user@host:path/to/repo - path relative to current directory
59 user@host:~/path/to/repo - path relative to user's home directory
61 user@host:~other/path/to/repo - path relative to other's home directory
63 Note: giving user@host:/./path/to/repo or user@host:/~/path/to/repo or
65 user@host:/~other/path/to/repo is also supported, but not required
66 here.
67 Remote repositories with relative paths, alternative syntax with port:
69 ssh://user@host:port/./path/to/repo - path relative to current direc‐
71 tory
72 ssh://user@host:port/~/path/to/repo - path relative to user's home
74 directory
75 ssh://user@host:port/~other/path/to/repo - path relative to other's
77 home directory
78 If you frequently need the same repo URL, it is a good idea to set the
80 BORG_REPO environment variable to set a default for the repo URL:
81 export BORG_REPO='ssh://user@host:port/path/to/repo'
83 Then just leave away the repo URL if only a repo URL is needed and you
85 want to use the default - it will be read from BORG_REPO then.
86 Use :: syntax to give the repo URL when syntax requires giving a posi‐
88 tional argument for the repo (e.g. borg mount :: /mnt).
89 Repository / Archive Locations
91 Many commands want either a repository (just give the repo URL, see
92 above) or an archive location, which is a repo URL followed by ::ar‐
93 chive_name.
94 Archive names must not contain the / (slash) character. For simplicity,
96 maybe also avoid blanks or other characters that have special meaning
97 on the shell or in a filesystem (borg mount will use the archive name
98 as directory name).
99 If you have set BORG_REPO (see above) and an archive location is
101 needed, use ::archive_name - the repo URL part is then read from
102 BORG_REPO.
103 Logging
105 Borg writes all log output to stderr by default. But please note that
106 something showing up on stderr does not indicate an error condition
107 just because it is on stderr. Please check the log levels of the mes‐
108 sages and the return code of borg for determining error, warning or
109 success conditions.
110 If you want to capture the log output to a file, just redirect it:
112 borg create repo::archive myfiles 2>> logfile
114 Custom logging configurations can be implemented via BORG_LOGGING_CONF.
116 The log level of the builtin logging configuration defaults to WARNING.
118 This is because we want Borg to be mostly silent and only output warn‐
119 ings, errors and critical messages, unless output has been requested by
120 supplying an option that implies output (e.g. --list or --progress).
121 Log levels: DEBUG < INFO < WARNING < ERROR < CRITICAL
123 Use --debug to set DEBUG log level - to get debug, info, warning, error
125 and critical level output.
126 Use --info (or -v or --verbose) to set INFO log level - to get info,
128 warning, error and critical level output.
129 Use --warning (default) to set WARNING log level - to get warning,
131 error and critical level output.
132 Use --error to set ERROR log level - to get error and critical level
134 output.
135 Use --critical to set CRITICAL log level - to get critical level out‐
137 put.
138 While you can set misc. log levels, do not expect that every command
140 will give different output on different log levels - it's just a possi‐
141 bility.
142 WARNING:
144 Options --critical and --error are provided for completeness, their
145 usage is not recommended as you might miss important information.
146 Return codes
148 Borg can exit with the following return codes (rc):
149 ┌────────────┬────────────────────────────┐
151 │Return code │ Meaning │
152 ├────────────┼────────────────────────────┤
153 │0 │ success (logged as INFO) │
154 ├────────────┼────────────────────────────┤
155 │1 │ warning (operation reached │
156 │ │ its normal end, but there │
157 │ │ were warnings -- you │
158 │ │ should check the log, │
159 │ │ logged as WARNING) │
160 ├────────────┼────────────────────────────┤
161 │2 │ error (like a fatal error, │
162 │ │ a local or remote excep‐ │
163 │ │ tion, the operation did │
164 │ │ not reach its normal end, │
165 │ │ logged as ERROR) │
166 ├────────────┼────────────────────────────┤
167 │128+N │ killed by signal N (e.g. │
168 │ │ 137 == kill -9) │
169 └────────────┴────────────────────────────┘
170 If you use --show-rc, the return code is also logged at the indicated
172 level as the last log entry.
173 Environment Variables
175 Borg uses some environment variables for automation:
176 General:
178 BORG_REPO
180 When set, use the value to give the default repository
181 location. If a command needs an archive parameter, you
182 can abbreviate as ::archive. If a command needs a reposi‐
183 tory parameter, you can either leave it away or abbrevi‐
184 ate as ::, if a positional parameter is required.
185 BORG_PASSPHRASE
187 When set, use the value to answer the passphrase question
188 for encrypted repositories. It is used when a passphrase
189 is needed to access an encrypted repo as well as when a
190 new passphrase should be initially set when initializing
191 an encrypted repo. See also BORG_NEW_PASSPHRASE.
192 BORG_PASSCOMMAND
194 When set, use the standard output of the command (trail‐
195 ing newlines are stripped) to answer the passphrase ques‐
196 tion for encrypted repositories. It is used when a
197 passphrase is needed to access an encrypted repo as well
198 as when a new passphrase should be initially set when
199 initializing an encrypted repo. Note that the command is
200 executed without a shell. So variables, like $HOME will
201 work, but ~ won't. If BORG_PASSPHRASE is also set, it
202 takes precedence. See also BORG_NEW_PASSPHRASE.
203 BORG_PASSPHRASE_FD
205 When set, specifies a file descriptor to read a
206 passphrase from. Programs starting borg may choose to
207 open an anonymous pipe and use it to pass a passphrase.
208 This is safer than passing via BORG_PASSPHRASE, because
209 on some systems (e.g. Linux) environment can be examined
210 by other processes. If BORG_PASSPHRASE or BORG_PASSCOM‐
211 MAND are also set, they take precedence.
212 BORG_NEW_PASSPHRASE
214 When set, use the value to answer the passphrase question
215 when a new passphrase is asked for. This variable is
216 checked first. If it is not set, BORG_PASSPHRASE and
217 BORG_PASSCOMMAND will also be checked. Main usecase for
218 this is to fully automate borg change-passphrase.
219 BORG_DISPLAY_PASSPHRASE
221 When set, use the value to answer the "display the
222 passphrase for verification" question when defining a new
223 passphrase for encrypted repositories.
224 BORG_HOSTNAME_IS_UNIQUE=no
226 Borg assumes that it can derive a unique hostname / iden‐
227 tity (see borg debug info). If this is not the case or
228 you do not want Borg to automatically remove stale locks,
229 set this to no.
230 BORG_HOST_ID
232 Borg usually computes a host id from the FQDN plus the
233 results of uuid.getnode() (which usually returns a unique
234 id based on the MAC address of the network interface.
235 Except if that MAC happens to be all-zero - in that case
236 it returns a random value, which is not what we want
237 (because it kills automatic stale lock removal). So, if
238 you have a all-zero MAC address or other reasons to bet‐
239 ter externally control the host id, just set this envi‐
240 ronment variable to a unique value. If all your FQDNs are
241 unique, you can just use the FQDN. If not, use
242 fqdn@uniqueid.
243 BORG_LOGGING_CONF
245 When set, use the given filename as INI-style logging
246 configuration. A basic example conf can be found at
247 docs/misc/logging.conf.
248 BORG_RSH
250 When set, use this command instead of ssh. This can be
251 used to specify ssh options, such as a custom identity
252 file ssh -i /path/to/private/key. See man ssh for other
253 options. Using the --rsh CMD commandline option overrides
254 the environment variable.
255 BORG_REMOTE_PATH
257 When set, use the given path as borg executable on the
258 remote (defaults to "borg" if unset). Using
259 --remote-path PATH commandline option overrides the envi‐
260 ronment variable.
261 BORG_FILES_CACHE_TTL
263 When set to a numeric value, this determines the maximum
264 "time to live" for the files cache entries (default: 20).
265 The files cache is used to quickly determine whether a
266 file is unchanged. The FAQ explains this more detailed
267 in: always_chunking
268 BORG_SHOW_SYSINFO
270 When set to no (default: yes), system information (like
271 OS, Python version, ...) in exceptions is not shown.
272 Please only use for good reasons as it makes issues
273 harder to analyze.
274 TMPDIR where temporary files are stored (might need a lot of
276 temporary space for some operations), see tempfile for
277 details
278 Some automatic answerers (if set, they automatically answer confirma‐
280 tion questions):
281 BORG_UNKNOWN_UNENCRYPTED_REPO_ACCESS_IS_OK=no (or =yes)
283 For "Warning: Attempting to access a previously unknown
284 unencrypted repository"
285 BORG_RELOCATED_REPO_ACCESS_IS_OK=no (or =yes)
287 For "Warning: The repository at location ... was previ‐
288 ously located at ..."
289 BORG_CHECK_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
291 For "Warning: 'check --repair' is an experimental feature
292 that might result in data loss."
293 BORG_DELETE_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
295 For "You requested to completely DELETE the repository
296 including all archives it contains:"
297 BORG_RECREATE_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
299 For "recreate is an experimental feature."
300 Note: answers are case sensitive. setting an invalid answer
302 value might either give the default answer or ask you interac‐
303 tively, depending on whether retries are allowed (they by
304 default are allowed). So please test your scripts interactively
305 before making them a non-interactive script.
306 Directories and files:
308 BORG_BASE_DIR
310 Default to '$HOME', '~$USER', '~' (in that order)'. If
311 we refer to ~ below, we in fact mean BORG_BASE_DIR.
312 BORG_CONFIG_DIR
314 Default to '~/.config/borg'. This directory contains the
315 whole config directories.
316 BORG_CACHE_DIR
318 Default to '~/.cache/borg'. This directory contains the
319 local cache and might need a lot of space for dealing
320 with big repositories.
321 BORG_SECURITY_DIR
323 Default to '~/.config/borg/security'. This directory con‐
324 tains information borg uses to track its usage of NONCES
325 ("numbers used once" - usually in encryption context) and
326 other security relevant data.
327 BORG_KEYS_DIR
329 Default to '~/.config/borg/keys'. This directory contains
330 keys for encrypted repositories.
331 BORG_KEY_FILE
333 When set, use the given filename as repository key file.
334 Building:
336 BORG_OPENSSL_PREFIX
338 Adds given OpenSSL header file directory to the default
339 locations (setup.py).
340 BORG_LIBLZ4_PREFIX
342 Adds given prefix directory to the default locations. If
343 a 'include/lz4.h' is found Borg will be linked against
344 the system liblz4 instead of a bundled implementation.
345 (setup.py)
346 BORG_LIBB2_PREFIX
348 Adds given prefix directory to the default locations. If
349 a 'include/blake2.h' is found Borg will be linked against
350 the system libb2 instead of a bundled implementation.
351 (setup.py)
352 BORG_LIBZSTD_PREFIX
354 Adds given prefix directory to the default locations. If
355 a 'include/zstd.h' is found Borg will be linked against
356 the system libzstd instead of a bundled implementation.
357 (setup.py)
358 Please note:
360 · be very careful when using the "yes" sayers, the warnings with prompt
362 exist for your / your data's security/safety
363 · also be very careful when putting your passphrase into a script, make
365 sure it has appropriate file permissions (e.g. mode 600, root:root).
366 File systems
368 We strongly recommend against using Borg (or any other database-like
369 software) on non-journaling file systems like FAT, since it is not pos‐
370 sible to assume any consistency in case of power failures (or a sudden
371 disconnect of an external drive or similar failures).
372 While Borg uses a data store that is resilient against these failures
374 when used on journaling file systems, it is not possible to guarantee
375 this with some hardware -- independent of the software used. We don't
376 know a list of affected hardware.
377 If you are suspicious whether your Borg repository is still consistent
379 and readable after one of the failures mentioned above occurred, run
380 borg check --verify-data to make sure it is consistent.
381 Requirements for Borg repository file systems
383 · Long file names
385 · At least three directory levels with short names
387 · Typically, file sizes up to a few hundred MB. Large repositories may
389 require large files (>2 GB).
390 · Up to 1000 files per directory (10000 for repositories initialized
392 with Borg 1.0)
393 · mkdir(2) should be atomic, since it is used for locking
395 · Hardlinks are needed for borg_upgrade --inplace
397 Units
399 To display quantities, Borg takes care of respecting the usual conven‐
400 tions of scale. Disk sizes are displayed in decimal, using powers of
401 ten (so kB means 1000 bytes). For memory usage, binary prefixes are
402 used, and are indicated using the IEC binary prefixes, using powers of
403 two (so KiB means 1024 bytes).
404 Date and Time
406 We format date and time conforming to ISO-8601, that is: YYYY-MM-DD and
407 HH:MM:SS (24h clock).
408 For more information about that, see: https://xkcd.com/1179/
410 Unless otherwise noted, we display local date and time. Internally, we
412 store and process date and time as UTC.
413 Resource Usage
415 Borg might use a lot of resources depending on the size of the data set
416 it is dealing with.
417 If one uses Borg in a client/server way (with a ssh: repository), the
419 resource usage occurs in part on the client and in another part on the
420 server.
421 If one uses Borg as a single process (with a filesystem repo), all the
423 resource usage occurs in that one process, so just add up client +
424 server to get the approximate resource usage.
425 CPU client:
427 · borg create: does chunking, hashing, compression, crypto (high
429 CPU usage)
430 · chunks cache sync: quite heavy on CPU, doing lots of hashtable
432 operations.
433 · borg extract: crypto, decompression (medium to high CPU usage)
435 · borg check: similar to extract, but depends on options given.
