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 https://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_SUFFIX
263 When set to a value at least one character long,
264 instructs borg to use a specifically named (based on the
265 suffix) alternative files cache. This can be used to
266 avoid loading and saving cache entries for backup sources
267 other than the current sources.
268 BORG_FILES_CACHE_TTL
270 When set to a numeric value, this determines the maximum
271 "time to live" for the files cache entries (default: 20).
272 The files cache is used to quickly determine whether a
273 file is unchanged. The FAQ explains this more detailed
274 in: always_chunking
275 BORG_SHOW_SYSINFO
277 When set to no (default: yes), system information (like
278 OS, Python version, ...) in exceptions is not shown.
279 Please only use for good reasons as it makes issues
280 harder to analyze.
281 BORG_WORKAROUNDS
283 A list of comma separated strings that trigger work‐
284 arounds in borg, e.g. to work around bugs in other soft‐
285 ware.
286 Currently known strings are:
288 basesyncfile
290 Use the more simple BaseSyncFile code to avoid
291 issues with sync_file_range. You might need this
292 to run borg on WSL (Windows Subsystem for Linux)
293 or in systemd.nspawn containers on some architec‐
294 tures (e.g. ARM). Using this does not affect data
295 safety, but might result in a more bursty write to
296 disk behaviour (not continuously streaming to
297 disk).
298 Some automatic answerers (if set, they automatically answer confirma‐
300 tion questions):
301 BORG_UNKNOWN_UNENCRYPTED_REPO_ACCESS_IS_OK=no (or =yes)
303 For "Warning: Attempting to access a previously unknown
304 unencrypted repository"
305 BORG_RELOCATED_REPO_ACCESS_IS_OK=no (or =yes)
307 For "Warning: The repository at location ... was previ‐
308 ously located at ..."
309 BORG_CHECK_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
311 For "This is a potentially dangerous function..." (check
312 --repair)
313 BORG_DELETE_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
315 For "You requested to completely DELETE the repository
316 including all archives it contains:"
317 Note: answers are case sensitive. setting an invalid answer
319 value might either give the default answer or ask you interac‐
320 tively, depending on whether retries are allowed (they by
321 default are allowed). So please test your scripts interactively
322 before making them a non-interactive script.
323 Directories and files:
325 BORG_BASE_DIR
327 Defaults to $HOME or ~$USER or ~ (in that order). If you
328 want to move all borg-specific folders to a custom path
329 at once, all you need to do is to modify BORG_BASE_DIR:
330 the other paths for cache, config etc. will adapt accord‐
331 ingly (assuming you didn't set them to a different custom
332 value).
333 BORG_CACHE_DIR
335 Defaults to $BORG_BASE_DIR/.cache/borg. If BORG_BASE_DIR
336 is not explicitly set while XDG env var XDG_CACHE_HOME is
337 set, then $XDG_CACHE_HOME/borg is being used instead.
338 This directory contains the local cache and might need a
339 lot of space for dealing with big repositories. Make sure
340 you're aware of the associated security aspects of the
341 cache location: cache_security
342 BORG_CONFIG_DIR
344 Defaults to $BORG_BASE_DIR/.config/borg. If BORG_BASE_DIR
345 is not explicitly set while XDG env var XDG_CONFIG_HOME
346 is set, then $XDG_CONFIG_HOME/borg is being used instead.
347 This directory contains all borg configuration directo‐
348 ries, see the FAQ for a security advisory about the data
349 in this directory: home_config_borg
350 BORG_SECURITY_DIR
352 Defaults to $BORG_CONFIG_DIR/security. This directory
353 contains information borg uses to track its usage of
354 NONCES ("numbers used once" - usually in encryption con‐
355 text) and other security relevant data.
356 BORG_KEYS_DIR
358 Defaults to $BORG_CONFIG_DIR/keys. This directory con‐
359 tains keys for encrypted repositories.
360 BORG_KEY_FILE
362 When set, use the given filename as repository key file.
363 TMPDIR This is where temporary files are stored (might need a
365 lot of temporary space for some operations), see tempfile
366 for details.
367 Building:
369 BORG_OPENSSL_PREFIX
371 Adds given OpenSSL header file directory to the default
372 locations (setup.py).
373 BORG_LIBLZ4_PREFIX
375 Adds given prefix directory to the default locations. If
376 a 'include/lz4.h' is found Borg will be linked against
377 the system liblz4 instead of a bundled implementation.
378 (setup.py)
379 BORG_LIBB2_PREFIX
381 Adds given prefix directory to the default locations. If
382 a 'include/blake2.h' is found Borg will be linked against
383 the system libb2 instead of a bundled implementation.
384 (setup.py)
385 BORG_LIBZSTD_PREFIX
387 Adds given prefix directory to the default locations. If
388 a 'include/zstd.h' is found Borg will be linked against
389 the system libzstd instead of a bundled implementation.
390 (setup.py)
391 Please note:
393 · Be very careful when using the "yes" sayers, the warnings with prompt
395 exist for your / your data's security/safety.
396 · Also be very careful when putting your passphrase into a script, make
398 sure it has appropriate file permissions (e.g. mode 600, root:root).
399 File systems
401 We strongly recommend against using Borg (or any other database-like
402 software) on non-journaling file systems like FAT, since it is not pos‐
403 sible to assume any consistency in case of power failures (or a sudden
404 disconnect of an external drive or similar failures).
405 While Borg uses a data store that is resilient against these failures
407 when used on journaling file systems, it is not possible to guarantee
408 this with some hardware -- independent of the software used. We don't
409 know a list of affected hardware.
410 If you are suspicious whether your Borg repository is still consistent
412 and readable after one of the failures mentioned above occurred, run
413 borg check --verify-data to make sure it is consistent.
414 Requirements for Borg repository file systems
416 · Long file names
418 · At least three directory levels with short names
420 · Typically, file sizes up to a few hundred MB. Large repositories may
422 require large files (>2 GB).
423 · Up to 1000 files per directory (10000 for repositories initialized
425 with Borg 1.0)
426 · mkdir(2) should be atomic, since it is used for locking
428 · Hardlinks are needed for borg_upgrade (if --inplace option is not
430 used). Also hardlinks are used for more safe and secure file updat‐
431 ing (e.g. of the repo config file), but the code tries to work also
432 if hardlinks are not supported.
433 Units
435 To display quantities, Borg takes care of respecting the usual conven‐
436 tions of scale. Disk sizes are displayed in decimal, using powers of
437 ten (so kB means 1000 bytes). For memory usage, binary prefixes are
438 used, and are indicated using the IEC binary prefixes, using powers of
439 two (so KiB means 1024 bytes).
440 Date and Time
442 We format date and time conforming to ISO-8601, that is: YYYY-MM-DD and
443 HH:MM:SS (24h clock).
444 For more information about that, see: https://xkcd.com/1179/
446 Unless otherwise noted, we display local date and time. Internally, we
448 store and process date and time as UTC.
449 Resource Usage
451 Borg might use a lot of resources depending on the size of the data set
452 it is dealing with.
453 If one uses Borg in a client/server way (with a ssh: repository), the
455 resource usage occurs in part on the client and in another part on the
456 server.
457 If one uses Borg as a single process (with a filesystem repo), all the
459 resource usage occurs in that one process, so just add up client +
460 server to get the approximate resource usage.
461 CPU client:
463 · borg create: does chunking, hashing, compression, crypto (high
465 CPU usage)
466 · chunks cache sync: quite heavy on CPU, doing lots of hashtable
468 operations.
469 · borg extract: crypto, decompression (medium to high CPU usage)
471 · borg check: similar to extract, but depends on options given.
473 · borg prune / borg delete archive: low to medium CPU usage
475 · borg delete repo: done on the server
477 It won't go beyond 100% of 1 core as the code is currently sin‐
479 gle-threaded. Especially higher zlib and lzma compression lev‐
480 els use significant amounts of CPU cycles. Crypto might be cheap
481 on the CPU (if hardware accelerated) or expensive (if not).
482 CPU server:
484 It usually doesn't need much CPU, it just deals with the
485 key/value store (repository) and uses the repository index for
486 that.
487 borg check: the repository check computes the checksums of all
489 chunks (medium CPU usage) borg delete repo: low CPU usage
490 CPU (only for client/server operation):
492 When using borg in a client/server way with a ssh:-type repo,
493 the ssh processes used for the transport layer will need some
494 CPU on the client and on the server due to the crypto they are
495 doing - esp. if you are pumping big amounts of data.
496 Memory (RAM) client:
498 The chunks index and the files index are read into memory for
499 performance reasons. Might need big amounts of memory (see
500 below). Compression, esp. lzma compression with high levels
501 might need substantial amounts of memory.
502 Memory (RAM) server:
504 The server process will load the repository index into memory.
505 Might need considerable amounts of memory, but less than on the
506 client (see below).
507 Chunks index (client only):
509 Proportional to the amount of data chunks in your repo. Lots of
510 chunks in your repo imply a big chunks index. It is possible to
511 tweak the chunker params (see create options).
512 Files index (client only):
514 Proportional to the amount of files in your last backups. Can be
515 switched off (see create options), but next backup might be much
516 slower if you do. The speed benefit of using the files cache is
517 proportional to file size.
518 Repository index (server only):
520 Proportional to the amount of data chunks in your repo. Lots of
521 chunks in your repo imply a big repository index. It is possi‐
522 ble to tweak the chunker params (see create options) to influ‐
523 ence the amount of chunks being created.
524 Temporary files (client):
526 Reading data and metadata from a FUSE mounted repository will
527 consume up to the size of all deduplicated, small chunks in the
528 repository. Big chunks won't be locally cached.
529 Temporary files (server):
531 A non-trivial amount of data will be stored on the remote temp
532 directory for each client that connects to it. For some remotes,
533 this can fill the default temporary directory at /tmp. This can
534 be remediated by ensuring the $TMPDIR, $TEMP, or $TMP environ‐
535 ment variable is properly set for the sshd process. For some
536 OSes, this can be done just by setting the correct value in the
537 .bashrc (or equivalent login config file for other shells), how‐
538 ever in other cases it may be necessary to first enable Permi‐
539 tUserEnvironment yes in your sshd_config file, then add environ‐
540 ment="TMPDIR=/my/big/tmpdir" at the start of the public key to
541 be used in the authorized_hosts file.
542 Cache files (client only):
544 Contains the chunks index and files index (plus a collection of
545 single- archive chunk indexes which might need huge amounts of
546 disk space, depending on archive count and size - see FAQ about
547 how to reduce).
548 Network (only for client/server operation):
550 If your repository is remote, all deduplicated (and optionally
551 compressed/ encrypted) data of course has to go over the connec‐
552 tion (ssh:// repo url). If you use a locally mounted network
553 filesystem, additionally some copy operations used for transac‐
554 tion support also go over the connection. If you backup multiple
555 sources to one target repository, additional traffic happens for
556 cache resynchronization.
557 Support for file metadata
559 Besides regular file and directory structures, Borg can preserve
560 · symlinks (stored as symlink, the symlink is not followed)
562 · special files:
564 · character and block device files (restored via mknod)
566 · FIFOs ("named pipes")
568 · special file contents can be backed up in --read-special mode. By
570 default the metadata to create them with mknod(2), mkfifo(2) etc.
571 is stored.
572 · hardlinked regular files, devices, FIFOs (considering all items in
574 the same archive)
575 · timestamps in nanosecond precision: mtime, atime, ctime
577 · other timestamps: birthtime (on platforms supporting it)
579 · permissions:
581 · IDs of owning user and owning group
583 · names of owning user and owning group (if the IDs can be resolved)
585 · Unix Mode/Permissions (u/g/o permissions, suid, sgid, sticky)
587 On some platforms additional features are supported:
589 ┌─────────────────┬──────────┬───────────┬───────────┐
591 │Platform │ ACLs [5] │ xattr [6] │ Flags [7] │
592 ├─────────────────┼──────────┼───────────┼───────────┤
593 │Linux │ Yes │ Yes │ Yes [1] │
594 ├─────────────────┼──────────┼───────────┼───────────┤
595 │Mac OS X │ Yes │ Yes │ Yes (all) │
596 ├─────────────────┼──────────┼───────────┼───────────┤
597 │FreeBSD │ Yes │ Yes │ Yes (all) │
598 ├─────────────────┼──────────┼───────────┼───────────┤
599 │OpenBSD │ n/a │ n/a │ Yes (all) │
600 ├─────────────────┼──────────┼───────────┼───────────┤
601 │NetBSD │ n/a │ No [2] │ Yes (all) │
602 └─────────────────┴──────────┴───────────┴───────────┘
603 │Solaris and de‐ │ No [3] │ No [3] │ n/a │
605 │rivatives │ │ │ │
606 ├─────────────────┼──────────┼───────────┼───────────┤
607 │Windows (cygwin) │ No [4] │ No │ No │
608 └─────────────────┴──────────┴───────────┴───────────┘
609 Other Unix-like operating systems may work as well, but have not been
611 tested at all.
612 Note that most of the platform-dependent features also depend on the
614 file system. For example, ntfs-3g on Linux isn't able to convey NTFS
615 ACLs.
616 [1] Only "nodump", "immutable", "compressed" and "append" are sup‐
618 ported. Feature request #618 for more flags.
619 [2] Feature request #1332
621 [3] Feature request #1337
623 [4] Cygwin tries to map NTFS ACLs to permissions with varying degrees
625 of success.
626 [5] The native access control list mechanism of the OS. This normally
628 limits access to non-native ACLs. For example, NTFS ACLs aren't
629 completely accessible on Linux with ntfs-3g.
630 [6] extended attributes; key-value pairs attached to a file, mainly
632 used by the OS. This includes resource forks on Mac OS X.
633 [7] aka BSD flags. The Linux set of flags [1] is portable across plat‐
635 forms. The BSDs define additional flags.
636 In case you are interested in more details (like formulas), please
638 see internals. For details on the available JSON output, refer to
639 json_output.
640 Common options
642 All Borg commands share these options:
643 -h, --help
645 show this help message and exit
646 --critical
648 work on log level CRITICAL
649 --error
651 work on log level ERROR
652 --warning
654 work on log level WARNING (default)
655 --info, -v, --verbose
657 work on log level INFO
658 --debug
660 enable debug output, work on log level DEBUG
661 --debug-topic TOPIC
663 enable TOPIC debugging (can be specified multiple times). The
664 logger path is borg.debug.<TOPIC> if TOPIC is not fully quali‐
665 fied.
666 -p, --progress
668 show progress information
669 --log-json
671 Output one JSON object per log line instead of formatted text.
672 --lock-wait SECONDS
674 wait at most SECONDS for acquiring a repository/cache lock
675 (default: 1).
676 --bypass-lock
678 Bypass locking mechanism
679 --show-version
681 show/log the borg version
682 --show-rc
684 show/log the return code (rc)
685 --umask M
687 set umask to M (local and remote, default: 0077)
688 --remote-path PATH
690 use PATH as borg executable on the remote (default: "borg")
691 --remote-ratelimit RATE
693 set remote network upload rate limit in kiByte/s (default:
694 0=unlimited)
695 --consider-part-files
697 treat part files like normal files (e.g. to list/extract them)
698 --debug-profile FILE
700 Write execution profile in Borg format into FILE. For local use
701 a Python-compatible file can be generated by suffixing FILE with
702 ".pyprof".
703 --rsh RSH
705 Use this command to connect to the 'borg serve' process
706 (default: 'ssh')
707 Option --bypass-lock allows you to access the repository while bypass‐
709 ing borg's locking mechanism. This is necessary if your repository is
710 on a read-only storage where you don't have write permissions or capa‐
711 bilities and therefore cannot create a lock. Examples are repositories
712 stored on a Bluray disc or a read-only network storage. Avoid this
713 option if you are able to use locks as that is the safer way; see the
714 warning below.
715 WARNING:
717 If you do use --bypass-lock, you are responsible to ensure that no
718 other borg instances have write access to the repository. Otherwise,
719 you might experience errors and read broken data if changes to that
720 repository are being made at the same time.
721 Examples
723 # Create an archive and log: borg version, files list, return code
724 $ borg create --show-version --list --show-rc /path/to/repo::my-files files
725
727 borg [common options] init [options] [REPOSITORY]
728 Description
730 This command initializes an empty repository. A repository is a
731 filesystem directory containing the deduplicated data from zero or more
732 archives.
733 Encryption can be enabled at repository init time. It cannot be changed
735 later.
736 It is not recommended to work without encryption. Repository encryption
738 protects you e.g. against the case that an attacker has access to your
739 backup repository.
740 Borg relies on randomly generated key material and uses that for chunk‐
742 ing, id generation, encryption and authentication. The key material is
743 encrypted using the passphrase you give before it is stored on-disk.
744 You need to be careful with the key / the passphrase:
746 If you want "passphrase-only" security, use one of the repokey modes.
748 The key will be stored inside the repository (in its "config" file). In
749 above mentioned attack scenario, the attacker will have the key (but
750 not the passphrase).
751 If you want "passphrase and having-the-key" security, use one of the
753 keyfile modes. The key will be stored in your home directory (in .con‐
754 fig/borg/keys). In the attack scenario, the attacker who has just
755 access to your repo won't have the key (and also not the passphrase).
756 Make a backup copy of the key file (keyfile mode) or repo config file
758 (repokey mode) and keep it at a safe place, so you still have the key
759 in case it gets corrupted or lost. Also keep the passphrase at a safe
760 place. The backup that is encrypted with that key won't help you with
761 that, of course.
762 Make sure you use a good passphrase. Not too short, not too simple. The
764 real encryption / decryption key is encrypted with / locked by your
765 passphrase. If an attacker gets your key, he can't unlock and use it
766 without knowing the passphrase.
767 Be careful with special or non-ascii characters in your passphrase:
769 · Borg processes the passphrase as unicode (and encodes it as utf-8),
771 so it does not have problems dealing with even the strangest charac‐
772 ters.
773 · BUT: that does not necessarily apply to your OS / VM / keyboard con‐
775 figuration.
776 So better use a long passphrase made from simple ascii chars than one
778 that includes non-ascii stuff or characters that are hard/impossible to
779 enter on a different keyboard layout.
780 You can change your passphrase for existing repos at any time, it won't
782 affect the encryption/decryption key or other secrets.
783 Encryption modes
785 You can choose from the encryption modes seen in the table below on a
786 per-repo basis. The mode determines encryption algorithm, hash/MAC
787 algorithm and also the key storage location.
788 Example: borg init --encryption repokey ...
790 ┌─────────┬──────────────────┬──────────────────┬──────────────────┐
792 │Hash/MAC │ Not encrypted no │ Not encrypted, │ Encrypted (AEAD │
793 │ │ auth │ but authenti‐ │ w/ AES) and │
794 │ │ │ cated │ authenticated │
795 ├─────────┼──────────────────┼──────────────────┼──────────────────┤
796 │SHA-256 │ none │ authenticated │ repokey keyfile │
797 ├─────────┼──────────────────┼──────────────────┼──────────────────┤
798 │BLAKE2b │ n/a │ authenti‐ │ repokey-blake2 │
799 │ │ │ cated-blake2 │ keyfile-blake2 │
800 └─────────┴──────────────────┴──────────────────┴──────────────────┘
801 Modes marked like this in the above table are new in Borg 1.1 and are
803 not backwards-compatible with Borg 1.0.x.
804 On modern Intel/AMD CPUs (except very cheap ones), AES is usually hard‐
806 ware-accelerated. BLAKE2b is faster than SHA256 on Intel/AMD 64-bit
807 CPUs (except AMD Ryzen and future CPUs with SHA extensions), which
808 makes authenticated-blake2 faster than none and authenticated.
809 On modern ARM CPUs, NEON provides hardware acceleration for SHA256 mak‐
811 ing it faster than BLAKE2b-256 there. NEON accelerates AES as well.
812 Hardware acceleration is always used automatically when available.
814 repokey and keyfile use AES-CTR-256 for encryption and HMAC-SHA256 for
816 authentication in an encrypt-then-MAC (EtM) construction. The chunk ID
817 hash is HMAC-SHA256 as well (with a separate key). These modes are
818 compatible with Borg 1.0.x.
819 repokey-blake2 and keyfile-blake2 are also authenticated encryption
821 modes, but use BLAKE2b-256 instead of HMAC-SHA256 for authentication.
822 The chunk ID hash is a keyed BLAKE2b-256 hash. These modes are new and
823 not compatible with Borg 1.0.x.
824 authenticated mode uses no encryption, but authenticates repository
826 contents through the same HMAC-SHA256 hash as the repokey and keyfile
827 modes (it uses it as the chunk ID hash). The key is stored like
828 repokey. This mode is new and not compatible with Borg 1.0.x.
829 authenticated-blake2 is like authenticated, but uses the keyed
831 BLAKE2b-256 hash from the other blake2 modes. This mode is new and not
832 compatible with Borg 1.0.x.
833 none mode uses no encryption and no authentication. It uses SHA256 as
835 chunk ID hash. This mode is not recommended, you should rather consider
836 using an authenticated or authenticated/encrypted mode. This mode has
837 possible denial-of-service issues when running borg create on contents
838 controlled by an attacker. Use it only for new repositories where no
839 encryption is wanted and when compatibility with 1.0.x is important. If
840 compatibility with 1.0.x is not important, use authenticated-blake2 or
841 authenticated instead. This mode is compatible with Borg 1.0.x.
842 Examples
844 # Local repository, repokey encryption, BLAKE2b (often faster, since Borg 1.1)
845 $ borg init --encryption=repokey-blake2 /path/to/repo
846 # Local repository (no encryption)
848 $ borg init --encryption=none /path/to/repo
849 # Remote repository (accesses a remote borg via ssh)
851 # repokey: stores the (encrypted) key into <REPO_DIR>/config
852 $ borg init --encryption=repokey-blake2 user@hostname:backup
853 # Remote repository (accesses a remote borg via ssh)
855 # keyfile: stores the (encrypted) key into ~/.config/borg/keys/
856 $ borg init --encryption=keyfile user@hostname:backup
857
859 borg [common options] create [options] ARCHIVE [PATH...]
860 Description
862 This command creates a backup archive containing all files found while
863 recursively traversing all paths specified. Paths are added to the ar‐
864 chive as they are given, that means if relative paths are desired, the
865 command has to be run from the correct directory.
866 When giving '-' as path, borg will read data from standard input and
868 create a file 'stdin' in the created archive from that data. See sec‐
869 tion Reading from stdin below for details.
870 The archive will consume almost no disk space for files or parts of
872 files that have already been stored in other archives.
873 The archive name needs to be unique. It must not end in '.checkpoint'
875 or '.checkpoint.N' (with N being a number), because these names are
876 used for checkpoints and treated in special ways.
877 In the archive name, you may use the following placeholders: {now},
879 {utcnow}, {fqdn}, {hostname}, {user} and some others.