437 · borg prune / borg delete archive: low to medium CPU usage
439 · borg delete repo: done on the server
441 It won't go beyond 100% of 1 core as the code is currently sin‐
443 gle-threaded. Especially higher zlib and lzma compression lev‐
444 els use significant amounts of CPU cycles. Crypto might be cheap
445 on the CPU (if hardware accelerated) or expensive (if not).
446 CPU server:
448 It usually doesn't need much CPU, it just deals with the
449 key/value store (repository) and uses the repository index for
450 that.
451 borg check: the repository check computes the checksums of all
453 chunks (medium CPU usage) borg delete repo: low CPU usage
454 CPU (only for client/server operation):
456 When using borg in a client/server way with a ssh:-type repo,
457 the ssh processes used for the transport layer will need some
458 CPU on the client and on the server due to the crypto they are
459 doing - esp. if you are pumping big amounts of data.
460 Memory (RAM) client:
462 The chunks index and the files index are read into memory for
463 performance reasons. Might need big amounts of memory (see
464 below). Compression, esp. lzma compression with high levels
465 might need substantial amounts of memory.
466 Memory (RAM) server:
468 The server process will load the repository index into memory.
469 Might need considerable amounts of memory, but less than on the
470 client (see below).
471 Chunks index (client only):
473 Proportional to the amount of data chunks in your repo. Lots of
474 chunks in your repo imply a big chunks index. It is possible to
475 tweak the chunker params (see create options).
476 Files index (client only):
478 Proportional to the amount of files in your last backups. Can be
479 switched off (see create options), but next backup might be much
480 slower if you do. The speed benefit of using the files cache is
481 proportional to file size.
482 Repository index (server only):
484 Proportional to the amount of data chunks in your repo. Lots of
485 chunks in your repo imply a big repository index. It is possi‐
486 ble to tweak the chunker params (see create options) to influ‐
487 ence the amount of chunks being created.
488 Temporary files (client):
490 Reading data and metadata from a FUSE mounted repository will
491 consume up to the size of all deduplicated, small chunks in the
492 repository. Big chunks won't be locally cached.
493 Temporary files (server):
495 None.
496 Cache files (client only):
498 Contains the chunks index and files index (plus a collection of
499 single- archive chunk indexes which might need huge amounts of
500 disk space, depending on archive count and size - see FAQ about
501 how to reduce).
502 Network (only for client/server operation):
504 If your repository is remote, all deduplicated (and optionally
505 compressed/ encrypted) data of course has to go over the connec‐
506 tion (ssh:// repo url). If you use a locally mounted network
507 filesystem, additionally some copy operations used for transac‐
508 tion support also go over the connection. If you backup multiple
509 sources to one target repository, additional traffic happens for
510 cache resynchronization.
511 Support for file metadata
513 Besides regular file and directory structures, Borg can preserve
514 · symlinks (stored as symlink, the symlink is not followed)
516 · special files:
518 · character and block device files (restored via mknod)
520 · FIFOs ("named pipes")
522 · special file contents can be backed up in --read-special mode. By
524 default the metadata to create them with mknod(2), mkfifo(2) etc.
525 is stored.
526 · hardlinked regular files, devices, FIFOs (considering all items in
528 the same archive)
529 · timestamps in nanosecond precision: mtime, atime, ctime
531 · other timestamps: birthtime (on platforms supporting it)
533 · permissions:
535 · IDs of owning user and owning group
537 · names of owning user and owning group (if the IDs can be resolved)
539 · Unix Mode/Permissions (u/g/o permissions, suid, sgid, sticky)
541 On some platforms additional features are supported:
543 ┌─────────────────┬──────────┬───────────┬───────────┐
545 │Platform │ ACLs [5] │ xattr [6] │ Flags [7] │
546 ├─────────────────┼──────────┼───────────┼───────────┤
547 │Linux │ Yes │ Yes │ Yes [1] │
548 ├─────────────────┼──────────┼───────────┼───────────┤
549 │Mac OS X │ Yes │ Yes │ Yes (all) │
550 ├─────────────────┼──────────┼───────────┼───────────┤
551 │FreeBSD │ Yes │ Yes │ Yes (all) │
552 ├─────────────────┼──────────┼───────────┼───────────┤
553 │OpenBSD │ n/a │ n/a │ Yes (all) │
554 ├─────────────────┼──────────┼───────────┼───────────┤
555 │NetBSD │ n/a │ No [2] │ Yes (all) │
556 ├─────────────────┼──────────┼───────────┼───────────┤
557 │Solaris and de‐ │ No [3] │ No [3] │ n/a │
558 │rivatives │ │ │ │
559 ├─────────────────┼──────────┼───────────┼───────────┤
560 │Windows (cygwin) │ No [4] │ No │ No │
561 └─────────────────┴──────────┴───────────┴───────────┘
562 Other Unix-like operating systems may work as well, but have not been
564 tested at all.
565 Note that most of the platform-dependent features also depend on the
567 file system. For example, ntfs-3g on Linux isn't able to convey NTFS
568 ACLs.
569 [1] Only "nodump", "immutable", "compressed" and "append" are sup‐
571 ported. Feature request #618 for more flags.
572 [2] Feature request #1332
574 [3] Feature request #1337
576 [4] Cygwin tries to map NTFS ACLs to permissions with varying degress
578 of success.
579 [5] The native access control list mechanism of the OS. This normally
581 limits access to non-native ACLs. For example, NTFS ACLs aren't
582 completely accessible on Linux with ntfs-3g.
583 [6] extended attributes; key-value pairs attached to a file, mainly
585 used by the OS. This includes resource forks on Mac OS X.
586 [7] aka BSD flags. The Linux set of flags [1] is portable across plat‐
588 forms. The BSDs define additional flags.
589 In case you are interested in more details (like formulas), please
591 see internals. For details on the available JSON output, refer to
592 json_output.
593 Common options
595 All Borg commands share these options:
596 -h, --help
598 show this help message and exit
599 --critical
601 work on log level CRITICAL
602 --error
604 work on log level ERROR
605 --warning
607 work on log level WARNING (default)
608 --info, -v, --verbose
610 work on log level INFO
611 --debug
613 enable debug output, work on log level DEBUG
614 --debug-topic TOPIC
616 enable TOPIC debugging (can be specified multiple times). The
617 logger path is borg.debug.<TOPIC> if TOPIC is not fully quali‐
618 fied.
619 -p, --progress
621 show progress information
622 --log-json
624 Output one JSON object per log line instead of formatted text.
625 --lock-wait SECONDS
627 wait at most SECONDS for acquiring a repository/cache lock
628 (default: 1).
629 --show-version
631 show/log the borg version
632 --show-rc
634 show/log the return code (rc)
635 --umask M
637 set umask to M (local and remote, default: 0077)
638 --remote-path PATH
640 use PATH as borg executable on the remote (default: "borg")
641 --remote-ratelimit RATE
643 set remote network upload rate limit in kiByte/s (default:
644 0=unlimited)
645 --consider-part-files
647 treat part files like normal files (e.g. to list/extract them)
648 --debug-profile FILE
650 Write execution profile in Borg format into FILE. For local use
651 a Python-compatible file can be generated by suffixing FILE with
652 ".pyprof".
653 --rsh RSH
655 Use this command to connect to the 'borg serve' process
656 (default: 'ssh')
657 Examples
659 # Create an archive and log: borg version, files list, return code
660 $ borg create --show-version --list --show-rc /path/to/repo::my-files files
661
663 borg [common options] init [options] [REPOSITORY]
664 Description
666 This command initializes an empty repository. A repository is a
667 filesystem directory containing the deduplicated data from zero or more
668 archives.
669 Encryption can be enabled at repository init time. It cannot be changed
671 later.
672 It is not recommended to work without encryption. Repository encryption
674 protects you e.g. against the case that an attacker has access to your
675 backup repository.
676 But be careful with the key / the passphrase:
678 If you want "passphrase-only" security, use one of the repokey modes.
680 The key will be stored inside the repository (in its "config" file). In
681 above mentioned attack scenario, the attacker will have the key (but
682 not the passphrase).
683 If you want "passphrase and having-the-key" security, use one of the
685 keyfile modes. The key will be stored in your home directory (in .con‐
686 fig/borg/keys). In the attack scenario, the attacker who has just
687 access to your repo won't have the key (and also not the passphrase).
688 Make a backup copy of the key file (keyfile mode) or repo config file
690 (repokey mode) and keep it at a safe place, so you still have the key
691 in case it gets corrupted or lost. Also keep the passphrase at a safe
692 place. The backup that is encrypted with that key won't help you with
693 that, of course.
694 Make sure you use a good passphrase. Not too short, not too simple. The
696 real encryption / decryption key is encrypted with / locked by your
697 passphrase. If an attacker gets your key, he can't unlock and use it
698 without knowing the passphrase.
699 Be careful with special or non-ascii characters in your passphrase:
701 · Borg processes the passphrase as unicode (and encodes it as utf-8),
703 so it does not have problems dealing with even the strangest charac‐
704 ters.
705 · BUT: that does not necessarily apply to your OS / VM / keyboard con‐
707 figuration.
708 So better use a long passphrase made from simple ascii chars than one
710 that includes non-ascii stuff or characters that are hard/impossible to
711 enter on a different keyboard layout.
712 You can change your passphrase for existing repos at any time, it won't
714 affect the encryption/decryption key or other secrets.
715 Encryption modes
717 ┌─────────┬──────────────────┬──────────────────┬──────────────────┐
718 │Hash/MAC │ Not encrypted no │ Not encrypted, │ Encrypted (AEAD │
719 │ │ auth │ but authenti‐ │ w/ AES) and │
720 │ │ │ cated │ authenticated │
721 ├─────────┼──────────────────┼──────────────────┼──────────────────┤
722 │SHA-256 │ none │ authenticated │ repokey keyfile │
723 ├─────────┼──────────────────┼──────────────────┼──────────────────┤
724 │BLAKE2b │ n/a │ authenti‐ │ repokey-blake2 │
725 │ │ │ cated-blake2 │ keyfile-blake2 │
726 └─────────┴──────────────────┴──────────────────┴──────────────────┘
727 Marked modes are new in Borg 1.1 and are not backwards-compatible with
729 Borg 1.0.x.
730 On modern Intel/AMD CPUs (except very cheap ones), AES is usually hard‐
732 ware-accelerated. BLAKE2b is faster than SHA256 on Intel/AMD 64-bit
733 CPUs (except AMD Ryzen and future CPUs with SHA extensions), which
734 makes authenticated-blake2 faster than none and authenticated.
735 On modern ARM CPUs, NEON provides hardware acceleration for SHA256 mak‐
737 ing it faster than BLAKE2b-256 there. NEON accelerates AES as well.
738 Hardware acceleration is always used automatically when available.
740 repokey and keyfile use AES-CTR-256 for encryption and HMAC-SHA256 for
742 authentication in an encrypt-then-MAC (EtM) construction. The chunk ID
743 hash is HMAC-SHA256 as well (with a separate key). These modes are
744 compatible with Borg 1.0.x.
745 repokey-blake2 and keyfile-blake2 are also authenticated encryption
747 modes, but use BLAKE2b-256 instead of HMAC-SHA256 for authentication.
748 The chunk ID hash is a keyed BLAKE2b-256 hash. These modes are new and
749 not compatible with Borg 1.0.x.
750 authenticated mode uses no encryption, but authenticates repository
752 contents through the same HMAC-SHA256 hash as the repokey and keyfile
753 modes (it uses it as the chunk ID hash). The key is stored like
754 repokey. This mode is new and not compatible with Borg 1.0.x.
755 authenticated-blake2 is like authenticated, but uses the keyed
757 BLAKE2b-256 hash from the other blake2 modes. This mode is new and not
758 compatible with Borg 1.0.x.
759 none mode uses no encryption and no authentication. It uses SHA256 as
761 chunk ID hash. Not recommended, rather consider using an authenticated
762 or authenticated/encrypted mode. This mode has possible denial-of-ser‐
763 vice issues when running borg create on contents controlled by an
764 attacker. Use it only for new repositories where no encryption is
765 wanted and when compatibility with 1.0.x is important. If compatibility
766 with 1.0.x is not important, use authenticated-blake2 or authenticated
767 instead. This mode is compatible with Borg 1.0.x.
768 Examples
770 # Local repository, repokey encryption, BLAKE2b (often faster, since Borg 1.1)
771 $ borg init --encryption=repokey-blake2 /path/to/repo
772 # Local repository (no encryption)
774 $ borg init --encryption=none /path/to/repo
775 # Remote repository (accesses a remote borg via ssh)
777 $ borg init --encryption=repokey-blake2 user@hostname:backup
778 # Remote repository (store the key your home dir)
780 $ borg init --encryption=keyfile user@hostname:backup
781
783 borg [common options] create [options] ARCHIVE [PATH...]
784 Description
786 This command creates a backup archive containing all files found while
787 recursively traversing all paths specified. Paths are added to the ar‐
788 chive as they are given, that means if relative paths are desired, the
789 command has to be run from the correct directory.
790 When giving '-' as path, borg will read data from standard input and
792 create a file 'stdin' in the created archive from that data.
793 The archive will consume almost no disk space for files or parts of
795 files that have already been stored in other archives.
796 The archive name needs to be unique. It must not end in '.checkpoint'
798 or '.checkpoint.N' (with N being a number), because these names are
799 used for checkpoints and treated in special ways.
800 In the archive name, you may use the following placeholders: {now},
802 {utcnow}, {fqdn}, {hostname}, {user} and some others.