880 Backup speed is increased by not reprocessing files that are already
882 part of existing archives and weren't modified. The detection of unmod‐
883 ified files is done by comparing multiple file metadata values with
884 previous values kept in the files cache.
885 This comparison can operate in different modes as given by
887 --files-cache:
888 · ctime,size,inode (default)
890 · mtime,size,inode (default behaviour of borg versions older than
892 1.1.0rc4)
893 · ctime,size (ignore the inode number)
895 · mtime,size (ignore the inode number)
897 · rechunk,ctime (all files are considered modified - rechunk, cache
899 ctime)
900 · rechunk,mtime (all files are considered modified - rechunk, cache
902 mtime)
903 · disabled (disable the files cache, all files considered modified -
905 rechunk)
906 inode number: better safety, but often unstable on network filesystems
908 Normally, detecting file modifications will take inode information into
910 consideration to improve the reliability of file change detection.
911 This is problematic for files located on sshfs and similar network file
912 systems which do not provide stable inode numbers, such files will
913 always be considered modified. You can use modes without inode in this
914 case to improve performance, but reliability of change detection might
915 be reduced.
916 ctime vs. mtime: safety vs. speed
918 · ctime is a rather safe way to detect changes to a file (metadata and
920 contents) as it can not be set from userspace. But, a metadata-only
921 change will already update the ctime, so there might be some unneces‐
922 sary chunking/hashing even without content changes. Some filesystems
923 do not support ctime (change time). E.g. doing a chown or chmod to a
924 file will change its ctime.
925 · mtime usually works and only updates if file contents were changed.
927 But mtime can be arbitrarily set from userspace, e.g. to set mtime
928 back to the same value it had before a content change happened. This
929 can be used maliciously as well as well-meant, but in both cases
930 mtime based cache modes can be problematic.
931 The mount points of filesystems or filesystem snapshots should be the
933 same for every creation of a new archive to ensure fast operation. This
934 is because the file cache that is used to determine changed files
935 quickly uses absolute filenames. If this is not possible, consider
936 creating a bind mount to a stable location.
937 The --progress option shows (from left to right) Original, Compressed
939 and Deduplicated (O, C and D, respectively), then the Number of files
940 (N) processed so far, followed by the currently processed path.
941 When using --stats, you will get some statistics about how much data
943 was added - the "This Archive" deduplicated size there is most inter‐
944 esting as that is how much your repository will grow. Please note that
945 the "All archives" stats refer to the state after creation. Also, the
946 --stats and --dry-run options are mutually exclusive because the data
947 is not actually compressed and deduplicated during a dry run.
948 See the output of the "borg help patterns" command for more help on
950 exclude patterns.
951 See the output of the "borg help placeholders" command for more help on
953 placeholders.
954 The --exclude patterns are not like tar. In tar --exclude .bundler/gems
956 will exclude foo/.bundler/gems. In borg it will not, you need to use
957 --exclude '*/.bundler/gems' to get the same effect. See borg help pat‐
958 terns for more information.
959 In addition to using --exclude patterns, it is possible to use
961 --exclude-if-present to specify the name of a filesystem object (e.g. a
962 file or folder name) which, when contained within another folder, will
963 prevent the containing folder from being backed up. By default, the
964 containing folder and all of its contents will be omitted from the
965 backup. If, however, you wish to only include the objects specified by
966 --exclude-if-present in your backup, and not include any other contents
967 of the containing folder, this can be enabled through using the
968 --keep-exclude-tags option.
969 The -x or --one-file-system option excludes directories, that are
971 mountpoints (and everything in them). It detects mountpoints by com‐
972 paring the device number from the output of stat() of the directory and
973 its parent directory. Specifically, it excludes directories for which
974 stat() reports a device number different from the device number of
975 their parent. Be aware that in Linux (and possibly elsewhere) there are
976 directories with device number different from their parent, which the
977 kernel does not consider a mountpoint and also the other way around.
978 Examples are bind mounts (possibly same device number, but always a
979 mountpoint) and ALL subvolumes of a btrfs (different device number from
980 parent but not necessarily a mountpoint). Therefore when using
981 --one-file-system, one should make doubly sure that the backup works as
982 intended especially when using btrfs. This is even more important, if
983 the btrfs layout was created by someone else, e.g. a distribution in‐
984 staller.
985 Item flags
987 --list outputs a list of all files, directories and other file system
988 items it considered (no matter whether they had content changes or
989 not). For each item, it prefixes a single-letter flag that indicates
990 type and/or status of the item.
991 If you are interested only in a subset of that output, you can give
993 e.g. --filter=AME and it will only show regular files with A, M or E
994 status (see below).
995 A uppercase character represents the status of a regular file relative
997 to the "files" cache (not relative to the repo -- this is an issue if
998 the files cache is not used). Metadata is stored in any case and for
999 'A' and 'M' also new data chunks are stored. For 'U' all data chunks
1000 refer to already existing chunks.
1001 · 'A' = regular file, added (see also a_status_oddity in the FAQ)
1003 · 'M' = regular file, modified
1005 · 'U' = regular file, unchanged
1007 · 'E' = regular file, an error happened while accessing/reading this
1009 file
1010 A lowercase character means a file type other than a regular file, borg
1012 usually just stores their metadata:
1013 · 'd' = directory
1015 · 'b' = block device
1017 · 'c' = char device
1019 · 'h' = regular file, hardlink (to already seen inodes)
1021 · 's' = symlink
1023 · 'f' = fifo
1025 Other flags used include:
1027 · 'i' = backup data was read from standard input (stdin)
1029 · '-' = dry run, item was not backed up
1031 · 'x' = excluded, item was not backed up
1033 · '?' = missing status code (if you see this, please file a bug
1035 report!)
1036 Reading from stdin
1038 To read from stdin, specify - as path and pipe directly to borg:
1039 backup-vm --id myvm --stdout | borg create REPO::ARCHIVE -
1041 Note that piping to borg creates an archive even if the command piping
1043 to borg exits with a failure. In this case, one can end up with trun‐
1044 cated output being backed up.
1045 Reading from stdin yields just a stream of data without file metadata
1047 associated with it, and the files cache is not needed at all. So it is
1048 safe to disable it via --no-files-cache and speed up backup creation a
1049 bit.
1050 By default, the content read from stdin is stored in a file called
1052 'stdin'. Use --stdin-name to change the name.
1053 Examples
1055 # Backup ~/Documents into an archive named "my-documents"
1056 $ borg create /path/to/repo::my-documents ~/Documents
1057 # same, but list all files as we process them
1059 $ borg create --list /path/to/repo::my-documents ~/Documents
1060 # Backup ~/Documents and ~/src but exclude pyc files
1062 $ borg create /path/to/repo::my-files \
1063 ~/Documents \
1064 ~/src \
1065 --exclude '*.pyc'
1066 # Backup home directories excluding image thumbnails (i.e. only
1068 # /home/<one directory>/.thumbnails is excluded, not /home/*/*/.thumbnails etc.)
1069 $ borg create /path/to/repo::my-files /home \
1070 --exclude 'sh:/home/*/.thumbnails'
1071 # Backup the root filesystem into an archive named "root-YYYY-MM-DD"
1073 # use zlib compression (good, but slow) - default is lz4 (fast, low compression ratio)
1074 $ borg create -C zlib,6 --one-file-system /path/to/repo::root-{now:%Y-%m-%d} /
1075 # Backup a remote host locally ("pull" style) using sshfs
1077 $ mkdir sshfs-mount
1078 $ sshfs root@example.com:/ sshfs-mount
1079 $ cd sshfs-mount
1080 $ borg create /path/to/repo::example.com-root-{now:%Y-%m-%d} .
1081 $ cd ..
1082 $ fusermount -u sshfs-mount
1083 # Make a big effort in fine granular deduplication (big chunk management
1085 # overhead, needs a lot of RAM and disk space, see formula in internals
1086 # docs - same parameters as borg < 1.0 or attic):
1087 $ borg create --chunker-params 10,23,16,4095 /path/to/repo::small /smallstuff
1088 # Backup a raw device (must not be active/in use/mounted at that time)
1090 $ dd if=/dev/sdx bs=10M | borg create /path/to/repo::my-sdx -
1091 # No compression (none)
1093 $ borg create --compression none /path/to/repo::arch ~
1094 # Super fast, low compression (lz4, default)
1096 $ borg create /path/to/repo::arch ~
1097 # Less fast, higher compression (zlib, N = 0..9)
1099 $ borg create --compression zlib,N /path/to/repo::arch ~
1100 # Even slower, even higher compression (lzma, N = 0..9)
1102 $ borg create --compression lzma,N /path/to/repo::arch ~
1103 # Only compress compressible data with lzma,N (N = 0..9)
1105 $ borg create --compression auto,lzma,N /path/to/repo::arch ~
1106 # Use short hostname, user name and current time in archive name
1108 $ borg create /path/to/repo::{hostname}-{user}-{now} ~
1109 # Similar, use the same datetime format that is default as of borg 1.1
1110 $ borg create /path/to/repo::{hostname}-{user}-{now:%Y-%m-%dT%H:%M:%S} ~
1111 # As above, but add nanoseconds
1112 $ borg create /path/to/repo::{hostname}-{user}-{now:%Y-%m-%dT%H:%M:%S.%f} ~
1113 # Backing up relative paths by moving into the correct directory first
1115 $ cd /home/user/Documents
1116 # The root directory of the archive will be "projectA"
1117 $ borg create /path/to/repo::daily-projectA-{now:%Y-%m-%d} projectA
1118
1120 borg [common options] extract [options] ARCHIVE [PATH...]
1121 Description
1123 This command extracts the contents of an archive. By default the entire
1124 archive is extracted but a subset of files and directories can be
1125 selected by passing a list of PATHs as arguments. The file selection
1126 can further be restricted by using the --exclude option.
1127 See the output of the "borg help patterns" command for more help on
1129 exclude patterns.
1130 By using --dry-run, you can do all extraction steps except actually
1132 writing the output data: reading metadata and data chunks from the
1133 repo, checking the hash/hmac, decrypting, decompressing.
1134 --progress can be slower than no progress display, since it makes one
1136 additional pass over the archive metadata.
1137 NOTE:
1139 Currently, extract always writes into the current working directory
1140 ("."), so make sure you cd to the right place before calling borg
1141 extract.
1142 Examples
1144 # Extract entire archive
1145 $ borg extract /path/to/repo::my-files
1146 # Extract entire archive and list files while processing
1148 $ borg extract --list /path/to/repo::my-files
1149 # Verify whether an archive could be successfully extracted, but do not write files to disk
1151 $ borg extract --dry-run /path/to/repo::my-files
1152 # Extract the "src" directory
1154 $ borg extract /path/to/repo::my-files home/USERNAME/src
1155 # Extract the "src" directory but exclude object files
1157 $ borg extract /path/to/repo::my-files home/USERNAME/src --exclude '*.o'
1158 # Restore a raw device (must not be active/in use/mounted at that time)
1160 $ borg extract --stdout /path/to/repo::my-sdx | dd of=/dev/sdx bs=10M
1161
1163 borg [common options] check [options] [REPOSITORY_OR_ARCHIVE]
1164 Description
1166 The check command verifies the consistency of a repository and the cor‐
1167 responding archives.
1168 check --repair is a potentially dangerous function and might lead to
1170 data loss (for kinds of corruption it is not capable of dealing with).
1171 BE VERY CAREFUL!
1172 First, the underlying repository data files are checked:
1174 · For all segments, the segment magic header is checked.