803 Backup speed is increased by not reprocessing files that are already
805 part of existing archives and weren't modified. The detection of unmod‐
806 ified files is done by comparing multiple file metadata values with
807 previous values kept in the files cache.
808 This comparison can operate in different modes as given by
810 --files-cache:
811 · ctime,size,inode (default)
813 · mtime,size,inode (default behaviour of borg versions older than
815 1.1.0rc4)
816 · ctime,size (ignore the inode number)
818 · mtime,size (ignore the inode number)
820 · rechunk,ctime (all files are considered modified - rechunk, cache
822 ctime)
823 · rechunk,mtime (all files are considered modified - rechunk, cache
825 mtime)
826 · disabled (disable the files cache, all files considered modified -
828 rechunk)
829 inode number: better safety, but often unstable on network filesystems
831 Normally, detecting file modifications will take inode information into
833 consideration to improve the reliability of file change detection.
834 This is problematic for files located on sshfs and similar network file
835 systems which do not provide stable inode numbers, such files will
836 always be considered modified. You can use modes without inode in this
837 case to improve performance, but reliability of change detection might
838 be reduced.
839 ctime vs. mtime: safety vs. speed
841 · ctime is a rather safe way to detect changes to a file (metadata and
843 contents) as it can not be set from userspace. But, a metadata-only
844 change will already update the ctime, so there might be some unneces‐
845 sary chunking/hashing even without content changes. Some filesystems
846 do not support ctime (change time).
847 · mtime usually works and only updates if file contents were changed.
849 But mtime can be arbitrarily set from userspace, e.g. to set mtime
850 back to the same value it had before a content change happened. This
851 can be used maliciously as well as well-meant, but in both cases
852 mtime based cache modes can be problematic.
853 The mount points of filesystems or filesystem snapshots should be the
855 same for every creation of a new archive to ensure fast operation. This
856 is because the file cache that is used to determine changed files
857 quickly uses absolute filenames. If this is not possible, consider
858 creating a bind mount to a stable location.
859 The --progress option shows (from left to right) Original, Compressed
861 and Deduplicated (O, C and D, respectively), then the Number of files
862 (N) processed so far, followed by the currently processed path.
863 When using --stats, you will get some statistics about how much data
865 was added - the "This Archive" deduplicated size there is most inter‐
866 esting as that is how much your repository will grow. Please note that
867 the "All archives" stats refer to the state after creation. Also, the
868 --stats and --dry-run options are mutually exclusive because the data
869 is not actually compressed and deduplicated during a dry run.
870 See the output of the "borg help patterns" command for more help on
872 exclude patterns. See the output of the "borg help placeholders" com‐
873 mand for more help on placeholders.
874 The --exclude patterns are not like tar. In tar --exclude .bundler/gems
876 will exclude foo/.bundler/gems. In borg it will not, you need to use
877 --exclude '*/.bundler/gems' to get the same effect. See borg help pat‐
878 terns for more information.
879 In addition to using --exclude patterns, it is possible to use
881 --exclude-if-present to specify the name of a filesystem object (e.g. a
882 file or folder name) which, when contained within another folder, will
883 prevent the containing folder from being backed up. By default, the
884 containing folder and all of its contents will be omitted from the
885 backup. If, however, you wish to only include the objects specified by
886 --exclude-if-present in your backup, and not include any other contents
887 of the containing folder, this can be enabled through using the
888 --keep-exclude-tags option.
889 Item flags
891 --list outputs a list of all files, directories and other file system
892 items it considered (no matter whether they had content changes or
893 not). For each item, it prefixes a single-letter flag that indicates
894 type and/or status of the item.
895 If you are interested only in a subset of that output, you can give
897 e.g. --filter=AME and it will only show regular files with A, M or E
898 status (see below).
899 A uppercase character represents the status of a regular file relative
901 to the "files" cache (not relative to the repo -- this is an issue if
902 the files cache is not used). Metadata is stored in any case and for
903 'A' and 'M' also new data chunks are stored. For 'U' all data chunks
904 refer to already existing chunks.
905 · 'A' = regular file, added (see also a_status_oddity in the FAQ)
907 · 'M' = regular file, modified
909 · 'U' = regular file, unchanged
911 · 'E' = regular file, an error happened while accessing/reading this
913 file
914 A lowercase character means a file type other than a regular file, borg
916 usually just stores their metadata:
917 · 'd' = directory
919 · 'b' = block device
921 · 'c' = char device
923 · 'h' = regular file, hardlink (to already seen inodes)
925 · 's' = symlink
927 · 'f' = fifo
929 Other flags used include:
931 · 'i' = backup data was read from standard input (stdin)
933 · '-' = dry run, item was not backed up
935 · 'x' = excluded, item was not backed up
937 · '?' = missing status code (if you see this, please file a bug
939 report!)
940 Examples
942 # Backup ~/Documents into an archive named "my-documents"
943 $ borg create /path/to/repo::my-documents ~/Documents
944 # same, but list all files as we process them
946 $ borg create --list /path/to/repo::my-documents ~/Documents
947 # Backup ~/Documents and ~/src but exclude pyc files
949 $ borg create /path/to/repo::my-files \
950 ~/Documents \
951 ~/src \
952 --exclude '*.pyc'
953 # Backup home directories excluding image thumbnails (i.e. only
955 # /home/<one directory>/.thumbnails is excluded, not /home/*/*/.thumbnails etc.)
956 $ borg create /path/to/repo::my-files /home \
957 --exclude 'sh:/home/*/.thumbnails'
958 # Backup the root filesystem into an archive named "root-YYYY-MM-DD"
960 # use zlib compression (good, but slow) - default is lz4 (fast, low compression ratio)
961 $ borg create -C zlib,6 --one-file-system /path/to/repo::root-{now:%Y-%m-%d} /
962 # Backup a remote host locally ("pull" style) using sshfs
964 $ mkdir sshfs-mount
965 $ sshfs root@example.com:/ sshfs-mount
966 $ cd sshfs-mount
967 $ borg create /path/to/repo::example.com-root-{now:%Y-%m-%d} .
968 $ cd ..
969 $ fusermount -u sshfs-mount
970 # Make a big effort in fine granular deduplication (big chunk management
972 # overhead, needs a lot of RAM and disk space, see formula in internals
973 # docs - same parameters as borg < 1.0 or attic):
974 $ borg create --chunker-params 10,23,16,4095 /path/to/repo::small /smallstuff
975 # Backup a raw device (must not be active/in use/mounted at that time)
977 $ dd if=/dev/sdx bs=10M | borg create /path/to/repo::my-sdx -
978 # No compression (none)
980 $ borg create --compression none /path/to/repo::arch ~
981 # Super fast, low compression (lz4, default)
983 $ borg create /path/to/repo::arch ~
984 # Less fast, higher compression (zlib, N = 0..9)
986 $ borg create --compression zlib,N /path/to/repo::arch ~
987 # Even slower, even higher compression (lzma, N = 0..9)
989 $ borg create --compression lzma,N /path/to/repo::arch ~
990 # Only compress compressible data with lzma,N (N = 0..9)
992 $ borg create --compression auto,lzma,N /path/to/repo::arch ~
993 # Use short hostname, user name and current time in archive name
995 $ borg create /path/to/repo::{hostname}-{user}-{now} ~
996 # Similar, use the same datetime format as borg 1.1 will have as default
997 $ borg create /path/to/repo::{hostname}-{user}-{now:%Y-%m-%dT%H:%M:%S} ~
998 # As above, but add nanoseconds
999 $ borg create /path/to/repo::{hostname}-{user}-{now:%Y-%m-%dT%H:%M:%S.%f} ~
1000 # Backing up relative paths by moving into the correct directory first
1002 $ cd /home/user/Documents
1003 # The root directory of the archive will be "projectA"
1004 $ borg create /path/to/repo::daily-projectA-{now:%Y-%m-%d} projectA
1005
1007 borg [common options] extract [options] ARCHIVE [PATH...]
1008 Description
1010 This command extracts the contents of an archive. By default the entire
1011 archive is extracted but a subset of files and directories can be
1012 selected by passing a list of PATHs as arguments. The file selection
1013 can further be restricted by using the --exclude option.
1014 See the output of the "borg help patterns" command for more help on
1016 exclude patterns.
1017 By using --dry-run, you can do all extraction steps except actually
1019 writing the output data: reading metadata and data chunks from the
1020 repo, checking the hash/hmac, decrypting, decompressing.
1021 --progress can be slower than no progress display, since it makes one
1023 additional pass over the archive metadata.
1024 NOTE:
1026 Currently, extract always writes into the current working directory
1027 ("."), so make sure you cd to the right place before calling borg
1028 extract.
1029 Examples
1031 # Extract entire archive
1032 $ borg extract /path/to/repo::my-files
1033 # Extract entire archive and list files while processing
1035 $ borg extract --list /path/to/repo::my-files
1036 # Verify whether an archive could be successfully extracted, but do not write files to disk
1038 $ borg extract --dry-run /path/to/repo::my-files
1039 # Extract the "src" directory
1041 $ borg extract /path/to/repo::my-files home/USERNAME/src
1042 # Extract the "src" directory but exclude object files
1044 $ borg extract /path/to/repo::my-files home/USERNAME/src --exclude '*.o'
1045 # Restore a raw device (must not be active/in use/mounted at that time)
1047 $ borg extract --stdout /path/to/repo::my-sdx | dd of=/dev/sdx bs=10M
1048
1050 borg [common options] check [options] [REPOSITORY_OR_ARCHIVE]
1051 Description
1053 The check command verifies the consistency of a repository and the cor‐
1054 responding archives.
1055 First, the underlying repository data files are checked:
1057 · For all segments the segment magic (header) is checked
1059 · For all objects stored in the segments, all metadata (e.g. crc and
1061 size) and all data is read. The read data is checked by size and CRC.
1062 Bit rot and other types of accidental damage can be detected this
1063 way.
1064 · If we are in repair mode and a integrity error is detected for a seg‐
1066 ment, we try to recover as many objects from the segment as possible.
1067 · In repair mode, it makes sure that the index is consistent with the
1069 data stored in the segments.
1070 · If you use a remote repo server via ssh:, the repo check is executed
1072 on the repo server without causing significant network traffic.
1073 · The repository check can be skipped using the --archives-only option.
1075 Second, the consistency and correctness of the archive metadata is ver‐
1077 ified:
1078 · Is the repo manifest present? If not, it is rebuilt from archive
1080 metadata chunks (this requires reading and decrypting of all metadata
1081 and data).
1082 · Check if archive metadata chunk is present. if not, remove archive
1084 from manifest.
1085 · For all files (items) in the archive, for all chunks referenced by
1087 these files, check if chunk is present. If a chunk is not present
1088 and we are in repair mode, replace it with a same-size replacement
1089 chunk of zeros. If a previously lost chunk reappears (e.g. via a
1090 later backup) and we are in repair mode, the all-zero replacement
1091 chunk will be replaced by the correct chunk. This requires reading
1092 of archive and file metadata, but not data.
1093 · If we are in repair mode and we checked all the archives: delete
1095 orphaned chunks from the repo.
1096 · if you use a remote repo server via ssh:, the archive check is exe‐
1098 cuted on the client machine (because if encryption is enabled, the
1099 checks will require decryption and this is always done client-side,
1100 because key access will be required).
1101 · The archive checks can be time consuming, they can be skipped using
1103 the --repository-only option.
1104 The --verify-data option will perform a full integrity verification (as
1106 opposed to checking the CRC32 of the segment) of data, which means
1107 reading the data from the repository, decrypting and decompressing it.
1108 This is a cryptographic verification, which will detect (accidental)
1109 corruption. For encrypted repositories it is tamper-resistant as well,
1110 unless the attacker has access to the keys.
1111 It is also very slow.
1113
1115 borg [common options] rename [options] ARCHIVE NEWNAME
1116 Description
1118 This command renames an archive in the repository.
1119 This results in a different archive ID.
1121 Examples
1123 $ borg create /path/to/repo::archivename ~
1124 $ borg list /path/to/repo
1125 archivename Mon, 2016-02-15 19:50:19
1126 $ borg rename /path/to/repo::archivename newname
1128 $ borg list /path/to/repo
1129 newname Mon, 2016-02-15 19:50:19
1130
1132 borg [common options] list [options] [REPOSITORY_OR_ARCHIVE] [PATH...]
1133 Description
1135 This command lists the contents of a repository or an archive.
1136 See the "borg help patterns" command for more help on exclude patterns.
1138 The following keys are available for --format:
1140 · NEWLINE: OS dependent line separator
1142 · NL: alias of NEWLINE
1144 · NUL: NUL character for creating print0 / xargs -0 like output, see
1146 barchive/bpath
1147 · SPACE
1149 · TAB
1151 · CR
1153 · LF
1155 Keys for listing repository archives:
1157 · archive: archive name interpreted as text (might be missing non-text
1159 characters, see barchive)
1160 · name: alias of "archive"
1162 · barchive: verbatim archive name, can contain any character except NUL
1164 · comment: archive comment interpreted as text (might be missing
1166 non-text characters, see bcomment)
1167 · bcomment: verbatim archive comment, can contain any character except
1169 NUL
1170 · id: internal ID of the archive
1172 · start: time (start) of creation of the archive
1174 · time: alias of "start"
1176 · end: time (end) of creation of the archive
1178 · hostname: hostname of host on which this archive was created
1180 · username: username of user who created this archive
1182 Keys for listing archive files:
1184 · type
1186 · mode
1188 · uid
1190 · gid
1192 · user
1194 · group
1196 · path: path interpreted as text (might be missing non-text characters,
1198 see bpath)
1199 · bpath: verbatim POSIX path, can contain any character except NUL
1201 · source: link target for links (identical to linktarget)
1203 · linktarget
1205 · flags
1207 · size
1209 · csize: compressed size
1211 · dsize: deduplicated size
1213 · dcsize: deduplicated compressed size
1215 · num_chunks: number of chunks in this file
1217 · unique_chunks: number of unique chunks in this file
1219 · mtime
1221 · ctime
1223 · atime
1225 · isomtime
1227 · isoctime
1229 · isoatime
1231 · md5
1233 · sha1
1235 · sha224
1237 · sha256
1239 · sha384
1241 · sha512
1243 · archiveid
1245 · archivename
1247 · extra: prepends {source} with " -> " for soft links and " link to "
1249 for hard links
1250 · health: either "healthy" (file ok) or "broken" (if file has all-zero
1252 replacement chunks)
1253 Examples
1255 $ borg list /path/to/repo
1256 Monday Mon, 2016-02-15 19:15:11
1257 repo Mon, 2016-02-15 19:26:54
1258 root-2016-02-15 Mon, 2016-02-15 19:36:29
1259 newname Mon, 2016-02-15 19:50:19
1260 ...
1261 $ borg list /path/to/repo::root-2016-02-15
1263 drwxr-xr-x root root 0 Mon, 2016-02-15 17:44:27 .
1264 drwxrwxr-x root root 0 Mon, 2016-02-15 19:04:49 bin
1265 -rwxr-xr-x root root 1029624 Thu, 2014-11-13 00:08:51 bin/bash
1266 lrwxrwxrwx root root 0 Fri, 2015-03-27 20:24:26 bin/bzcmp -> bzdiff
1267 -rwxr-xr-x root root 2140 Fri, 2015-03-27 20:24:22 bin/bzdiff
1268 ...
1269 $ borg list /path/to/repo::archiveA --format="{mode} {user:6} {group:6} {size:8d} {isomtime} {path}{extra}{NEWLINE}"
1271 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 .
1272 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 code
1273 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 code/myproject
1274 -rw-rw-r-- user user 1416192 Sun, 2015-02-01 11:00:00 code/myproject/file.ext
1275 ...
1276
1278 borg [common options] diff [options] REPO::ARCHIVE1 ARCHIVE2 [PATH...]
1279 Description
1281 This command finds differences (file contents, user/group/mode) between
1282 archives.
1283 A repository location and an archive name must be specified for
1285 REPO::ARCHIVE1. ARCHIVE2 is just another archive name in same reposi‐
1286 tory (no repository location allowed).
1287 For archives created with Borg 1.1 or newer diff automatically detects
1289 whether the archives are created with the same chunker params. If so,
1290 only chunk IDs are compared, which is very fast.
1291 For archives prior to Borg 1.1 chunk contents are compared by default.
1293 If you did not create the archives with different chunker params, pass
1294 --same-chunker-params. Note that the chunker params changed from Borg
1295 0.xx to 1.0.
1296 See the output of the "borg help patterns" command for more help on
1298 exclude patterns.
1299 Examples
1301 $ borg init -e=none testrepo
1302 $ mkdir testdir
1303 $ cd testdir
1304 $ echo asdf > file1
1305 $ dd if=/dev/urandom bs=1M count=4 > file2
1306 $ touch file3
1307 $ borg create ../testrepo::archive1 .
1308 $ chmod a+x file1
1310 $ echo "something" >> file2
1311 $ borg create ../testrepo::archive2 .
1312 $ rm file3
1314 $ touch file4
1315 $ borg create ../testrepo::archive3 .
1316 $ cd ..
1318 $ borg diff testrepo::archive1 archive2
1319 [-rw-r--r-- -> -rwxr-xr-x] file1
1320 +135 B -252 B file2
1321 $ borg diff testrepo::archive2 archive3
1323 added 0 B file4
1324 removed 0 B file3
1325 $ borg diff testrepo::archive1 archive3
1327 [-rw-r--r-- -> -rwxr-xr-x] file1
1328 +135 B -252 B file2
1329 added 0 B file4
1330 removed 0 B file3
1331
1333 borg [common options] delete [options] [TARGET] [ARCHIVE...]
1334 Description
1336 This command deletes an archive from the repository or the complete
1337 repository. Disk space is reclaimed accordingly. If you delete the
1338 complete repository, the local cache for it (if any) is also deleted.
1339 When using --stats, you will get some statistics about how much data
1341 was deleted - the "Deleted data" deduplicated size there is most inter‐
1342 esting as that is how much your repository will shrink. Please note
1343 that the "All archives" stats refer to the state after deletion.
1344 Examples
1346 # delete a single backup archive:
1347 $ borg delete /path/to/repo::Monday
1348 # delete the whole repository and the related local cache:
1350 $ borg delete /path/to/repo
1351 You requested to completely DELETE the repository *including* all archives it contains:
1352 repo Mon, 2016-02-15 19:26:54
1353 root-2016-02-15 Mon, 2016-02-15 19:36:29
1354 newname Mon, 2016-02-15 19:50:19
1355 Type 'YES' if you understand this and want to continue: YES
1356
1358 borg [common options] prune [options] [REPOSITORY]
1359 Description
1361 The prune command prunes a repository by deleting all archives not
1362 matching any of the specified retention options. This command is nor‐
1363 mally used by automated backup scripts wanting to keep a certain number
1364 of historic backups.
1365 Also, prune automatically removes checkpoint archives (incomplete ar‐
1367 chives left behind by interrupted backup runs) except if the checkpoint
1368 is the latest archive (and thus still needed). Checkpoint archives are
1369 not considered when comparing archive counts against the retention lim‐
1370 its (--keep-X).
1371 If a prefix is set with -P, then only archives that start with the pre‐
1373 fix are considered for deletion and only those archives count towards
1374 the totals specified by the rules. Otherwise, all archives in the
1375 repository are candidates for deletion! There is no automatic distinc‐
1376 tion between archives representing different contents. These need to be
1377 distinguished by specifying matching prefixes.
1378 If you have multiple sequences of archives with different data sets
1380 (e.g. from different machines) in one shared repository, use one prune
1381 call per data set that matches only the respective archives using the
1382 -P option.
1383 The --keep-within option takes an argument of the form "<int><char>",
1385 where char is "H", "d", "w", "m", "y". For example, --keep-within 2d
1386 means to keep all archives that were created within the past 48 hours.
1387 "1m" is taken to mean "31d". The archives kept with this option do not
1388 count towards the totals specified by any other options.
1389 A good procedure is to thin out more and more the older your backups
1391 get. As an example, --keep-daily 7 means to keep the latest backup on
1392 each day, up to 7 most recent days with backups (days without backups
1393 do not count). The rules are applied from secondly to yearly, and
1394 backups selected by previous rules do not count towards those of later
1395 rules. The time that each backup starts is used for pruning purposes.
1396 Dates and times are interpreted in the local timezone, and weeks go
1397 from Monday to Sunday. Specifying a negative number of archives to keep
1398 means that there is no limit.
1399 The --keep-last N option is doing the same as --keep-secondly N (and it
1401 will keep the last N archives under the assumption that you do not cre‐
1402 ate more than one backup archive in the same second).
1403 When using --stats, you will get some statistics about how much data
1405 was deleted - the "Deleted data" deduplicated size there is most inter‐
1406 esting as that is how much your repository will shrink. Please note
1407 that the "All archives" stats refer to the state after pruning.
1408 Examples
1410 Be careful, prune is a potentially dangerous command, it will remove
1411 backup archives.
1412 The default of prune is to apply to all archives in the repository
1414 unless you restrict its operation to a subset of the archives using
1415 --prefix. When using --prefix, be careful to choose a good prefix -
1416 e.g. do not use a prefix "foo" if you do not also want to match "foo‐
1417 bar".
1418 It is strongly recommended to always run prune -v --list --dry-run ...
1420 first so you will see what it would do without it actually doing any‐
1421 thing.
1422 # Keep 7 end of day and 4 additional end of week archives.
1424 # Do a dry-run without actually deleting anything.
1425 $ borg prune -v --list --dry-run --keep-daily=7 --keep-weekly=4 /path/to/repo
1426 # Same as above but only apply to archive names starting with the hostname
1428 # of the machine followed by a "-" character:
1429 $ borg prune -v --list --keep-daily=7 --keep-weekly=4 --prefix='{hostname}-' /path/to/repo
1430 # Keep 7 end of day, 4 additional end of week archives,
1432 # and an end of month archive for every month:
1433 $ borg prune -v --list --keep-daily=7 --keep-weekly=4 --keep-monthly=-1 /path/to/repo
1434 # Keep all backups in the last 10 days, 4 additional end of week archives,
1436 # and an end of month archive for every month:
1437 $ borg prune -v --list --keep-within=10d --keep-weekly=4 --keep-monthly=-1 /path/to/repo
1438 There is also a visualized prune example in docs/misc/prune-exam‐
1440 ple.txt:
1441 borg prune visualized
1443 =====================
1444 Assume it is 2016-01-01, today's backup has not yet been made and you have
1446 created at least one backup on each day in 2015 except on 2015-12-19 (no
1447 backup made on that day).
1448 This is what borg prune --keep-daily 14 --keep-monthly 6 would keep.
1450 Backups kept by the --keep-daily rule are marked by a "d" to the right,
1452 backups kept by the --keep-monthly rule are marked by a "m" to the right.
1453 Calendar view
1455 -------------
1456 2015
1458 January February March
1459 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1460 1 2 3 4 1 1
1461 5 6 7 8 9 10 11 2 3 4 5 6 7 8 2 3 4 5 6 7 8
1462 12 13 14 15 16 17 18 9 10 11 12 13 14 15 9 10 11 12 13 14 15
1463 19 20 21 22 23 24 25 16 17 18 19 20 21 22 16 17 18 19 20 21 22
1464 26 27 28 29 30 31 23 24 25 26 27 28 23 24 25 26 27 28 29
1465 30 31
1466 April May June
1468 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1469 1 2 3 4 5 1 2 3 1 2 3 4 5 6 7
1470 6 7 8 9 10 11 12 4 5 6 7 8 9 10 8 9 10 11 12 13 14
1471 13 14 15 16 17 18 19 11 12 13 14 15 16 17 15 16 17 18 19 20 21
1472 20 21 22 23 24 25 26 18 19 20 21 22 23 24 22 23 24 25 26 27 28
1473 27 28 29 30 25 26 27 28 29 30 31 29 30m
1474 July August September
1477 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1478 1 2 3 4 5 1 2 1 2 3 4 5 6
1479 6 7 8 9 10 11 12 3 4 5 6 7 8 9 7 8 9 10 11 12 13
1480 13 14 15 16 17 18 19 10 11 12 13 14 15 16 14 15 16 17 18 19 20
1481 20 21 22 23 24 25 26 17 18 19 20 21 22 23 21 22 23 24 25 26 27
1482 27 28 29 30 31m 24 25 26 27 28 29 30 28 29 30m
1483 31m
1484 October November December
1486 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1487 1 2 3 4 1 1 2 3 4 5 6
1488 5 6 7 8 9 10 11 2 3 4 5 6 7 8 7 8 9 10 11 12 13
1489 12 13 14 15 16 17 18 9 10 11 12 13 14 15 14 15 16 17d18d19 20d
1490 19 20 21 22 23 24 25 16 17 18 19 20 21 22 21d22d23d24d25d26d27d
1491 26 27 28 29 30 31m 23 24 25 26 27 28 29 28d29d30d31d
1492 30m
1493 List view
1495 ---------
1496 --keep-daily 14 --keep-monthly 6
1498 -------------------------------------------------
1499 1. 2015-12-31 (2015-12-31 kept by daily rule)
1500 2. 2015-12-30 1. 2015-11-30
1501 3. 2015-12-29 2. 2015-10-31
1502 4. 2015-12-28 3. 2015-09-30
1503 5. 2015-12-27 4. 2015-08-31
1504 6. 2015-12-26 5. 2015-07-31
1505 7. 2015-12-25 6. 2015-06-30
1506 8. 2015-12-24
1507 9. 2015-12-23
1508 10. 2015-12-22
1509 11. 2015-12-21
1510 12. 2015-12-20
1511 (no backup made on 2015-12-19)
1512 13. 2015-12-18
1513 14. 2015-12-17
1514 Notes
1517 -----
1518 2015-12-31 is kept due to the --keep-daily 14 rule (because it is applied
1520 first), not due to the --keep-monthly rule.
1521 Because of that, the --keep-monthly 6 rule keeps Nov, Oct, Sep, Aug, Jul and
1523 Jun. December is not considered for this rule, because that backup was already
1524 kept because of the daily rule.
1525 2015-12-17 is kept to satisfy the --keep-daily 14 rule - because no backup was
1527 made on 2015-12-19. If a backup had been made on that day, it would not keep
1528 the one from 2015-12-17.
1529 We did not include yearly, weekly, hourly, minutely or secondly rules to keep
1531 this example simple. They all work in basically the same way.
1532 The weekly rule is easy to understand roughly, but hard to understand in all
1534 details. If interested, read "ISO 8601:2000 standard week-based year".