1176 · For all objects stored in the segments, all metadata (e.g. CRC and
1178 size) and all data is read. The read data is checked by size and CRC.
1179 Bit rot and other types of accidental damage can be detected this
1180 way.
1181 · In repair mode, if an integrity error is detected in a segment, try
1183 to recover as many objects from the segment as possible.
1184 · In repair mode, make sure that the index is consistent with the data
1186 stored in the segments.
1187 · If checking a remote repo via ssh:, the repo check is executed on the
1189 server without causing significant network traffic.
1190 · The repository check can be skipped using the --archives-only option.
1192 Second, the consistency and correctness of the archive metadata is ver‐
1194 ified:
1195 · Is the repo manifest present? If not, it is rebuilt from archive
1197 metadata chunks (this requires reading and decrypting of all metadata
1198 and data).
1199 · Check if archive metadata chunk is present; if not, remove archive
1201 from manifest.
1202 · For all files (items) in the archive, for all chunks referenced by
1204 these files, check if chunk is present. In repair mode, if a chunk is
1205 not present, replace it with a same-size replacement chunk of zeroes.
1206 If a previously lost chunk reappears (e.g. via a later backup), in
1207 repair mode the all-zero replacement chunk will be replaced by the
1208 correct chunk. This requires reading of archive and file metadata,
1209 but not data.
1210 · In repair mode, when all the archives were checked, orphaned chunks
1212 are deleted from the repo. One cause of orphaned chunks are input
1213 file related errors (like read errors) in the archive creation
1214 process.
1215 · If checking a remote repo via ssh:, the archive check is executed on
1217 the client machine because it requires decryption, and this is always
1218 done client-side as key access is needed.
1219 · The archive checks can be time consuming; they can be skipped using
1221 the --repository-only option.
1222 The --verify-data option will perform a full integrity verification (as
1224 opposed to checking the CRC32 of the segment) of data, which means
1225 reading the data from the repository, decrypting and decompressing it.
1226 This is a cryptographic verification, which will detect (accidental)
1227 corruption. For encrypted repositories it is tamper-resistant as well,
1228 unless the attacker has access to the keys. It is also very slow.
1229
1231 borg [common options] rename [options] ARCHIVE NEWNAME
1232 Description
1234 This command renames an archive in the repository.
1235 This results in a different archive ID.
1237 Examples
1239 $ borg create /path/to/repo::archivename ~
1240 $ borg list /path/to/repo
1241 archivename Mon, 2016-02-15 19:50:19
1242 $ borg rename /path/to/repo::archivename newname
1244 $ borg list /path/to/repo
1245 newname Mon, 2016-02-15 19:50:19
1246
1248 borg [common options] list [options] [REPOSITORY_OR_ARCHIVE] [PATH...]
1249 Description
1251 This command lists the contents of a repository or an archive.
1252 See the "borg help patterns" command for more help on exclude patterns.
1254 The following keys are available for --format:
1256 · NEWLINE: OS dependent line separator
1258 · NL: alias of NEWLINE
1260 · NUL: NUL character for creating print0 / xargs -0 like output, see
1262 barchive/bpath
1263 · SPACE
1265 · TAB
1267 · CR
1269 · LF
1271 Keys for listing repository archives:
1273 · archive: archive name interpreted as text (might be missing non-text
1275 characters, see barchive)
1276 · name: alias of "archive"
1278 · barchive: verbatim archive name, can contain any character except NUL
1280 · comment: archive comment interpreted as text (might be missing
1282 non-text characters, see bcomment)
1283 · bcomment: verbatim archive comment, can contain any character except
1285 NUL
1286 · id: internal ID of the archive
1288 · start: time (start) of creation of the archive
1290 · time: alias of "start"
1292 · end: time (end) of creation of the archive
1294 · hostname: hostname of host on which this archive was created
1296 · username: username of user who created this archive
1298 Keys for listing archive files:
1300 · type
1302 · mode
1304 · uid
1306 · gid
1308 · user
1310 · group
1312 · path: path interpreted as text (might be missing non-text characters,
1314 see bpath)
1315 · bpath: verbatim POSIX path, can contain any character except NUL
1317 · source: link target for links (identical to linktarget)
1319 · linktarget
1321 · flags
1323 · size
1325 · csize: compressed size
1327 · dsize: deduplicated size
1329 · dcsize: deduplicated compressed size
1331 · num_chunks: number of chunks in this file
1333 · unique_chunks: number of unique chunks in this file
1335 · mtime
1337 · ctime
1339 · atime
1341 · isomtime
1343 · isoctime
1345 · isoatime
1347 · blake2b
1349 · blake2s
1351 · md5
1353 · sha1
1355 · sha224
1357 · sha256
1359 · sha384
1361 · sha3_224
1363 · sha3_256
1365 · sha3_384
1367 · sha3_512
1369 · sha512
1371 · shake_128
1373 · shake_256
1375 · archiveid
1377 · archivename
1379 · extra: prepends {source} with " -> " for soft links and " link to "
1381 for hard links
1382 · health: either "healthy" (file ok) or "broken" (if file has all-zero
1384 replacement chunks)
1385 Examples
1387 $ borg list /path/to/repo
1388 Monday Mon, 2016-02-15 19:15:11
1389 repo Mon, 2016-02-15 19:26:54
1390 root-2016-02-15 Mon, 2016-02-15 19:36:29
1391 newname Mon, 2016-02-15 19:50:19
1392 ...
1393 $ borg list /path/to/repo::root-2016-02-15
1395 drwxr-xr-x root root 0 Mon, 2016-02-15 17:44:27 .
1396 drwxrwxr-x root root 0 Mon, 2016-02-15 19:04:49 bin
1397 -rwxr-xr-x root root 1029624 Thu, 2014-11-13 00:08:51 bin/bash
1398 lrwxrwxrwx root root 0 Fri, 2015-03-27 20:24:26 bin/bzcmp -> bzdiff
1399 -rwxr-xr-x root root 2140 Fri, 2015-03-27 20:24:22 bin/bzdiff
1400 ...
1401 $ borg list /path/to/repo::root-2016-02-15 --pattern "- bin/ba*"
1403 drwxr-xr-x root root 0 Mon, 2016-02-15 17:44:27 .
1404 drwxrwxr-x root root 0 Mon, 2016-02-15 19:04:49 bin
1405 lrwxrwxrwx root root 0 Fri, 2015-03-27 20:24:26 bin/bzcmp -> bzdiff
1406 -rwxr-xr-x root root 2140 Fri, 2015-03-27 20:24:22 bin/bzdiff
1407 ...
1408 $ borg list /path/to/repo::archiveA --format="{mode} {user:6} {group:6} {size:8d} {isomtime} {path}{extra}{NEWLINE}"
1410 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 .
1411 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 code
1412 drwxrwxr-x user user 0 Sun, 2015-02-01 11:00:00 code/myproject
1413 -rw-rw-r-- user user 1416192 Sun, 2015-02-01 11:00:00 code/myproject/file.ext
1414 ...
1415
1417 borg [common options] diff [options] REPO::ARCHIVE1 ARCHIVE2 [PATH...]
1418 Description
1420 This command finds differences (file contents, user/group/mode) between
1421 archives.
1422 A repository location and an archive name must be specified for
1424 REPO::ARCHIVE1. ARCHIVE2 is just another archive name in same reposi‐
1425 tory (no repository location allowed).
1426 For archives created with Borg 1.1 or newer diff automatically detects
1428 whether the archives are created with the same chunker params. If so,
1429 only chunk IDs are compared, which is very fast.
1430 For archives prior to Borg 1.1 chunk contents are compared by default.
1432 If you did not create the archives with different chunker params, pass
1433 --same-chunker-params. Note that the chunker params changed from Borg
1434 0.xx to 1.0.
1435 See the output of the "borg help patterns" command for more help on
1437 exclude patterns.
1438 Examples
1440 $ borg init -e=none testrepo
1441 $ mkdir testdir
1442 $ cd testdir
1443 $ echo asdf > file1
1444 $ dd if=/dev/urandom bs=1M count=4 > file2
1445 $ touch file3
1446 $ borg create ../testrepo::archive1 .
1447 $ chmod a+x file1
1449 $ echo "something" >> file2
1450 $ borg create ../testrepo::archive2 .
1451 $ echo "testing 123" >> file1
1453 $ rm file3
1454 $ touch file4
1455 $ borg create ../testrepo::archive3 .
1456 $ cd ..
1458 $ borg diff testrepo::archive1 archive2
1459 [-rw-r--r-- -> -rwxr-xr-x] file1
1460 +135 B -252 B file2
1461 $ borg diff testrepo::archive2 archive3
1463 +17 B -5 B file1
1464 added 0 B file4
1465 removed 0 B file3
1466 $ borg diff testrepo::archive1 archive3
1468 +17 B -5 B [-rw-r--r-- -> -rwxr-xr-x] file1
1469 +135 B -252 B file2
1470 added 0 B file4
1471 removed 0 B file3
1472 $ borg diff --json-lines testrepo::archive1 archive3
1474 {"path": "file1", "changes": [{"type": "modified", "added": 17, "removed": 5}, {"type": "mode", "old_mode": "-rw-r--r--", "new_mode": "-rwxr-xr-x"}]}
1475 {"path": "file2", "changes": [{"type": "modified", "added": 135, "removed": 252}]}
1476 {"path": "file4", "changes": [{"type": "added", "size": 0}]}
1477 {"path": "file3", "changes": [{"type": "removed", "size": 0}]}
1478
1480 borg [common options] delete [options] [REPOSITORY_OR_ARCHIVE] [ARCHIVE...]
1481 Description
1483 This command deletes an archive from the repository or the complete
1484 repository. Disk space is reclaimed accordingly. If you delete the
1485 complete repository, the local cache for it (if any) is also deleted.
1486 When using --stats, you will get some statistics about how much data
1488 was deleted - the "Deleted data" deduplicated size there is most inter‐
1489 esting as that is how much your repository will shrink. Please note
1490 that the "All archives" stats refer to the state after deletion.
1491 Examples
1493 # delete a single backup archive:
1494 $ borg delete /path/to/repo::Monday
1495 # delete the whole repository and the related local cache:
1497 $ borg delete /path/to/repo
1498 You requested to completely DELETE the repository *including* all archives it contains:
1499 repo Mon, 2016-02-15 19:26:54
1500 root-2016-02-15 Mon, 2016-02-15 19:36:29
1501 newname Mon, 2016-02-15 19:50:19
1502 Type 'YES' if you understand this and want to continue: YES
1503
1505 borg [common options] prune [options] [REPOSITORY]
1506 Description
1508 The prune command prunes a repository by deleting all archives not
1509 matching any of the specified retention options. This command is nor‐
1510 mally used by automated backup scripts wanting to keep a certain number
1511 of historic backups.
1512 Also, prune automatically removes checkpoint archives (incomplete ar‐
1514 chives left behind by interrupted backup runs) except if the checkpoint
1515 is the latest archive (and thus still needed). Checkpoint archives are
1516 not considered when comparing archive counts against the retention lim‐
1517 its (--keep-X).
1518 If a prefix is set with -P, then only archives that start with the pre‐
1520 fix are considered for deletion and only those archives count towards
1521 the totals specified by the rules. Otherwise, all archives in the
1522 repository are candidates for deletion! There is no automatic distinc‐
1523 tion between archives representing different contents. These need to be
1524 distinguished by specifying matching prefixes.
1525 If you have multiple sequences of archives with different data sets
1527 (e.g. from different machines) in one shared repository, use one prune
1528 call per data set that matches only the respective archives using the
1529 -P option.
1530 The --keep-within option takes an argument of the form "<int><char>",
1532 where char is "H", "d", "w", "m", "y". For example, --keep-within 2d
1533 means to keep all archives that were created within the past 48 hours.
1534 "1m" is taken to mean "31d". The archives kept with this option do not
1535 count towards the totals specified by any other options.
1536 A good procedure is to thin out more and more the older your backups
1538 get. As an example, --keep-daily 7 means to keep the latest backup on
1539 each day, up to 7 most recent days with backups (days without backups
1540 do not count). The rules are applied from secondly to yearly, and
1541 backups selected by previous rules do not count towards those of later
1542 rules. The time that each backup starts is used for pruning purposes.
1543 Dates and times are interpreted in the local timezone, and weeks go
1544 from Monday to Sunday. Specifying a negative number of archives to keep
1545 means that there is no limit.
1546 The --keep-last N option is doing the same as --keep-secondly N (and it
1548 will keep the last N archives under the assumption that you do not cre‐
1549 ate more than one backup archive in the same second).
1550 When using --stats, you will get some statistics about how much data
1552 was deleted - the "Deleted data" deduplicated size there is most inter‐
1553 esting as that is how much your repository will shrink. Please note
1554 that the "All archives" stats refer to the state after pruning.
1555 Examples
1557 Be careful, prune is a potentially dangerous command, it will remove
1558 backup archives.
1559 The default of prune is to apply to all archives in the repository
1561 unless you restrict its operation to a subset of the archives using
1562 --prefix. When using --prefix, be careful to choose a good prefix -
1563 e.g. do not use a prefix "foo" if you do not also want to match "foo‐
1564 bar".
1565 It is strongly recommended to always run prune -v --list --dry-run ...
1567 first so you will see what it would do without it actually doing any‐
1568 thing.
1569 # Keep 7 end of day and 4 additional end of week archives.
1571 # Do a dry-run without actually deleting anything.
1572 $ borg prune -v --list --dry-run --keep-daily=7 --keep-weekly=4 /path/to/repo
1573 # Same as above but only apply to archive names starting with the hostname
1575 # of the machine followed by a "-" character:
1576 $ borg prune -v --list --keep-daily=7 --keep-weekly=4 --prefix='{hostname}-' /path/to/repo
1577 # Keep 7 end of day, 4 additional end of week archives,
1579 # and an end of month archive for every month:
1580 $ borg prune -v --list --keep-daily=7 --keep-weekly=4 --keep-monthly=-1 /path/to/repo
1581 # Keep all backups in the last 10 days, 4 additional end of week archives,
1583 # and an end of month archive for every month:
1584 $ borg prune -v --list --keep-within=10d --keep-weekly=4 --keep-monthly=-1 /path/to/repo
1585 There is also a visualized prune example in docs/misc/prune-exam‐
1587 ple.txt:
1588 borg prune visualized
1590 =====================
1591 Assume it is 2016-01-01, today's backup has not yet been made and you have
1593 created at least one backup on each day in 2015 except on 2015-12-19 (no
1594 backup made on that day).
1595 This is what borg prune --keep-daily 14 --keep-monthly 6 would keep.
1597 Backups kept by the --keep-daily rule are marked by a "d" to the right,
1599 backups kept by the --keep-monthly rule are marked by a "m" to the right.
1600 Calendar view
1602 -------------
1603 2015
1605 January February March
1606 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1607 1 2 3 4 1 1
1608 5 6 7 8 9 10 11 2 3 4 5 6 7 8 2 3 4 5 6 7 8
1609 12 13 14 15 16 17 18 9 10 11 12 13 14 15 9 10 11 12 13 14 15
1610 19 20 21 22 23 24 25 16 17 18 19 20 21 22 16 17 18 19 20 21 22
1611 26 27 28 29 30 31 23 24 25 26 27 28 23 24 25 26 27 28 29
1612 30 31
1613 April May June
1615 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1616 1 2 3 4 5 1 2 3 1 2 3 4 5 6 7
1617 6 7 8 9 10 11 12 4 5 6 7 8 9 10 8 9 10 11 12 13 14
1618 13 14 15 16 17 18 19 11 12 13 14 15 16 17 15 16 17 18 19 20 21
1619 20 21 22 23 24 25 26 18 19 20 21 22 23 24 22 23 24 25 26 27 28
1620 27 28 29 30 25 26 27 28 29 30 31 29 30m
1621 July August September
1624 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1625 1 2 3 4 5 1 2 1 2 3 4 5 6
1626 6 7 8 9 10 11 12 3 4 5 6 7 8 9 7 8 9 10 11 12 13
1627 13 14 15 16 17 18 19 10 11 12 13 14 15 16 14 15 16 17 18 19 20
1628 20 21 22 23 24 25 26 17 18 19 20 21 22 23 21 22 23 24 25 26 27
1629 27 28 29 30 31m 24 25 26 27 28 29 30 28 29 30m
1630 31m
1631 October November December
1633 Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su Mo Tu We Th Fr Sa Su
1634 1 2 3 4 1 1 2 3 4 5 6
1635 5 6 7 8 9 10 11 2 3 4 5 6 7 8 7 8 9 10 11 12 13
1636 12 13 14 15 16 17 18 9 10 11 12 13 14 15 14 15 16 17d18d19 20d
1637 19 20 21 22 23 24 25 16 17 18 19 20 21 22 21d22d23d24d25d26d27d
1638 26 27 28 29 30 31m 23 24 25 26 27 28 29 28d29d30d31d
1639 30m
1640 List view
1642 ---------
1643 --keep-daily 14 --keep-monthly 6
1645 -------------------------------------------------
1646 1. 2015-12-31 (2015-12-31 kept by daily rule)
1647 2. 2015-12-30 1. 2015-11-30
1648 3. 2015-12-29 2. 2015-10-31
1649 4. 2015-12-28 3. 2015-09-30
1650 5. 2015-12-27 4. 2015-08-31
1651 6. 2015-12-26 5. 2015-07-31
1652 7. 2015-12-25 6. 2015-06-30
1653 8. 2015-12-24
1654 9. 2015-12-23
1655 10. 2015-12-22
1656 11. 2015-12-21
1657 12. 2015-12-20
1658 (no backup made on 2015-12-19)
1659 13. 2015-12-18
1660 14. 2015-12-17
1661 Notes
1664 -----
1665 2015-12-31 is kept due to the --keep-daily 14 rule (because it is applied
1667 first), not due to the --keep-monthly rule.
1668 Because of that, the --keep-monthly 6 rule keeps Nov, Oct, Sep, Aug, Jul and
1670 Jun. December is not considered for this rule, because that backup was already
1671 kept because of the daily rule.
1672 2015-12-17 is kept to satisfy the --keep-daily 14 rule - because no backup was
1674 made on 2015-12-19. If a backup had been made on that day, it would not keep
1675 the one from 2015-12-17.
1676 We did not include yearly, weekly, hourly, minutely or secondly rules to keep
1678 this example simple. They all work in basically the same way.
1679 The weekly rule is easy to understand roughly, but hard to understand in all
1681 details. If interested, read "ISO 8601:2000 standard week-based year".
1682
1685 borg [common options] info [options] [REPOSITORY_OR_ARCHIVE]
1686 Description
1688 This command displays detailed information about the specified archive
1689 or repository.
1690 Please note that the deduplicated sizes of the individual archives do
1692 not add up to the deduplicated size of the repository ("all archives"),
1693 because the two are meaning different things:
1694 This archive / deduplicated size = amount of data stored ONLY for this
1696 archive = unique chunks of this archive. All archives / deduplicated
1697 size = amount of data stored in the repo = all chunks in the reposi‐
1698 tory.
1699 Borg archives can only contain a limited amount of file metadata. The
1701 size of an archive relative to this limit depends on a number of fac‐
1702 tors, mainly the number of files, the lengths of paths and other meta‐
1703 data stored for files. This is shown as utilization of maximum sup‐
1704 ported archive size.
1705 Examples
1707 $ borg info /path/to/repo::2017-06-29T11:00-srv
1708 Archive name: 2017-06-29T11:00-srv
1709 Archive fingerprint: b2f1beac2bd553b34e06358afa45a3c1689320d39163890c5bbbd49125f00fe5
1710 Comment:
1711 Hostname: myhostname
1712 Username: root
1713 Time (start): Thu, 2017-06-29 11:03:07
1714 Time (end): Thu, 2017-06-29 11:03:13
1715 Duration: 5.66 seconds
1716 Number of files: 17037
1717 Command line: /usr/sbin/borg create /path/to/repo::2017-06-29T11:00-srv /srv
1718 Utilization of max. archive size: 0%
1719 ------------------------------------------------------------------------------
1720 Original size Compressed size Deduplicated size
1721 This archive: 12.53 GB 12.49 GB 1.62 kB
1722 All archives: 121.82 TB 112.41 TB 215.42 GB
1723 Unique chunks Total chunks
1725 Chunk index: 1015213 626934122
1726 $ borg info /path/to/repo --last 1
1728 Archive name: 2017-06-29T11:00-srv
1729 Archive fingerprint: b2f1beac2bd553b34e06358afa45a3c1689320d39163890c5bbbd49125f00fe5
1730 Comment:
1731 Hostname: myhostname
1732 Username: root
1733 Time (start): Thu, 2017-06-29 11:03:07
1734 Time (end): Thu, 2017-06-29 11:03:13
1735 Duration: 5.66 seconds
1736 Number of files: 17037
1737 Command line: /usr/sbin/borg create /path/to/repo::2017-06-29T11:00-srv /srv
1738 Utilization of max. archive size: 0%
1739 ------------------------------------------------------------------------------
1740 Original size Compressed size Deduplicated size
1741 This archive: 12.53 GB 12.49 GB 1.62 kB
1742 All archives: 121.82 TB 112.41 TB 215.42 GB
1743 Unique chunks Total chunks
1745 Chunk index: 1015213 626934122
1746 $ borg info /path/to/repo
1748 Repository ID: d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1749 Location: /path/to/repo
1750 Encrypted: Yes (repokey)
1751 Cache: /root/.cache/borg/d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1752 Security dir: /root/.config/borg/security/d857ce5788c51272c61535062e89eac4e8ef5a884ffbe976e0af9d8765dedfa5
1753 ------------------------------------------------------------------------------
1754 Original size Compressed size Deduplicated size
1755 All archives: 121.82 TB 112.41 TB 215.42 GB
1756 Unique chunks Total chunks
1758 Chunk index: 1015213 626934122
1759
1761 borg [common options] mount [options] REPOSITORY_OR_ARCHIVE MOUNTPOINT [PATH...]
1762 Description
1764 This command mounts an archive as a FUSE filesystem. This can be useful
1765 for browsing an archive or restoring individual files. Unless the
1766 --foreground option is given the command will run in the background
1767 until the filesystem is umounted.
1768 The command borgfs provides a wrapper for borg mount. This can also be
1770 used in fstab entries: /path/to/repo /mnt/point fuse.borgfs
1771 defaults,noauto 0 0
1772 To allow a regular user to use fstab entries, add the user option:
1774 /path/to/repo /mnt/point fuse.borgfs defaults,noauto,user 0 0
1775 For FUSE configuration and mount options, see the mount.fuse(8) manual
1777 page.
1778 Additional mount options supported by borg:
1780 · versions: when used with a repository mount, this gives a merged,
1782 versioned view of the files in the archives. EXPERIMENTAL, layout may
1783 change in future.