1535
1538 borg [common options] info [options] [REPOSITORY_OR_ARCHIVE]
1539 Description
1541 This command displays detailed information about the specified archive
1542 or repository.
1543 Please note that the deduplicated sizes of the individual archives do
1545 not add up to the deduplicated size of the repository ("all archives"),
1546 because the two are meaning different things:
1547 This archive / deduplicated size = amount of data stored ONLY for this
1549 archive = unique chunks of this archive. All archives / deduplicated
1550 size = amount of data stored in the repo = all chunks in the reposi‐
1551 tory.
1552 Borg archives can only contain a limited amount of file metadata. The
1554 size of an archive relative to this limit depends on a number of fac‐
1555 tors, mainly the number of files, the lengths of paths and other meta‐
1556 data stored for files. This is shown as utilization of maximum sup‐
1557 ported archive size.
1558 Examples
1560 $ borg info /path/to/repo::2017-06-29T11:00-srv
1561 Archive name: 2017-06-29T11:00-srv
1562 Archive fingerprint: b2f1beac2bd553b34e06358afa45a3c1689320d39163890c5bbbd49125f00fe5
1563 Comment:
1564 Hostname: myhostname
1565 Username: root
1566 Time (start): Thu, 2017-06-29 11:03:07
1567 Time (end): Thu, 2017-06-29 11:03:13
1568 Duration: 5.66 seconds
1569 Number of files: 17037
1570 Command line: /usr/sbin/borg create /path/to/repo::2017-06-29T11:00-srv /srv
1571 Utilization of max. archive size: 0%
1572 ------------------------------------------------------------------------------
1573 Original size Compressed size Deduplicated size
1574 This archive: 12.53 GB 12.49 GB 1.62 kB
1575 All archives: 121.82 TB 112.41 TB 215.42 GB
1576 Unique chunks Total chunks
1578 Chunk index: 1015213 626934122
1579 $ borg info /path/to/repo --last 1
1581 Archive name: 2017-06-29T11:00-srv
1582 Archive fingerprint: b2f1beac2bd553b34e06358afa45a3c1689320d39163890c5bbbd49125f00fe5
1583 Comment:
1584 Hostname: myhostname
1585 Username: root
1586 Time (start): Thu, 2017-06-29 11:03:07
1587 Time (end): Thu, 2017-06-29 11:03:13
1588 Duration: 5.66 seconds
1589 Number of files: 17037
1590 Command line: /usr/sbin/borg create /path/to/repo::2017-06-29T11:00-srv /srv
1591 Utilization of max. archive size: 0%
1592 ------------------------------------------------------------------------------
1593 Original size Compressed size Deduplicated size
1594 This archive: 12.53 GB 12.49 GB 1.62 kB
1595 All archives: 121.82 TB 112.41 TB 215.42 GB
1596 Unique chunks Total chunks
1598 Chunk index: 1015213 626934122
1599 $ borg info /path/to/repo
1601 Repository ID: d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1602 Location: /path/to/repo
1603 Encrypted: Yes (repokey)
1604 Cache: /root/.cache/borg/d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1605 Security dir: /root/.config/borg/security/d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1606 ------------------------------------------------------------------------------
1607 Original size Compressed size Deduplicated size
1608 All archives: 121.82 TB 112.41 TB 215.42 GB
1609 Unique chunks Total chunks
1611 Chunk index: 1015213 626934122
1612
1614 borg [common options] mount [options] REPOSITORY_OR_ARCHIVE MOUNTPOINT [PATH...]
1615 Description
1617 This command mounts an archive as a FUSE filesystem. This can be useful
1618 for browsing an archive or restoring individual files. Unless the
1619 --foreground option is given the command will run in the background
1620 until the filesystem is umounted.
1621 The command borgfs provides a wrapper for borg mount. This can also be
1623 used in fstab entries: /path/to/repo /mnt/point fuse.borgfs
1624 defaults,noauto 0 0
1625 To allow a regular user to use fstab entries, add the user option:
1627 /path/to/repo /mnt/point fuse.borgfs defaults,noauto,user 0 0
1628 For FUSE configuration and mount options, see the mount.fuse(8) manual
1630 page.
1631 Additional mount options supported by borg:
1633 · versions: when used with a repository mount, this gives a merged,
1635 versioned view of the files in the archives. EXPERIMENTAL, layout may
1636 change in future.
1637 · allow_damaged_files: by default damaged files (where missing chunks
1639 were replaced with runs of zeros by borg check --repair) are not
1640 readable and return EIO (I/O error). Set this option to read such
1641 files.
1642 · ignore_permissions: for security reasons the "default_permissions"
1644 mount option is internally enforced by borg. "ignore_permissions" can
1645 be given to not enforce "default_permissions".
1646 The BORG_MOUNT_DATA_CACHE_ENTRIES environment variable is meant for
1648 advanced users to tweak the performance. It sets the number of cached
1649 data chunks; additional memory usage can be up to ~8 MiB times this
1650 number. The default is the number of CPU cores.
1651 When the daemonized process receives a signal or crashes, it does not
1653 unmount. Unmounting in these cases could cause an active rsync or sim‐
1654 ilar process to unintentionally delete data.
1655 When running in the foreground ^C/SIGINT unmounts cleanly, but other
1657 signals or crashes do not.
1658
1660 borg [common options] umount [options] MOUNTPOINT
1661 Description
1663 This command un-mounts a FUSE filesystem that was mounted with borg
1664 mount.
1665 This is a convenience wrapper that just calls the platform-specific
1667 shell command - usually this is either umount or fusermount -u.
1668 Examples
1670 # Mounting the repository shows all archives.
1671 # Archives are loaded lazily, expect some delay when navigating to an archive
1672 # for the first time.
1673 $ borg mount /path/to/repo /tmp/mymountpoint
1674 $ ls /tmp/mymountpoint
1675 root-2016-02-14 root-2016-02-15
1676 $ borg umount /tmp/mymountpoint
1677 # Mounting a specific archive is possible as well.
1679 $ borg mount /path/to/repo::root-2016-02-15 /tmp/mymountpoint
1680 $ ls /tmp/mymountpoint
1681 bin boot etc home lib lib64 lost+found media mnt opt
1682 root sbin srv tmp usr var
1683 $ borg umount /tmp/mymountpoint
1684 # The experimental "versions view" merges all archives in the repository
1686 # and provides a versioned view on files.
1687 $ borg mount -o versions /path/to/repo /tmp/mymountpoint
1688 $ ls -l /tmp/mymountpoint/home/user/doc.txt/
1689 total 24
1690 -rw-rw-r-- 1 user group 12357 Aug 26 21:19 doc.cda00bc9.txt
1691 -rw-rw-r-- 1 user group 12204 Aug 26 21:04 doc.fa760f28.txt
1692 $ borg umount /tmp/mymountpoint
1693 # Archive filters are supported.
1695 # These are especially handy for the "versions view",
1696 # which does not support lazy processing of archives.
1697 $ borg mount -o versions --glob-archives '*-my-home' --last 10 /path/to/repo /tmp/mymountpoint
1698 # Exclusion options are supported.
1700 # These can speed up mounting and lower memory needs significantly.
1701 $ borg mount /path/to/repo /tmp/mymountpoint only/that/path
1702 $ borg mount --exclude '...' /path/to/repo /tmp/mymountpoint
1703 borgfs
1705 $ echo '/mnt/backup /tmp/myrepo fuse.borgfs defaults,noauto 0 0' >> /etc/fstab
1706 $ echo '/mnt/backup::root-2016-02-15 /tmp/myarchive fuse.borgfs defaults,noauto 0 0' >> /etc/fstab
1707 $ mount /tmp/myrepo
1708 $ mount /tmp/myarchive
1709 $ ls /tmp/myrepo
1710 root-2016-02-01 root-2016-02-2015
1711 $ ls /tmp/myarchive
1712 bin boot etc home lib lib64 lost+found media mnt opt root sbin srv tmp usr var
1713 NOTE:
1715 borgfs will be automatically provided if you used a distribution
1716 package, pip or setup.py to install Borg. Users of the standalone
1717 binary will have to manually create a symlink (see pyin‐
1718 staller-binary).
1719
1721 borg [common options] key change-passphrase [options] [REPOSITORY]
1722 Description
1724 The key files used for repository encryption are optionally passphrase
1725 protected. This command can be used to change this passphrase.
1726 Please note that this command only changes the passphrase, but not any
1728 secret protected by it (like e.g. encryption/MAC keys or chunker seed).
1729 Thus, changing the passphrase after passphrase and borg key got compro‐
1730 mised does not protect future (nor past) backups to the same reposi‐
1731 tory.
1732 Examples
1734 # Create a key file protected repository
1735 $ borg init --encryption=keyfile -v /path/to/repo
1736 Initializing repository at "/path/to/repo"
1737 Enter new passphrase:
1738 Enter same passphrase again:
1739 Remember your passphrase. Your data will be inaccessible without it.
1740 Key in "/root/.config/borg/keys/mnt_backup" created.
1741 Keep this key safe. Your data will be inaccessible without it.
1742 Synchronizing chunks cache...
1743 Archives: 0, w/ cached Idx: 0, w/ outdated Idx: 0, w/o cached Idx: 0.
1744 Done.
1745 # Change key file passphrase
1747 $ borg key change-passphrase -v /path/to/repo
1748 Enter passphrase for key /root/.config/borg/keys/mnt_backup:
1749 Enter new passphrase:
1750 Enter same passphrase again:
1751 Remember your passphrase. Your data will be inaccessible without it.
1752 Key updated
1753 Fully automated using environment variables:
1755 $ BORG_NEW_PASSPHRASE=old borg init -e=repokey repo
1757 # now "old" is the current passphrase.
1758 $ BORG_PASSPHRASE=old BORG_NEW_PASSPHRASE=new borg key change-passphrase repo
1759 # now "new" is the current passphrase.
1760
1762 borg [common options] key export [options] [REPOSITORY] [PATH]
1763 Description
1765 If repository encryption is used, the repository is inaccessible with‐
1766 out the key. This command allows to backup this essential key. Note
1767 that the backup produced does not include the passphrase itself (i.e.
1768 the exported key stays encrypted). In order to regain access to a
1769 repository, one needs both the exported key and the original
1770 passphrase.
1771 There are two backup formats. The normal backup format is suitable for
1773 digital storage as a file. The --paper backup format is optimized for
1774 printing and typing in while importing, with per line checks to reduce
1775 problems with manual input.
1776 For repositories using keyfile encryption the key is saved locally on
1778 the system that is capable of doing backups. To guard against loss of
1779 this key, the key needs to be backed up independently of the main data
1780 backup.
1781 For repositories using the repokey encryption the key is saved in the
1783 repository in the config file. A backup is thus not strictly needed,
1784 but guards against the repository becoming inaccessible if the file is
1785 damaged for some reason.
1786
1788 borg [common options] key import [options] [REPOSITORY] [PATH]
1789 Description
1791 This command allows to restore a key previously backed up with the
1792 export command.
1793 If the --paper option is given, the import will be an interactive
1795 process in which each line is checked for plausibility before proceed‐
1796 ing to the next line. For this format PATH must not be given.
1797
1799 borg [common options] upgrade [options] [REPOSITORY]
1800 Description
1802 Upgrade an existing, local Borg repository.
1803 When you do not need borg upgrade
1805 Not every change requires that you run borg upgrade.
1806 You do not need to run it when:
1808 · moving your repository to a different place
1810 · upgrading to another point release (like 1.0.x to 1.0.y), except when
1812 noted otherwise in the changelog
1813 · upgrading from 1.0.x to 1.1.x, except when noted otherwise in the
1815 changelog
1816 Borg 1.x.y upgrades
1818 Use borg upgrade --tam REPO to require manifest authentication intro‐
1819 duced with Borg 1.0.9 to address security issues. This means that modi‐
1820 fying the repository after doing this with a version prior to 1.0.9
1821 will raise a validation error, so only perform this upgrade after
1822 updating all clients using the repository to 1.0.9 or newer.
1823 This upgrade should be done on each client for safety reasons.
1825 If a repository is accidentally modified with a pre-1.0.9 client after
1827 this upgrade, use borg upgrade --tam --force REPO to remedy it.
1828 If you routinely do this you might not want to enable this upgrade
1830 (which will leave you exposed to the security issue). You can reverse
1831 the upgrade by issuing borg upgrade --disable-tam REPO.
1832 See
1834 https://borgbackup.readthedocs.io/en/stable/changes.html#pre-1-0-9-manifest-spoofing-vulnerability
1835 for details.
1836 Attic and Borg 0.xx to Borg 1.x
1838 This currently supports converting an Attic repository to Borg and also
1839 helps with converting Borg 0.xx to 1.0.
1840 Currently, only LOCAL repositories can be upgraded (issue #465).
1842 Please note that borg create (since 1.0.0) uses bigger chunks by
1844 default than old borg or attic did, so the new chunks won't deduplicate
1845 with the old chunks in the upgraded repository. See --chunker-params
1846 option of borg create and borg recreate.
1847 borg upgrade will change the magic strings in the repository's segments
1849 to match the new Borg magic strings. The keyfiles found in
1850 $ATTIC_KEYS_DIR or ~/.attic/keys/ will also be converted and copied to
1851 $BORG_KEYS_DIR or ~/.config/borg/keys.
1852 The cache files are converted, from $ATTIC_CACHE_DIR or ~/.cache/attic
1854 to $BORG_CACHE_DIR or ~/.cache/borg, but the cache layout between Borg
1855 and Attic changed, so it is possible the first backup after the conver‐
1856 sion takes longer than expected due to the cache resync.