1784 · allow_damaged_files: by default damaged files (where missing chunks
1786 were replaced with runs of zeros by borg check --repair) are not
1787 readable and return EIO (I/O error). Set this option to read such
1788 files.
1789 · ignore_permissions: for security reasons the "default_permissions"
1791 mount option is internally enforced by borg. "ignore_permissions" can
1792 be given to not enforce "default_permissions".
1793 The BORG_MOUNT_DATA_CACHE_ENTRIES environment variable is meant for
1795 advanced users to tweak the performance. It sets the number of cached
1796 data chunks; additional memory usage can be up to ~8 MiB times this
1797 number. The default is the number of CPU cores.
1798 When the daemonized process receives a signal or crashes, it does not
1800 unmount. Unmounting in these cases could cause an active rsync or sim‐
1801 ilar process to unintentionally delete data.
1802 When running in the foreground ^C/SIGINT unmounts cleanly, but other
1804 signals or crashes do not.
1805
1807 borg [common options] umount [options] MOUNTPOINT
1808 Description
1810 This command un-mounts a FUSE filesystem that was mounted with borg
1811 mount.
1812 This is a convenience wrapper that just calls the platform-specific
1814 shell command - usually this is either umount or fusermount -u.
1815 Examples
1817 # Mounting the repository shows all archives.
1818 # Archives are loaded lazily, expect some delay when navigating to an archive
1819 # for the first time.
1820 $ borg mount /path/to/repo /tmp/mymountpoint
1821 $ ls /tmp/mymountpoint
1822 root-2016-02-14 root-2016-02-15
1823 $ borg umount /tmp/mymountpoint
1824 # Mounting a specific archive is possible as well.
1826 $ borg mount /path/to/repo::root-2016-02-15 /tmp/mymountpoint
1827 $ ls /tmp/mymountpoint
1828 bin boot etc home lib lib64 lost+found media mnt opt
1829 root sbin srv tmp usr var
1830 $ borg umount /tmp/mymountpoint
1831 # The experimental "versions view" merges all archives in the repository
1833 # and provides a versioned view on files.
1834 $ borg mount -o versions /path/to/repo /tmp/mymountpoint
1835 $ ls -l /tmp/mymountpoint/home/user/doc.txt/
1836 total 24
1837 -rw-rw-r-- 1 user group 12357 Aug 26 21:19 doc.cda00bc9.txt
1838 -rw-rw-r-- 1 user group 12204 Aug 26 21:04 doc.fa760f28.txt
1839 $ borg umount /tmp/mymountpoint
1840 # Archive filters are supported.
1842 # These are especially handy for the "versions view",
1843 # which does not support lazy processing of archives.
1844 $ borg mount -o versions --glob-archives '*-my-home' --last 10 /path/to/repo /tmp/mymountpoint
1845 # Exclusion options are supported.
1847 # These can speed up mounting and lower memory needs significantly.
1848 $ borg mount /path/to/repo /tmp/mymountpoint only/that/path
1849 $ borg mount --exclude '...' /path/to/repo /tmp/mymountpoint
1850 borgfs
1852 $ echo '/mnt/backup /tmp/myrepo fuse.borgfs defaults,noauto 0 0' >> /etc/fstab
1853 $ echo '/mnt/backup::root-2016-02-15 /tmp/myarchive fuse.borgfs defaults,noauto 0 0' >> /etc/fstab
1854 $ mount /tmp/myrepo
1855 $ mount /tmp/myarchive
1856 $ ls /tmp/myrepo
1857 root-2016-02-01 root-2016-02-2015
1858 $ ls /tmp/myarchive
1859 bin boot etc home lib lib64 lost+found media mnt opt root sbin srv tmp usr var
1860 NOTE:
1862 borgfs will be automatically provided if you used a distribution
1863 package, pip or setup.py to install Borg. Users of the standalone
1864 binary will have to manually create a symlink (see pyin‐
1865 staller-binary).
1866
1868 borg [common options] key change-passphrase [options] [REPOSITORY]
1869 Description
1871 The key files used for repository encryption are optionally passphrase
1872 protected. This command can be used to change this passphrase.
1873 Please note that this command only changes the passphrase, but not any
1875 secret protected by it (like e.g. encryption/MAC keys or chunker seed).
1876 Thus, changing the passphrase after passphrase and borg key got compro‐
1877 mised does not protect future (nor past) backups to the same reposi‐
1878 tory.
1879 Examples
1881 # Create a key file protected repository
1882 $ borg init --encryption=keyfile -v /path/to/repo
1883 Initializing repository at "/path/to/repo"
1884 Enter new passphrase:
1885 Enter same passphrase again:
1886 Remember your passphrase. Your data will be inaccessible without it.
1887 Key in "/root/.config/borg/keys/mnt_backup" created.
1888 Keep this key safe. Your data will be inaccessible without it.
1889 Synchronizing chunks cache...
1890 Archives: 0, w/ cached Idx: 0, w/ outdated Idx: 0, w/o cached Idx: 0.
1891 Done.
1892 # Change key file passphrase
1894 $ borg key change-passphrase -v /path/to/repo
1895 Enter passphrase for key /root/.config/borg/keys/mnt_backup:
1896 Enter new passphrase:
1897 Enter same passphrase again:
1898 Remember your passphrase. Your data will be inaccessible without it.
1899 Key updated
1900 Fully automated using environment variables:
1902 $ BORG_NEW_PASSPHRASE=old borg init -e=repokey repo
1904 # now "old" is the current passphrase.
1905 $ BORG_PASSPHRASE=old BORG_NEW_PASSPHRASE=new borg key change-passphrase repo
1906 # now "new" is the current passphrase.
1907
1909 borg [common options] key export [options] [REPOSITORY] [PATH]
1910 Description
1912 If repository encryption is used, the repository is inaccessible with‐
1913 out the key. This command allows to backup this essential key. Note
1914 that the backup produced does not include the passphrase itself (i.e.
1915 the exported key stays encrypted). In order to regain access to a
1916 repository, one needs both the exported key and the original
1917 passphrase.
1918 There are two backup formats. The normal backup format is suitable for
1920 digital storage as a file. The --paper backup format is optimized for
1921 printing and typing in while importing, with per line checks to reduce
1922 problems with manual input.
1923 For repositories using keyfile encryption the key is saved locally on
1925 the system that is capable of doing backups. To guard against loss of
1926 this key, the key needs to be backed up independently of the main data
1927 backup.
1928 For repositories using the repokey encryption the key is saved in the
1930 repository in the config file. A backup is thus not strictly needed,
1931 but guards against the repository becoming inaccessible if the file is
1932 damaged for some reason.
1933
1935 borg [common options] key import [options] [REPOSITORY] [PATH]
1936 Description
1938 This command allows to restore a key previously backed up with the
1939 export command.
1940 If the --paper option is given, the import will be an interactive
1942 process in which each line is checked for plausibility before proceed‐
1943 ing to the next line. For this format PATH must not be given.
1944
1946 borg [common options] upgrade [options] [REPOSITORY]
1947 Description
1949 Upgrade an existing, local Borg repository.
1950 When you do not need borg upgrade
1952 Not every change requires that you run borg upgrade.
1953 You do not need to run it when:
1955 · moving your repository to a different place
1957 · upgrading to another point release (like 1.0.x to 1.0.y), except when
1959 noted otherwise in the changelog
1960 · upgrading from 1.0.x to 1.1.x, except when noted otherwise in the
1962 changelog
1963 Borg 1.x.y upgrades
1965 Use borg upgrade --tam REPO to require manifest authentication intro‐
1966 duced with Borg 1.0.9 to address security issues. This means that modi‐
1967 fying the repository after doing this with a version prior to 1.0.9
1968 will raise a validation error, so only perform this upgrade after
1969 updating all clients using the repository to 1.0.9 or newer.
1970 This upgrade should be done on each client for safety reasons.
1972 If a repository is accidentally modified with a pre-1.0.9 client after
1974 this upgrade, use borg upgrade --tam --force REPO to remedy it.
1975 If you routinely do this you might not want to enable this upgrade
1977 (which will leave you exposed to the security issue). You can reverse
1978 the upgrade by issuing borg upgrade --disable-tam REPO.
1979 See
1981 https://borgbackup.readthedocs.io/en/stable/changes.html#pre-1-0-9-manifest-spoofing-vulnerability
1982 for details.
1983 Attic and Borg 0.xx to Borg 1.x
1985 This currently supports converting an Attic repository to Borg and also
1986 helps with converting Borg 0.xx to 1.0.
1987 Currently, only LOCAL repositories can be upgraded (issue #465).
1989 Please note that borg create (since 1.0.0) uses bigger chunks by
1991 default than old borg or attic did, so the new chunks won't deduplicate
1992 with the old chunks in the upgraded repository. See --chunker-params
1993 option of borg create and borg recreate.
1994 borg upgrade will change the magic strings in the repository's segments
1996 to match the new Borg magic strings. The keyfiles found in
1997 $ATTIC_KEYS_DIR or ~/.attic/keys/ will also be converted and copied to
1998 $BORG_KEYS_DIR or ~/.config/borg/keys.
1999 The cache files are converted, from $ATTIC_CACHE_DIR or ~/.cache/attic
2001 to $BORG_CACHE_DIR or ~/.cache/borg, but the cache layout between Borg
2002 and Attic changed, so it is possible the first backup after the conver‐
2003 sion takes longer than expected due to the cache resync.
2004 Upgrade should be able to resume if interrupted, although it will still
2006 iterate over all segments. If you want to start from scratch, use borg
2007 delete over the copied repository to make sure the cache files are also
2008 removed:
2009 borg delete borg
2011 Unless --inplace is specified, the upgrade process first creates a
2013 backup copy of the repository, in REPOSITORY.before-upgrade-DATETIME,
2014 using hardlinks. This requires that the repository and its parent
2015 directory reside on same filesystem so the hardlink copy can work.
2016 This takes longer than in place upgrades, but is much safer and gives
2017 progress information (as opposed to cp -al). Once you are satisfied
2018 with the conversion, you can safely destroy the backup copy.
2019 WARNING: Running the upgrade in place will make the current copy unus‐
2021 able with older version, with no way of going back to previous ver‐
2022 sions. This can PERMANENTLY DAMAGE YOUR REPOSITORY! Attic CAN NOT READ
2023 BORG REPOSITORIES, as the magic strings have changed. You have been
2024 warned.
2025 Examples
2027 # Upgrade the borg repository to the most recent version.
2028 $ borg upgrade -v /path/to/repo
2029 making a hardlink copy in /path/to/repo.before-upgrade-2016-02-15-20:51:55
2030 opening attic repository with borg and converting
2031 no key file found for repository
2032 converting repo index /path/to/repo/index.0
2033 converting 1 segments...
2034 converting borg 0.xx to borg current
2035 no key file found for repository
2036 Upgrading a passphrase encrypted attic repo
2038 attic offered a "passphrase" encryption mode, but this was removed in
2039 borg 1.0 and replaced by the "repokey" mode (which stores the
2040 passphrase-protected encryption key into the repository config).
2041 Thus, to upgrade a "passphrase" attic repo to a "repokey" borg repo, 2
2043 steps are needed, in this order:
2044 · borg upgrade repo
2046 · borg key migrate-to-repokey repo
2048
2050 borg [common options] recreate [options] [REPOSITORY_OR_ARCHIVE] [PATH...]
2051 Description
2053 Recreate the contents of existing archives.
2054 recreate is a potentially dangerous function and might lead to data
2056 loss (if used wrongly). BE VERY CAREFUL!
2057 --exclude, --exclude-from, --exclude-if-present, --keep-exclude-tags
2059 and PATH have the exact same semantics as in "borg create", but they
2060 only check for files in the archives and not in the local file system.
2061 If PATHs are specified, the resulting archives will only contain files
2062 from these PATHs.
2063 Note that all paths in an archive are relative, therefore absolute pat‐
2065 terns/paths will not match (--exclude, --exclude-from, PATHs).
2066 --recompress allows to change the compression of existing data in ar‐
2068 chives. Due to how Borg stores compressed size information this might
2069 display incorrect information for archives that were not recreated at
2070 the same time. There is no risk of data loss by this.
2071 --chunker-params will re-chunk all files in the archive, this can be
2073 used to have upgraded Borg 0.xx or Attic archives deduplicate with Borg
2074 1.x archives.
2075 USE WITH CAUTION. Depending on the PATHs and patterns given, recreate
2077 can be used to permanently delete files from archives. When in doubt,
2078 use --dry-run --verbose --list to see how patterns/PATHS are inter‐
2079 preted. See list_item_flags in borg create for details.
2080 The archive being recreated is only removed after the operation com‐
2082 pletes. The archive that is built during the operation exists at the
2083 same time at "<ARCHIVE>.recreate". The new archive will have a differ‐
2084 ent archive ID.
2085 With --target the original archive is not replaced, instead a new ar‐
2087 chive is created.
2088 When rechunking space usage can be substantial, expect at least the
2090 entire deduplicated size of the archives using the previous chunker
2091 params. When recompressing expect approx. (throughput / check‐
2092 point-interval) in space usage, assuming all chunks are recompressed.
2093 If you recently ran borg check --repair and it had to fix lost chunks
2095 with all-zero replacement chunks, please first run another backup for
2096 the same data and re-run borg check --repair afterwards to heal any ar‐
2097 chives that had lost chunks which are still generated from the input
2098 data.
2099 Important: running borg recreate to re-chunk will remove the
2101 chunks_healthy metadata of all items with replacement chunks, so heal‐
2102 ing will not be possible any more after re-chunking (it is also
2103 unlikely it would ever work: due to the change of chunking parameters,
2104 the missing chunk likely will never be seen again even if you still
2105 have the data that produced it).
2106 Examples
2108 # Make old (Attic / Borg 0.xx) archives deduplicate with Borg 1.x archives.
2109 # Archives created with Borg 1.1+ and the default chunker params are skipped
2110 # (archive ID stays the same).
2111 $ borg recreate /mnt/backup --chunker-params default --progress
2112 # Create a backup with little but fast compression
2114 $ borg create /mnt/backup::archive /some/files --compression lz4
2115 # Then compress it - this might take longer, but the backup has already completed,
2116 # so no inconsistencies from a long-running backup job.
2117 $ borg recreate /mnt/backup::archive --recompress --compression zlib,9
2118 # Remove unwanted files from all archives in a repository.
2120 # Note the relative path for the --exclude option - archives only contain relative paths.
2121 $ borg recreate /mnt/backup --exclude home/icke/Pictures/drunk_photos
2122 # Change archive comment
2124 $ borg create --comment "This is a comment" /mnt/backup::archivename ~
2125 $ borg info /mnt/backup::archivename
2126 Name: archivename
2127 Fingerprint: ...
2128 Comment: This is a comment
2129 ...
2130 $ borg recreate --comment "This is a better comment" /mnt/backup::archivename
2131 $ borg info /mnt/backup::archivename
2132 Name: archivename
2133 Fingerprint: ...
2134 Comment: This is a better comment
2135 ...
2136
2138 borg [common options] export-tar [options] ARCHIVE FILE [PATH...]
2139 Description
2141 This command creates a tarball from an archive.
2142 When giving '-' as the output FILE, Borg will write a tar stream to
2144 standard output.
2145 By default (--tar-filter=auto) Borg will detect whether the FILE should
2147 be compressed based on its file extension and pipe the tarball through
2148 an appropriate filter before writing it to FILE:
2149 · .tar.gz: gzip
2151 · .tar.bz2: bzip2
2153 · .tar.xz: xz
2155 Alternatively a --tar-filter program may be explicitly specified. It
2157 should read the uncompressed tar stream from stdin and write a com‐
2158 pressed/filtered tar stream to stdout.
2159 The generated tarball uses the GNU tar format.
2161 export-tar is a lossy conversion: BSD flags, ACLs, extended attributes
2163 (xattrs), atime and ctime are not exported. Timestamp resolution is
2164 limited to whole seconds, not the nanosecond resolution otherwise sup‐
2165 ported by Borg.
2166 A --sparse option (as found in borg extract) is not supported.
2168 By default the entire archive is extracted but a subset of files and
2170 directories can be selected by passing a list of PATHs as arguments.
2171 The file selection can further be restricted by using the --exclude
2172 option.
2173 See the output of the "borg help patterns" command for more help on
2175 exclude patterns.
2176 --progress can be slower than no progress display, since it makes one
2178 additional pass over the archive metadata.
2179 Examples
2181 # export as uncompressed tar
2182 $ borg export-tar /path/to/repo::Monday Monday.tar
2183 # exclude some types, compress using gzip
2185 $ borg export-tar /path/to/repo::Monday Monday.tar.gz --exclude '*.so'
2186 # use higher compression level with gzip
2188 $ borg export-tar --tar-filter="gzip -9" testrepo::linux Monday.tar.gz
2189 # export a tar, but instead of storing it on disk,
2191 # upload it to a remote site using curl.
2192 $ borg export-tar /path/to/repo::Monday - | curl --data-binary @- https://somewhere/to/POST
2193 # remote extraction via "tarpipe"
2195 $ borg export-tar /path/to/repo::Monday - | ssh somewhere "cd extracted; tar x"
2196
2198 borg [common options] serve [options]
2199 Description
2201 This command starts a repository server process. This command is usu‐
2202 ally not used manually.
2203 Examples
2205 borg serve has special support for ssh forced commands (see autho‐
2206 rized_keys example below): it will detect that you use such a forced
2207 command and extract the value of the --restrict-to-path option(s).
2208 It will then parse the original command that came from the client,
2210 makes sure that it is also borg serve and enforce path restriction(s)
2211 as given by the forced command. That way, other options given by the
2212 client (like --info or --umask) are preserved (and are not fixed by the
2213 forced command).
2214 Environment variables (such as BORG_HOSTNAME_IS_UNIQUE) contained in
2216 the original command sent by the client are not interpreted, but
2217 ignored. If BORG_XXX environment variables should be set on the borg
2218 serve side, then these must be set in system-specific locations like
2219 /etc/environment or in the forced command itself (example below).
2220 # Allow an SSH keypair to only run borg, and only have access to /path/to/repo.
2222 # Use key options to disable unneeded and potentially dangerous SSH functionality.
2223 # This will help to secure an automated remote backup system.
2224 $ cat ~/.ssh/authorized_keys
2225 command="borg serve --restrict-to-path /path/to/repo",restrict ssh-rsa AAAAB3[...]
2226 # Set a BORG_XXX environment variable on the "borg serve" side
2228 $ cat ~/.ssh/authorized_keys
2229 command="export BORG_XXX=value; borg serve [...]",restrict ssh-rsa [...]
2230 NOTE:
2232 The examples above use the restrict directive. This does automati‐
2233 cally block potential dangerous ssh features, even when they are
2234 added in a future update. Thus, this option should be preferred.
2235 If you're using openssh-server < 7.2, however, you have to explic‐
2237 itly specify the ssh features to restrict and cannot simply use the
2238 restrict option as it has been introduced in v7.2. We recommend to
2239 use no-port-forwarding,no-X11-forwarding,no-pty,no-agent-forward‐
2240 ing,no-user-rc in this case.
2241 SSH Configuration
2243 borg serve's pipes (stdin/stdout/stderr) are connected to the sshd
2244 process on the server side. In the event that the SSH connection
2245 between borg serve and the client is disconnected or stuck abnormally
2246 (for example, due to a network outage), it can take a long time for
2247 sshd to notice the client is disconnected. In the meantime, sshd con‐
2248 tinues running, and as a result so does the borg serve process holding
2249 the lock on the repository. This can cause subsequent borg operations
2250 on the remote repository to fail with the error: Failed to cre‐
2251 ate/acquire the lock.
2252 In order to avoid this, it is recommended to perform the following
2254 additional SSH configuration:
2255 Either in the client side's ~/.ssh/config file, or in the client's
2257 /etc/ssh/ssh_config file:
2258 Host backupserver
2260 ServerAliveInterval 10
2261 ServerAliveCountMax 30
2262 Replacing backupserver with the hostname, FQDN or IP address of the
2264 borg server.
2265 This will cause the client to send a keepalive to the server every 10
2267 seconds. If 30 consecutive keepalives are sent without a response (a
2268 time of 300 seconds), the ssh client process will be terminated, caus‐
2269 ing the borg process to terminate gracefully.
2270 On the server side's sshd configuration file (typically
2272 /etc/ssh/sshd_config):
2273 ClientAliveInterval 10
2275 ClientAliveCountMax 30
2276 This will cause the server to send a keep alive to the client every 10
2278 seconds. If 30 consecutive keepalives are sent without a response (a
2279 time of 300 seconds), the server's sshd process will be terminated,
2280 causing the borg serve process to terminate gracefully and release the
2281 lock on the repository.
2282 If you then run borg commands with --lock-wait 600, this gives suffi‐
2284 cient time for the borg serve processes to terminate after the SSH con‐
2285 nection is torn down after the 300 second wait for the keepalives to
2286 fail.
2287 You may, of course, modify the timeout values demonstrated above to
2289 values that suit your environment and use case.
2290
2292 borg [common options] config [options] [REPOSITORY] [NAME] [VALUE]
2293 Description
2295 This command gets and sets options in a local repository or cache con‐
2296 fig file. For security reasons, this command only works on local
2297 repositories.
2298 To delete a config value entirely, use --delete. To list the values of
2300 the configuration file or the default values, use --list. To get and
2301 existing key, pass only the key name. To set a key, pass both the key
2302 name and the new value. Keys can be specified in the format "sec‐
2303 tion.name" or simply "name"; the section will default to "repository"
2304 and "cache" for the repo and cache configs, respectively.
2305 By default, borg config manipulates the repository config file. Using
2307 --cache edits the repository cache's config file instead.
2308 NOTE:
2310 The repository & cache config files are some of the only directly
2311 manipulable parts of a repository that aren't versioned or backed
2312 up, so be careful when making changes!
2313 Examples
2315 # find cache directory
2316 $ cd ~/.cache/borg/$(borg config /path/to/repo id)
2317 # reserve some space
2319 $ borg config /path/to/repo additional_free_space 2G
2320 # make a repo append-only
2322 $ borg config /path/to/repo append_only 1
2323
2325 borg [common options] with-lock [options] REPOSITORY COMMAND [ARGS...]
2326 Description
2328 This command runs a user-specified command while the repository lock is
2329 held.
2330 It will first try to acquire the lock (make sure that no other opera‐
2332 tion is running in the repo), then execute the given command as a sub‐
2333 process and wait for its termination, release the lock and return the
2334 user command's return code as borg's return code.