1857 Upgrade should be able to resume if interrupted, although it will still
1859 iterate over all segments. If you want to start from scratch, use borg
1860 delete over the copied repository to make sure the cache files are also
1861 removed:
1862 borg delete borg
1863 Unless --inplace is specified, the upgrade process first creates a
1865 backup copy of the repository, in REPOSITORY.before-upgrade-DATETIME,
1866 using hardlinks. This requires that the repository and its parent
1867 directory reside on same filesystem so the hardlink copy can work.
1868 This takes longer than in place upgrades, but is much safer and gives
1869 progress information (as opposed to cp -al). Once you are satisfied
1870 with the conversion, you can safely destroy the backup copy.
1871 WARNING: Running the upgrade in place will make the current copy unus‐
1873 able with older version, with no way of going back to previous ver‐
1874 sions. This can PERMANENTLY DAMAGE YOUR REPOSITORY! Attic CAN NOT READ
1875 BORG REPOSITORIES, as the magic strings have changed. You have been
1876 warned.
1877 Examples
1879 # Upgrade the borg repository to the most recent version.
1880 $ borg upgrade -v /path/to/repo
1881 making a hardlink copy in /path/to/repo.before-upgrade-2016-02-15-20:51:55
1882 opening attic repository with borg and converting
1883 no key file found for repository
1884 converting repo index /path/to/repo/index.0
1885 converting 1 segments...
1886 converting borg 0.xx to borg current
1887 no key file found for repository
1888 Upgrading a passphrase encrypted attic repo
1890 attic offered a "passphrase" encryption mode, but this was removed in
1891 borg 1.0 and replaced by the "repokey" mode (which stores the
1892 passphrase-protected encryption key into the repository config).
1893 Thus, to upgrade a "passphrase" attic repo to a "repokey" borg repo, 2
1895 steps are needed, in this order:
1896 · borg upgrade repo
1898 · borg key migrate-to-repokey repo
1900
1902 borg [common options] recreate [options] [REPOSITORY_OR_ARCHIVE] [PATH...]
1903 Description
1905 Recreate the contents of existing archives.
1906 This is an experimental feature. Do not use this on your only backup.
1908 --exclude, --exclude-from, --exclude-if-present, --keep-exclude-tags,
1910 and PATH have the exact same semantics as in "borg create". If PATHs
1911 are specified the resulting archive will only contain files from these
1912 PATHs.
1913 Note that all paths in an archive are relative, therefore absolute pat‐
1915 terns/paths will not match (--exclude, --exclude-from, PATHs).
1916 --recompress allows to change the compression of existing data in ar‐
1918 chives. Due to how Borg stores compressed size information this might
1919 display incorrect information for archives that were not recreated at
1920 the same time. There is no risk of data loss by this.
1921 --chunker-params will re-chunk all files in the archive, this can be
1923 used to have upgraded Borg 0.xx or Attic archives deduplicate with Borg
1924 1.x archives.
1925 USE WITH CAUTION. Depending on the PATHs and patterns given, recreate
1927 can be used to permanently delete files from archives. When in doubt,
1928 use --dry-run --verbose --list to see how patterns/PATHS are inter‐
1929 preted.
1930 The archive being recreated is only removed after the operation com‐
1932 pletes. The archive that is built during the operation exists at the
1933 same time at "<ARCHIVE>.recreate". The new archive will have a differ‐
1934 ent archive ID.
1935 With --target the original archive is not replaced, instead a new ar‐
1937 chive is created.
1938 When rechunking space usage can be substantial, expect at least the
1940 entire deduplicated size of the archives using the previous chunker
1941 params. When recompressing expect approx. (throughput / check‐
1942 point-interval) in space usage, assuming all chunks are recompressed.
1943 If you recently ran borg check --repair and it had to fix lost chunks
1945 with all-zero replacement chunks, please first run another backup for
1946 the same data and re-run borg check --repair afterwards to heal any ar‐
1947 chives that had lost chunks which are still generated from the input
1948 data.
1949 Important: running borg recreate to re-chunk will remove the
1951 chunks_healthy metadata of all items with replacement chunks, so heal‐
1952 ing will not be possible any more after re-chunking (it is also
1953 unlikely it would ever work: due to the change of chunking parameters,
1954 the missing chunk likely will never be seen again even if you still
1955 have the data that produced it).
1956 Examples
1958 # Make old (Attic / Borg 0.xx) archives deduplicate with Borg 1.x archives.
1959 # Archives created with Borg 1.1+ and the default chunker params are skipped
1960 # (archive ID stays the same).
1961 $ borg recreate /mnt/backup --chunker-params default --progress
1962 # Create a backup with little but fast compression
1964 $ borg create /mnt/backup::archive /some/files --compression lz4
1965 # Then compress it - this might take longer, but the backup has already completed,
1966 # so no inconsistencies from a long-running backup job.
1967 $ borg recreate /mnt/backup::archive --recompress --compression zlib,9
1968 # Remove unwanted files from all archives in a repository.
1970 # Note the relative path for the --exclude option - archives only contain relative paths.
1971 $ borg recreate /mnt/backup --exclude home/icke/Pictures/drunk_photos
1972 # Change archive comment
1974 $ borg create --comment "This is a comment" /mnt/backup::archivename ~
1975 $ borg info /mnt/backup::archivename
1976 Name: archivename
1977 Fingerprint: ...
1978 Comment: This is a comment
1979 ...
1980 $ borg recreate --comment "This is a better comment" /mnt/backup::archivename
1981 $ borg info /mnt/backup::archivename
1982 Name: archivename
1983 Fingerprint: ...
1984 Comment: This is a better comment
1985 ...
1986
1988 borg [common options] export-tar [options] ARCHIVE FILE [PATH...]
1989 Description
1991 This command creates a tarball from an archive.
1992 When giving '-' as the output FILE, Borg will write a tar stream to
1994 standard output.
1995 By default (--tar-filter=auto) Borg will detect whether the FILE should
1997 be compressed based on its file extension and pipe the tarball through
1998 an appropriate filter before writing it to FILE:
1999 · .tar.gz: gzip
2001 · .tar.bz2: bzip2
2003 · .tar.xz: xz
2005 Alternatively a --tar-filter program may be explicitly specified. It
2007 should read the uncompressed tar stream from stdin and write a com‐
2008 pressed/filtered tar stream to stdout.
2009 The generated tarball uses the GNU tar format.
2011 export-tar is a lossy conversion: BSD flags, ACLs, extended attributes
2013 (xattrs), atime and ctime are not exported. Timestamp resolution is
2014 limited to whole seconds, not the nanosecond resolution otherwise sup‐
2015 ported by Borg.
2016 A --sparse option (as found in borg extract) is not supported.
2018 By default the entire archive is extracted but a subset of files and
2020 directories can be selected by passing a list of PATHs as arguments.
2021 The file selection can further be restricted by using the --exclude
2022 option.
2023 See the output of the "borg help patterns" command for more help on
2025 exclude patterns.
2026 --progress can be slower than no progress display, since it makes one
2028 additional pass over the archive metadata.
2029 Examples
2031 # export as uncompressed tar
2032 $ borg export-tar /path/to/repo::Monday Monday.tar
2033 # exclude some types, compress using gzip
2035 $ borg export-tar /path/to/repo::Monday Monday.tar.gz --exclude '*.so'
2036 # use higher compression level with gzip
2038 $ borg export-tar --tar-filter="gzip -9" testrepo::linux Monday.tar.gz
2039 # export a tar, but instead of storing it on disk,
2041 # upload it to a remote site using curl.
2042 $ borg export-tar /path/to/repo::Monday - | curl --data-binary @- https://somewhere/to/POST
2043 # remote extraction via "tarpipe"
2045 $ borg export-tar /path/to/repo::Monday - | ssh somewhere "cd extracted; tar x"
2046
2048 borg [common options] serve [options]
2049 Description
2051 This command starts a repository server process. This command is usu‐
2052 ally not used manually.
2053 Examples
2055 borg serve has special support for ssh forced commands (see autho‐
2056 rized_keys example below): it will detect that you use such a forced
2057 command and extract the value of the --restrict-to-path option(s).
2058 It will then parse the original command that came from the client,
2060 makes sure that it is also borg serve and enforce path restriction(s)
2061 as given by the forced command. That way, other options given by the
2062 client (like --info or --umask) are preserved (and are not fixed by the
2063 forced command).
2064 Environment variables (such as BORG_HOSTNAME_IS_UNIQUE) contained in
2066 the original command sent by the client are not interpreted, but
2067 ignored. If BORG_XXX environment variables should be set on the borg
2068 serve side, then these must be set in system-specific locations like
2069 /etc/environment or in the forced command itself (example below).
2070 # Allow an SSH keypair to only run borg, and only have access to /path/to/repo.
2072 # Use key options to disable unneeded and potentially dangerous SSH functionality.
2073 # This will help to secure an automated remote backup system.
2074 $ cat ~/.ssh/authorized_keys
2075 command="borg serve --restrict-to-path /path/to/repo",restrict ssh-rsa AAAAB3[...]
2076 # Set a BORG_XXX environment variable on the "borg serve" side
2078 $ cat ~/.ssh/authorized_keys
2079 command="export BORG_XXX=value; borg serve [...]",restrict ssh-rsa [...]
2080 NOTE:
2082 The examples above use the restrict directive. This does automati‐
2083 cally block potential dangerous ssh features, even when they are
2084 added in a future update. Thus, this option should be preferred.
2085 If you're using openssh-server < 7.2, however, you have to explic‐
2087 itly specify the ssh features to restrict and cannot simply use the
2088 restrict option as it has been introduced in v7.2. We recommend to
2089 use no-port-forwarding,no-X11-forwarding,no-pty,no-agent-forward‐
2090 ing,no-user-rc in this case.
2091 SSH Configuration
2093 borg serve's pipes (stdin/stdout/stderr) are connected to the sshd
2094 process on the server side. In the event that the SSH connection
2095 between borg serve and the client is disconnected or stuck abnormally
2096 (for example, due to a network outage), it can take a long time for
2097 sshd to notice the client is disconnected. In the meantime, sshd con‐
2098 tinues running, and as a result so does the borg serve process holding
2099 the lock on the repository. This can cause subsequent borg operations
2100 on the remote repository to fail with the error: Failed to cre‐
2101 ate/acquire the lock.
2102 In order to avoid this, it is recommended to perform the following
2104 additional SSH configuration:
2105 Either in the client side's ~/.ssh/config file, or in the client's
2107 /etc/ssh/ssh_config file:
2108 Host backupserver
2110 ServerAliveInterval 10
2111 ServerAliveCountMax 30
2112 Replacing backupserver with the hostname, FQDN or IP address of the
2114 borg server.
2115 This will cause the client to send a keepalive to the server every 10
2117 seconds. If 30 consecutive keepalives are sent without a response (a
2118 time of 300 seconds), the ssh client process will be terminated, caus‐
2119 ing the borg process to terminate gracefully.
2120 On the server side's sshd configuration file (typically
2122 /etc/ssh/sshd_config):
2123 ClientAliveInterval 10
2125 ClientAliveCountMax 30
2126 This will cause the server to send a keep alive to the client every 10
2128 seconds. If 30 consecutive keepalives are sent without a response (a
2129 time of 300 seconds), the server's sshd process will be terminated,
2130 causing the borg serve process to terminate gracefully and release the
2131 lock on the repository.
2132 If you then run borg commands with --lock-wait 600, this gives suffi‐
2134 cient time for the borg serve processes to terminate after the SSH con‐
2135 nection is torn down after the 300 second wait for the keepalives to
2136 fail.
2137 You may, of course, modify the timeout values demonstrated above to
2139 values that suit your environment and use case.
2140
2142 borg [common options] config [options] REPOSITORY [NAME] [VALUE]
2143 Description
2145 This command gets and sets options in a local repository or cache con‐
2146 fig file. For security reasons, this command only works on local
2147 repositories.
2148 To delete a config value entirely, use --delete. To list the values of
2150 the configuration file or the default values, use --list. To get and
2151 existing key, pass only the key name. To set a key, pass both the key
2152 name and the new value. Keys can be specified in the format "sec‐
2153 tion.name" or simply "name"; the section will default to "repository"
2154 and "cache" for the repo and cache configs, respectively.
2155 By default, borg config manipulates the repository config file. Using
2157 --cache edits the repository cache's config file instead.
2158 NOTE:
2160 The repository & cache config files are some of the only directly
2161 manipulable parts of a repository that aren't versioned or backed
2162 up, so be careful when making changes!
2163 Examples
2165 # find cache directory
2166 $ cd ~/.cache/borg/$(borg config /path/to/repo id)
2167 # reserve some space
2169 $ borg config /path/to/repo additional_free_space 2G
2170 # make a repo append-only
2172 $ borg config /path/to/repo append_only 1
2173
2175 borg [common options] with-lock [options] REPOSITORY COMMAND [ARGS...]
2176 Description
2178 This command runs a user-specified command while the repository lock is
2179 held.
2180 It will first try to acquire the lock (make sure that no other opera‐
2182 tion is running in the repo), then execute the given command as a sub‐
2183 process and wait for its termination, release the lock and return the
2184 user command's return code as borg's return code.
2185 NOTE:
2187 If you copy a repository with the lock held, the lock will be
2188 present in the copy. Thus, before using borg on the copy from a dif‐
2189 ferent host, you need to use "borg break-lock" on the copied reposi‐
2190 tory, because Borg is cautious and does not automatically remove
2191 stale locks made by a different host.
2192
2194 borg [common options] break-lock [options] [REPOSITORY]
2195 Description
2197 This command breaks the repository and cache locks. Please use care‐
2198 fully and only while no borg process (on any machine) is trying to
2199 access the Cache or the Repository.
2200
2202 borg [common options] benchmark crud [options] REPO PATH
2203 Description
2205 This command benchmarks borg CRUD (create, read, update, delete) opera‐
2206 tions.
2207 It creates input data below the given PATH and backups this data into
2209 the given REPO. The REPO must already exist (it could be a fresh empty
2210 repo or an existing repo, the command will create / read / update /
2211 delete some archives named borg-benchmark-crud* there.
2212 Make sure you have free space there, you'll need about 1GB each (+
2214 overhead).
2215 If your repository is encrypted and borg needs a passphrase to unlock
2217 the key, use:
2218 BORG_PASSPHRASE=mysecret borg benchmark crud REPO PATH
2220 Measurements are done with different input file sizes and counts. The
2222 file contents are very artificial (either all zero or all random), thus
2223 the measurement results do not necessarily reflect performance with
2224 real data. Also, due to the kind of content used, no compression is
2225 used in these benchmarks.
2226 C- == borg create (1st archive creation, no compression, do not use
2228 files cache)
2229 C-Z- == all-zero files. full dedup, this is primarily measuring
2230 reader/chunker/hasher. C-R- == random files. no dedup, measur‐
2231 ing throughput through all processing stages.
2232 R- == borg extract (extract archive, dry-run, do everything, but do not
2234 write files to disk)
2235 R-Z- == all zero files. Measuring heavily duplicated files.
2236 R-R- == random files. No duplication here, measuring throughput
2237 through all processing stages, except writing to disk.
2238 U- == borg create (2nd archive creation of unchanged input files, mea‐
2240 sure files cache speed)
2241 The throughput value is kind of virtual here, it does not actu‐
2242 ally read the file. U-Z- == needs to check the 2 all-zero
2243 chunks' existence in the repo. U-R- == needs to check existence
2244 of a lot of different chunks in the repo.
2245 D- == borg delete archive (delete last remaining archive, measure dele‐
2247 tion + compaction)
2248 D-Z- == few chunks to delete / few segments to compact/remove.
2249 D-R- == many chunks to delete / many segments to compact/remove.
2250 Please note that there might be quite some variance in these measure‐
2252 ments. Try multiple measurements and having a otherwise idle machine
2253 (and network, if you use it).
2254
2256 borg help patterns
2257 The path/filenames used as input for the pattern matching start from
2258 the currently active recursion root. You usually give the recursion
2259 root(s) when invoking borg and these can be either relative or absolute
2260 paths.
2261 So, when you give relative/ as root, the paths going into the matcher
2263 will look like relative/.../file.ext. When you give /absolute/ as root,
2264 they will look like /absolute/.../file.ext. This is meant when we talk
2265 about "full path" below.
2266 File patterns support these styles: fnmatch, shell, regular expres‐
2268 sions, path prefixes and path full-matches. By default, fnmatch is used
2269 for --exclude patterns and shell-style is used for the experimental
2270 --pattern option.
2271 If followed by a colon (':') the first two characters of a pattern are
2273 used as a style selector. Explicit style selection is necessary when a
2274 non-default style is desired or when the desired pattern starts with
2275 two alphanumeric characters followed by a colon (i.e. aa:something/*).
2276 Fnmatch, selector fm:
2278 This is the default style for --exclude and --exclude-from.
2279 These patterns use a variant of shell pattern syntax, with '*'
2280 matching any number of characters, '?' matching any single char‐
2281 acter, '[...]' matching any single character specified, includ‐
2282 ing ranges, and '[!...]' matching any character not specified.
2283 For the purpose of these patterns, the path separator (backslash
2284 for Windows and '/' on other systems) is not treated specially.
2285 Wrap meta-characters in brackets for a literal match (i.e. [?]
2286 to match the literal character ?). For a path to match a pat‐
2287 tern, the full path must match, or it must match from the start
2288 of the full path to just before a path separator. Except for the
2289 root path, paths will never end in the path separator when
2290 matching is attempted. Thus, if a given pattern ends in a path
2291 separator, a '*' is appended before matching is attempted.
2292 Shell-style patterns, selector sh:
2294 This is the default style for --pattern and --patterns-from.
2295 Like fnmatch patterns these are similar to shell patterns. The
2296 difference is that the pattern may include **/ for matching zero
2297 or more directory levels, * for matching zero or more arbitrary
2298 characters with the exception of any path separator.
2299 Regular expressions, selector re:
2301 Regular expressions similar to those found in Perl are sup‐
2302 ported. Unlike shell patterns regular expressions are not
2303 required to match the full path and any substring match is suf‐
2304 ficient. It is strongly recommended to anchor patterns to the
2305 start ('^'), to the end ('$') or both. Path separators (back‐
2306 slash for Windows and '/' on other systems) in paths are always
2307 normalized to a forward slash ('/') before applying a pattern.
2308 The regular expression syntax is described in the Python docu‐
2309 mentation for the re module.
2310 Path prefix, selector pp:
2312 This pattern style is useful to match whole sub-directories. The
2313 pattern pp:root/somedir matches root/somedir and everything
2314 therein.
2315 Path full-match, selector pf:
2317 This pattern style is (only) useful to match full paths. This
2318 is kind of a pseudo pattern as it can not have any variable or
2319 unspecified parts - the full path must be given.
2320 pf:root/file.ext matches root/file.txt only.
2321 Implementation note: this is implemented via very time-efficient
2323 O(1) hashtable lookups (this means you can have huge amounts of
2324 such patterns without impacting performance much). Due to that,
2325 this kind of pattern does not respect any context or order. If
2326 you use such a pattern to include a file, it will always be
2327 included (if the directory recursion encounters it). Other
2328 include/exclude patterns that would normally match will be
2329 ignored. Same logic applies for exclude.
2330 NOTE:
2332 re:, sh: and fm: patterns are all implemented on top of the Python
2333 SRE engine. It is very easy to formulate patterns for each of these
2334 types which requires an inordinate amount of time to match paths. If
2335 untrusted users are able to supply patterns, ensure they cannot sup‐
2336 ply re: patterns. Further, ensure that sh: and fm: patterns only
2337 contain a handful of wildcards at most.
2338 Exclusions can be passed via the command line option --exclude. When
2340 used from within a shell the patterns should be quoted to protect them
2341 from expansion.
2342 The --exclude-from option permits loading exclusion patterns from a
2344 text file with one pattern per line. Lines empty or starting with the
2345 number sign ('#') after removing whitespace on both ends are ignored.
2346 The optional style selector prefix is also supported for patterns
2347 loaded from a file. Due to whitespace removal paths with whitespace at
2348 the beginning or end can only be excluded using regular expressions.
2349 Examples:
2351 # Exclude '/home/user/file.o' but not '/home/user/file.odt':
2353 $ borg create -e '*.o' backup /
2354 # Exclude '/home/user/junk' and '/home/user/subdir/junk' but
2356 # not '/home/user/importantjunk' or '/etc/junk':
2357 $ borg create -e '/home/*/junk' backup /
2358 # Exclude the contents of '/home/user/cache' but not the directory itself:
2360 $ borg create -e /home/user/cache/ backup /
2361 # The file '/home/user/cache/important' is *not* backed up:
2363 $ borg create -e /home/user/cache/ backup / /home/user/cache/important
2364 # The contents of directories in '/home' are not backed up when their name
2366 # ends in '.tmp'
2367 $ borg create --exclude 're:^/home/[^/]+\.tmp/' backup /
2368 # Load exclusions from file
2370 $ cat >exclude.txt <<EOF
2371 # Comment line
2372 /home/*/junk
2373 *.tmp
2374 fm:aa:something/*
2375 re:^/home/[^/]\.tmp/
2376 sh:/home/*/.thumbnails
2377 EOF
2378 $ borg create --exclude-from exclude.txt backup /
2379 A more general and easier to use way to define filename matching pat‐
2380 terns exists with the experimental --pattern and --patterns-from
2381 options. Using these, you may specify the backup roots (starting
2382 points) and patterns for inclusion/exclusion. A root path starts with
2383 the prefix R, followed by a path (a plain path, not a file pattern). An
2384 include rule starts with the prefix +, an exclude rule starts with the
2385 prefix -, an exclude-norecurse rule starts with !, all followed by a
2386 pattern. Inclusion patterns are useful to include paths that are con‐
2387 tained in an excluded path. The first matching pattern is used so if an
2388 include pattern matches before an exclude pattern, the file is backed
2389 up. If an exclude-norecurse pattern matches a directory, it won't
2390 recurse into it and won't discover any potential matches for include
2391 rules below that directory.
2392 Note that the default pattern style for --pattern and --patterns-from
2394 is shell style (sh:), so those patterns behave similar to rsync
2395 include/exclude patterns. The pattern style can be set via the P pre‐
2396 fix.
2397 Patterns (--pattern) and excludes (--exclude) from the command line are
2399 considered first (in the order of appearance). Then patterns from
2400 --patterns-from are added. Exclusion patterns from --exclude-from files
2401 are appended last.
2402 Examples:
2404 # backup pics, but not the ones from 2018, except the good ones:
2406 # note: using = is essential to avoid cmdline argument parsing issues.
2407 borg create --pattern=+pics/2018/good --pattern=-pics/2018 repo::arch pics
2408 # use a file with patterns:
2410 borg create --patterns-from patterns.lst repo::arch
2411 The patterns.lst file could look like that:
2413 # "sh:" pattern style is the default, so the following line is not needed:
2415 P sh
2416 R /
2417 # can be rebuild
2418 - /home/*/.cache
2419 # they're downloads for a reason
2420 - /home/*/Downloads
2421 # susan is a nice person
2422 # include susans home
2423 + /home/susan
2424 # don't backup the other home directories
2425 - /home/*
2426 borg help placeholders
2428 Repository (or Archive) URLs, --prefix, --glob-archives, --comment and
2429 --remote-path values support these placeholders:
2430 {hostname}
2432 The (short) hostname of the machine.
2433 {fqdn} The full name of the machine.
2435 {reverse-fqdn}
2437 The full name of the machine in reverse domain name notation.
2438 {now} The current local date and time, by default in ISO-8601 format.
2440 You can also supply your own format string, e.g.
2441 {now:%Y-%m-%d_%H:%M:%S}
2442 {utcnow}
2444 The current UTC date and time, by default in ISO-8601 format.
2445 You can also supply your own format string, e.g. {utc‐
2446 now:%Y-%m-%d_%H:%M:%S}
2447 {user} The user name (or UID, if no name is available) of the user run‐
2449 ning borg.
2450 {pid} The current process ID.
2452 {borgversion}
2454 The version of borg, e.g.: 1.0.8rc1
2455 {borgmajor}
2457 The version of borg, only the major version, e.g.: 1
2458 {borgminor}
2460 The version of borg, only major and minor version, e.g.: 1.0
2461 {borgpatch}
2463 The version of borg, only major, minor and patch version, e.g.:
2464 1.0.8
2465 If literal curly braces need to be used, double them for escaping:
2467 borg create /path/to/repo::{{literal_text}}
2469 Examples:
2471 borg create /path/to/repo::{hostname}-{user}-{utcnow} ...
2473 borg create /path/to/repo::{hostname}-{now:%Y-%m-%d_%H:%M:%S} ...
2474 borg prune --prefix '{hostname}-' ...
2475 NOTE:
2477 systemd uses a difficult, non-standard syntax for command lines in
2478 unit files (refer to the systemd.unit(5) manual page).
2479 When invoking borg from unit files, pay particular attention to
2481 escaping, especially when using the now/utcnow placeholders, since
2482 systemd performs its own %-based variable replacement even in quoted
2483 text. To avoid interference from systemd, double all percent signs
2484 ({hostname}-{now:%Y-%m-%d_%H:%M:%S} becomes {host‐
2485 name}-{now:%%Y-%%m-%%d_%%H:%%M:%%S}).
2486 borg help compression
2488 It is no problem to mix different compression methods in one repo,
2489 deduplication is done on the source data chunks (not on the compressed
2490 or encrypted data).
2491 If some specific chunk was once compressed and stored into the repo,
2493 creating another backup that also uses this chunk will not change the
2494 stored chunk. So if you use different compression specs for the back‐
2495 ups, whichever stores a chunk first determines its compression. See
2496 also borg recreate.
2497 Compression is lz4 by default. If you want something else, you have to
2499 specify what you want.
2500 Valid compression specifiers are:
2502 none Do not compress.
2504 lz4 Use lz4 compression. Very high speed, very low compression.
2506 (default)
2507 zstd[,L]
2509 Use zstd ("zstandard") compression, a modern wide-range algo‐
2510 rithm. If you do not explicitely give the compression level L
2511 (ranging from 1 to 22), it will use level 3. Archives com‐
2512 pressed with zstd are not compatible with borg < 1.1.4.
2513 zlib[,L]
2515 Use zlib ("gz") compression. Medium speed, medium compression.
2516 If you do not explicitely give the compression level L (ranging
2517 from 0 to 9), it will use level 6. Giving level 0 (means "no
2518 compression", but still has zlib protocol overhead) is usually
2519 pointless, you better use "none" compression.
2520 lzma[,L]
2522 Use lzma ("xz") compression. Low speed, high compression. If
2523 you do not explicitely give the compression level L (ranging
2524 from 0 to 9), it will use level 6. Giving levels above 6 is
2525 pointless and counterproductive because it does not compress
2526 better due to the buffer size used by borg - but it wastes lots
2527 of CPU cycles and RAM.
2528 auto,C[,L]
2530 Use a built-in heuristic to decide per chunk whether to compress
2531 or not. The heuristic tries with lz4 whether the data is com‐
2532 pressible. For incompressible data, it will not use compression
2533 (uses "none"). For compressible data, it uses the given C[,L]
2534 compression - with C[,L] being any valid compression specifier.
2535 Examples:
2537 borg create --compression lz4 REPO::ARCHIVE data
2539 borg create --compression zstd REPO::ARCHIVE data
2540 borg create --compression zstd,10 REPO::ARCHIVE data
2541 borg create --compression zlib REPO::ARCHIVE data
2542 borg create --compression zlib,1 REPO::ARCHIVE data
2543 borg create --compression auto,lzma,6 REPO::ARCHIVE data
2544 borg create --compression auto,lzma ...
2545
2547 There is a borg debug command that has some subcommands which are all
2548 not intended for normal use and potentially very dangerous if used
2549 incorrectly.
2550 For example, borg debug put-obj and borg debug delete-obj will only do
2552 what their name suggests: put objects into repo / delete objects from
2553 repo.
2554 Please note:
2556 · they will not update the chunks cache (chunks index) about the object
2558 · they will not update the manifest (so no automatic chunks index
2560 resync is triggered)
2561 · they will not check whether the object is in use (e.g. before
2563 delete-obj)
2564 · they will not update any metadata which may point to the object
2566 They exist to improve debugging capabilities without direct system
2568 access, e.g. in case you ever run into some severe malfunction. Use
2569 them only if you know what you are doing or if a trusted Borg developer
2570 tells you what to do.
2571 Borg has a --debug-topic TOPIC option to enable specific debugging mes‐
2573 sages. Topics are generally not documented.
2574 A --debug-profile FILE option exists which writes a profile of the main
2576 program's execution to a file. The format of these files is not
2577 directly compatible with the Python profiling tools, since these use
2578 the "marshal" format, which is not intended to be secure (quoting the
2579 Python docs: "Never unmarshal data received from an untrusted or unau‐
2580 thenticated source.").
2581 The borg debug profile-convert command can be used to take a Borg pro‐
2583 file and convert it to a profile file that is compatible with the
2584 Python tools.
2585 Additionally, if the filename specified for --debug-profile ends with
2587 ".pyprof" a Python compatible profile is generated. This is only
2588 intended for local use by developers.
2589
2591 Here are misc. notes about topics that are maybe not covered in enough
2592 detail in the usage section.
2593 --chunker-params
2595 The chunker params influence how input files are cut into pieces
2596 (chunks) which are then considered for deduplication. They also have a
2597 big impact on resource usage (RAM and disk space) as the amount of
2598 resources needed is (also) determined by the total amount of chunks in
2599 the repository (see cache-memory-usage for details).
2600 --chunker-params=10,23,16,4095 results in a fine-grained deduplication|
2602 and creates a big amount of chunks and thus uses a lot of resources to
2603 manage them. This is good for relatively small data volumes and if the
2604 machine has a good amount of free RAM and disk space.
2605 --chunker-params=19,23,21,4095 (default) results in a coarse-grained
2607 deduplication and creates a much smaller amount of chunks and thus uses
2608 less resources. This is good for relatively big data volumes and if the
2609 machine has a relatively low amount of free RAM and disk space.
2610 If you already have made some archives in a repository and you then
2612 change chunker params, this of course impacts deduplication as the
2613 chunks will be cut differently.
2614 In the worst case (all files are big and were touched in between back‐
2616 ups), this will store all content into the repository again.
2617 Usually, it is not that bad though:
2619 · usually most files are not touched, so it will just re-use the old
2621 chunks it already has in the repo
2622 · files smaller than the (both old and new) minimum chunksize result in
2624 only one chunk anyway, so the resulting chunks are same and dedupli‐
2625 cation will apply
2626 If you switch chunker params to save resources for an existing repo
2628 that already has some backup archives, you will see an increasing
2629 effect over time, when more and more files have been touched and stored
2630 again using the bigger chunksize and all references to the smaller
2631 older chunks have been removed (by deleting / pruning archives).
2632 If you want to see an immediate big effect on resource usage, you bet‐
2634 ter start a new repository when changing chunker params.
2635 For more details, see chunker_details.
2637 --noatime / --noctime
2639 You can use these borg create options to not store the respective time‐
2640 stamp into the archive, in case you do not really need it.
2641 Besides saving a little space for the not archived timestamp, it might
2643 also affect metadata stream deduplication: if only this timestamp
2644 changes between backups and is stored into the metadata stream, the
2645 metadata stream chunks won't deduplicate just because of that.
2646 --nobsdflags
2648 You can use this to not query and store (or not extract and set) bsd‐
2649 flags - in case you don't need them or if they are broken somehow for
2650 your fs.
2651 On Linux, dealing with the bsflags needs some additional syscalls.
2653 Especially when dealing with lots of small files, this causes a notica‐
2654 ble overhead, so you can use this option also for speeding up opera‐
2655 tions.
2656 --umask
2658 If you use --umask, make sure that all repository-modifying borg com‐
2659 mands (create, delete, prune) that access the repository in question
2660 use the same --umask value.
2661 If multiple machines access the same repository, this should hold true
2663 for all of them.
2664 --read-special
2666 The --read-special option is special - you do not want to use it for
2667 normal full-filesystem backups, but rather after carefully picking some
2668 targets for it.
2669 The option --read-special triggers special treatment for block and char
2671 device files as well as FIFOs. Instead of storing them as such a device
2672 (or FIFO), they will get opened, their content will be read and in the
2673 backup archive they will show up like a regular file.
2674 Symlinks will also get special treatment if (and only if) they point to
2676 such a special file: instead of storing them as a symlink, the target
2677 special file will get processed as described above.
2678 One intended use case of this is backing up the contents of one or mul‐
2680 tiple block devices, like e.g. LVM snapshots or inactive LVs or disk
2681 partitions.
2682 You need to be careful about what you include when using --read-spe‐
2684 cial, e.g. if you include /dev/zero, your backup will never terminate.
2685 Restoring such files' content is currently only supported one at a time
2687 via --stdout option (and you have to redirect stdout to where ever it
2688 shall go, maybe directly into an existing device file of your choice or
2689 indirectly via dd).
2690 To some extent, mounting a backup archive with the backups of special
2692 files via borg mount and then loop-mounting the image files from inside
2693 the mount point will work. If you plan to access a lot of data in
2694 there, it likely will scale and perform better if you do not work via
2695 the FUSE mount.
2696 Example
2698 Imagine you have made some snapshots of logical volumes (LVs) you want
2699 to backup.
2700 NOTE:
2702 For some scenarios, this is a good method to get "crash-like" con‐
2703 sistency (I call it crash-like because it is the same as you would
2704 get if you just hit the reset button or your machine would abrubtly
2705 and completely crash). This is better than no consistency at all
2706 and a good method for some use cases, but likely not good enough if
2707 you have databases running.
2708 Then you create a backup archive of all these snapshots. The backup
2710 process will see a "frozen" state of the logical volumes, while the
2711 processes working in the original volumes continue changing the data
2712 stored there.
2713 You also add the output of lvdisplay to your backup, so you can see the
2715 LV sizes in case you ever need to recreate and restore them.
2716 After the backup has completed, you remove the snapshots again.
2718 $ # create snapshots here
2720 $ lvdisplay > lvdisplay.txt
2721 $ borg create --read-special /path/to/repo::arch lvdisplay.txt /dev/vg0/*-snapshot
2722 $ # remove snapshots here
2723 Now, let's see how to restore some LVs from such a backup.
2725 $ borg extract /path/to/repo::arch lvdisplay.txt
2727 $ # create empty LVs with correct sizes here (look into lvdisplay.txt).
2728 $ # we assume that you created an empty root and home LV and overwrite it now:
2729 $ borg extract --stdout /path/to/repo::arch dev/vg0/root-snapshot > /dev/vg0/root
2730 $ borg extract --stdout /path/to/repo::arch dev/vg0/home-snapshot > /dev/vg0/home
2731 Append-only mode
2733 A repository can be made "append-only", which means that Borg will
2734 never overwrite or delete committed data (append-only refers to the
2735 segment files, but borg will also reject to delete the repository com‐
2736 pletely). This is useful for scenarios where a backup client machine
2737 backups remotely to a backup server using borg serve, since a hacked
2738 client machine cannot delete backups on the server permanently.
2739 To activate append-only mode, set append_only to 1 in the repository
2741 config:
2742 borg config /path/to/repo append_only 1
2744 Note that you can go back-and-forth between normal and append-only
2746 operation with borg config; it's not a "one way trip."
2747 In append-only mode Borg will create a transaction log in the transac‐
2749 tions file, where each line is a transaction and a UTC timestamp.
2750 In addition, borg serve can act as if a repository is in append-only
2752 mode with its option --append-only. This can be very useful for
2753 fine-tuning access control in .ssh/authorized_keys
2754 command="borg serve --append-only ..." ssh-rsa <key used for not-always-trustable backup clients>
2756 command="borg serve ..." ssh-rsa <key used for backup management>
2757 Running borg init via a borg serve --append-only server will not create
2759 an append-only repository. Running borg init --append-only creates an
2760 append-only repository regardless of server settings.
2761 Example
2763 Suppose an attacker remotely deleted all backups, but your repository
2764 was in append-only mode. A transaction log in this situation might look
2765 like this:
2766 transaction 1, UTC time 2016-03-31T15:53:27.383532
2768 transaction 5, UTC time 2016-03-31T15:53:52.588922
2769 transaction 11, UTC time 2016-03-31T15:54:23.887256
2770 transaction 12, UTC time 2016-03-31T15:55:54.022540
2771 transaction 13, UTC time 2016-03-31T15:55:55.472564
2772 From your security logs you conclude the attacker gained access at
2774 15:54:00 and all the backups where deleted or replaced by compromised
2775 backups. From the log you know that transactions 11 and later are com‐
2776 promised. Note that the transaction ID is the name of the last file in
2777 the transaction. For example, transaction 11 spans files 6 to 11.
2778 In a real attack you'll likely want to keep the compromised repository
2780 intact to analyze what the attacker tried to achieve. It's also a good
2781 idea to make this copy just in case something goes wrong during the
2782 recovery. Since recovery is done by deleting some files, a hard link
2783 copy (cp -al) is sufficient.
2784 The first step to reset the repository to transaction 5, the last
2786 uncompromised transaction, is to remove the hints.N, index.N and
2787 ``
2788 integrity.N``files in the repository (these files are always expend‐
2789 able). In this example N is 13.
2790 Then remove or move all segment files from the segment directories in
2792 data/ starting with file 6:
2793 rm data/**/{6..13}
2795 That's all to do in the repository.
2797 If you want to access this rollbacked repository from a client that
2799 already has a cache for this repository, the cache will reflect a newer
2800 repository state than what you actually have in the repository now,
2801 after the rollback.
2802 Thus, you need to clear the cache:
2804 borg delete --cache-only repo
2806 The cache will get rebuilt automatically. Depending on repo size and
2808 archive count, it may take a while.
2809 You also will need to remove ~/.config/borg/security/REPOID/mani‐
2811 fest-timestamp.
2812 Drawbacks
2814 As data is only appended, and nothing removed, commands like prune or
2815 delete won't free disk space, they merely tag data as deleted in a new
2816 transaction.
2817 Be aware that as soon as you write to the repo in non-append-only mode
2819 (e.g. prune, delete or create archives from an admin machine), it will
2820 remove the deleted objects permanently (including the ones that were
2821 already marked as deleted, but not removed, in append-only mode). Auto‐
2822 mated edits to the repository (such as a cron job running borg prune)
2823 will render append-only mode moot if data is deleted.
2824 Even if an archive appears to be available, it is possible an attacker
2826 could delete just a few chunks from an archive and silently corrupt its
2827 data. While in append-only mode, this is reversible, but borg check
2828 should be run before a writing/pruning operation on an append-only
2829 repository to catch accidental or malicious corruption:
2830 # run without append-only mode
2832 borg check --verify-data repo && borg compact repo
2833 Aside from checking repository & archive integrity you may want to also
2835 manually check backups to ensure their content seems correct.
2836 Further considerations
2838 Append-only mode is not respected by tools other than Borg. rm still
2839 works on the repository. Make sure that backup client machines only get
2840 to access the repository via borg serve.
2841 Ensure that no remote access is possible if the repository is temporar‐
2843 ily set to normal mode for e.g. regular pruning.
2844 Further protections can be implemented, but are outside of Borg's
2846 scope. For example, file system snapshots or wrapping borg serve to set
2847 special permissions or ACLs on new data files.
2848 SSH batch mode
2850 When running Borg using an automated script, ssh might still ask for a
2851 password, even if there is an SSH key for the target server. Use this
2852 to make scripts more robust:
2853 export BORG_RSH='ssh -oBatchMode=yes'
2855
2857 The Borg Collective (see AUTHORS file)
2858
2860 2010-2019 Jonas Borgström, 2015-2019 The Borg Collective (see AUTHORS
2861 file)
28621.1.10 2019-07-24 BORG(1)