2335 NOTE:
2337 If you copy a repository with the lock held, the lock will be
2338 present in the copy. Thus, before using borg on the copy from a dif‐
2339 ferent host, you need to use "borg break-lock" on the copied reposi‐
2340 tory, because Borg is cautious and does not automatically remove
2341 stale locks made by a different host.
2342
2344 borg [common options] break-lock [options] [REPOSITORY]
2345 Description
2347 This command breaks the repository and cache locks. Please use care‐
2348 fully and only while no borg process (on any machine) is trying to
2349 access the Cache or the Repository.
2350
2352 borg [common options] benchmark crud [options] REPOSITORY PATH
2353 Description
2355 This command benchmarks borg CRUD (create, read, update, delete) opera‐
2356 tions.
2357 It creates input data below the given PATH and backups this data into
2359 the given REPO. The REPO must already exist (it could be a fresh empty
2360 repo or an existing repo, the command will create / read / update /
2361 delete some archives named borg-benchmark-crud* there.
2362 Make sure you have free space there, you'll need about 1GB each (+
2364 overhead).
2365 If your repository is encrypted and borg needs a passphrase to unlock
2367 the key, use:
2368 BORG_PASSPHRASE=mysecret borg benchmark crud REPO PATH
2370 Measurements are done with different input file sizes and counts. The
2372 file contents are very artificial (either all zero or all random), thus
2373 the measurement results do not necessarily reflect performance with
2374 real data. Also, due to the kind of content used, no compression is
2375 used in these benchmarks.
2376 C- == borg create (1st archive creation, no compression, do not use
2378 files cache)
2379 C-Z- == all-zero files. full dedup, this is primarily measuring
2380 reader/chunker/hasher. C-R- == random files. no dedup, measur‐
2381 ing throughput through all processing stages.
2382 R- == borg extract (extract archive, dry-run, do everything, but do not
2384 write files to disk)
2385 R-Z- == all zero files. Measuring heavily duplicated files.
2386 R-R- == random files. No duplication here, measuring throughput
2387 through all processing stages, except writing to disk.
2388 U- == borg create (2nd archive creation of unchanged input files, mea‐
2390 sure files cache speed)
2391 The throughput value is kind of virtual here, it does not actu‐
2392 ally read the file. U-Z- == needs to check the 2 all-zero
2393 chunks' existence in the repo. U-R- == needs to check existence
2394 of a lot of different chunks in the repo.
2395 D- == borg delete archive (delete last remaining archive, measure dele‐
2397 tion + compaction)
2398 D-Z- == few chunks to delete / few segments to compact/remove.
2399 D-R- == many chunks to delete / many segments to compact/remove.
2400 Please note that there might be quite some variance in these measure‐
2402 ments. Try multiple measurements and having a otherwise idle machine
2403 (and network, if you use it).
2404
2406 borg help patterns
2407 The path/filenames used as input for the pattern matching start from
2408 the currently active recursion root. You usually give the recursion
2409 root(s) when invoking borg and these can be either relative or absolute
2410 paths.
2411 So, when you give relative/ as root, the paths going into the matcher
2413 will look like relative/.../file.ext. When you give /absolute/ as root,
2414 they will look like /absolute/.../file.ext. This is meant when we talk
2415 about "full path" below.
2416 File paths in Borg archives are always stored normalized and relative.
2418 This means that e.g. borg create /path/to/repo ../some/path will store
2419 all files as some/path/.../file.ext and borg create /path/to/repo
2420 /home/user will store all files as home/user/.../file.ext. Therefore,
2421 always use relative paths in your patterns when matching archive con‐
2422 tent in commands like extract or mount. Starting with Borg 1.2 this be‐
2423 haviour will be changed to accept both absolute and relative paths.
2424 File patterns support these styles: fnmatch, shell, regular expres‐
2426 sions, path prefixes and path full-matches. By default, fnmatch is used
2427 for --exclude patterns and shell-style is used for the experimental
2428 --pattern option.
2429 If followed by a colon (':') the first two characters of a pattern are
2431 used as a style selector. Explicit style selection is necessary when a
2432 non-default style is desired or when the desired pattern starts with
2433 two alphanumeric characters followed by a colon (i.e. aa:something/*).
2434 Fnmatch, selector fm:
2436 This is the default style for --exclude and --exclude-from.
2437 These patterns use a variant of shell pattern syntax, with '*'
2438 matching any number of characters, '?' matching any single char‐
2439 acter, '[...]' matching any single character specified, includ‐
2440 ing ranges, and '[!...]' matching any character not specified.
2441 For the purpose of these patterns, the path separator (backslash
2442 for Windows and '/' on other systems) is not treated specially.
2443 Wrap meta-characters in brackets for a literal match (i.e. [?]
2444 to match the literal character ?). For a path to match a pat‐
2445 tern, the full path must match, or it must match from the start
2446 of the full path to just before a path separator. Except for the
2447 root path, paths will never end in the path separator when
2448 matching is attempted. Thus, if a given pattern ends in a path
2449 separator, a '*' is appended before matching is attempted.
2450 Shell-style patterns, selector sh:
2452 This is the default style for --pattern and --patterns-from.
2453 Like fnmatch patterns these are similar to shell patterns. The
2454 difference is that the pattern may include **/ for matching zero
2455 or more directory levels, * for matching zero or more arbitrary
2456 characters with the exception of any path separator.
2457 Regular expressions, selector re:
2459 Regular expressions similar to those found in Perl are sup‐
2460 ported. Unlike shell patterns regular expressions are not
2461 required to match the full path and any substring match is suf‐
2462 ficient. It is strongly recommended to anchor patterns to the
2463 start ('^'), to the end ('$') or both. Path separators (back‐
2464 slash for Windows and '/' on other systems) in paths are always
2465 normalized to a forward slash ('/') before applying a pattern.
2466 The regular expression syntax is described in the Python docu‐
2467 mentation for the re module.
2468 Path prefix, selector pp:
2470 This pattern style is useful to match whole sub-directories. The
2471 pattern pp:root/somedir matches root/somedir and everything
2472 therein.
2473 Path full-match, selector pf:
2475 This pattern style is (only) useful to match full paths. This
2476 is kind of a pseudo pattern as it can not have any variable or
2477 unspecified parts - the full path must be given.
2478 pf:root/file.ext matches root/file.ext only.
2479 Implementation note: this is implemented via very time-efficient
2481 O(1) hashtable lookups (this means you can have huge amounts of
2482 such patterns without impacting performance much). Due to that,
2483 this kind of pattern does not respect any context or order. If
2484 you use such a pattern to include a file, it will always be
2485 included (if the directory recursion encounters it). Other
2486 include/exclude patterns that would normally match will be
2487 ignored. Same logic applies for exclude.
2488 NOTE:
2490 re:, sh: and fm: patterns are all implemented on top of the Python
2491 SRE engine. It is very easy to formulate patterns for each of these
2492 types which requires an inordinate amount of time to match paths. If
2493 untrusted users are able to supply patterns, ensure they cannot sup‐
2494 ply re: patterns. Further, ensure that sh: and fm: patterns only
2495 contain a handful of wildcards at most.
2496 Exclusions can be passed via the command line option --exclude. When
2498 used from within a shell, the patterns should be quoted to protect them
2499 from expansion.
2500 The --exclude-from option permits loading exclusion patterns from a
2502 text file with one pattern per line. Lines empty or starting with the
2503 number sign ('#') after removing whitespace on both ends are ignored.
2504 The optional style selector prefix is also supported for patterns
2505 loaded from a file. Due to whitespace removal, paths with whitespace at
2506 the beginning or end can only be excluded using regular expressions.
2507 To test your exclusion patterns without performing an actual backup you
2509 can run borg create --list --dry-run ....
2510 Examples:
2512 # Exclude '/home/user/file.o' but not '/home/user/file.odt':
2514 $ borg create -e '*.o' backup /
2515 # Exclude '/home/user/junk' and '/home/user/subdir/junk' but
2517 # not '/home/user/importantjunk' or '/etc/junk':
2518 $ borg create -e '/home/*/junk' backup /
2519 # Exclude the contents of '/home/user/cache' but not the directory itself:
2521 $ borg create -e /home/user/cache/ backup /
2522 # The file '/home/user/cache/important' is *not* backed up:
2524 $ borg create -e /home/user/cache/ backup / /home/user/cache/important
2525 # The contents of directories in '/home' are not backed up when their name
2527 # ends in '.tmp'
2528 $ borg create --exclude 're:^/home/[^/]+\.tmp/' backup /
2529 # Load exclusions from file
2531 $ cat >exclude.txt <<EOF
2532 # Comment line
2533 /home/*/junk
2534 *.tmp
2535 fm:aa:something/*
2536 re:^home/[^/]\.tmp/
2537 sh:home/*/.thumbnails
2538 # Example with spaces, no need to escape as it is processed by borg
2539 some file with spaces.txt
2540 EOF
2541 $ borg create --exclude-from exclude.txt backup /
2542 A more general and easier to use way to define filename matching pat‐
2543 terns exists with the experimental --pattern and --patterns-from
2544 options. Using these, you may specify the backup roots (starting
2545 points) and patterns for inclusion/exclusion. A root path starts with
2546 the prefix R, followed by a path (a plain path, not a file pattern). An
2547 include rule starts with the prefix +, an exclude rule starts with the
2548 prefix -, an exclude-norecurse rule starts with !, all followed by a
2549 pattern.
2550 NOTE:
2552 Via --pattern or --patterns-from you can define BOTH inclusion and
2553 exclusion of files using pattern prefixes + and -. With --exclude
2554 and --exlude-from ONLY excludes are defined.
2555 Inclusion patterns are useful to include paths that are contained in an
2557 excluded path. The first matching pattern is used so if an include pat‐
2558 tern matches before an exclude pattern, the file is backed up. If an
2559 exclude-norecurse pattern matches a directory, it won't recurse into it
2560 and won't discover any potential matches for include rules below that
2561 directory.
2562 Note that the default pattern style for --pattern and --patterns-from
2564 is shell style (sh:), so those patterns behave similar to rsync
2565 include/exclude patterns. The pattern style can be set via the P pre‐
2566 fix.
2567 Patterns (--pattern) and excludes (--exclude) from the command line are
2569 considered first (in the order of appearance). Then patterns from
2570 --patterns-from are added. Exclusion patterns from --exclude-from files
2571 are appended last.
2572 Examples:
2574 # backup pics, but not the ones from 2018, except the good ones:
2576 # note: using = is essential to avoid cmdline argument parsing issues.
2577 borg create --pattern=+pics/2018/good --pattern=-pics/2018 repo::arch pics
2578 # use a file with patterns:
2580 borg create --patterns-from patterns.lst repo::arch
2581 The patterns.lst file could look like that:
2583 # "sh:" pattern style is the default, so the following line is not needed:
2585 P sh
2586 R /
2587 # can be rebuild
2588 - /home/*/.cache
2589 # they're downloads for a reason
2590 - /home/*/Downloads
2591 # susan is a nice person
2592 # include susans home
2593 + /home/susan
2594 # don't backup the other home directories
2595 - /home/*
2596 # don't even look in /proc
2597 ! /proc
2598 borg help placeholders
2600 Repository (or Archive) URLs, --prefix, --glob-archives, --comment and
2601 --remote-path values support these placeholders:
2602 {hostname}
2604 The (short) hostname of the machine.
2605 {fqdn} The full name of the machine.
2607 {reverse-fqdn}
2609 The full name of the machine in reverse domain name notation.
2610 {now} The current local date and time, by default in ISO-8601 format.
2612 You can also supply your own format string, e.g.
2613 {now:%Y-%m-%d_%H:%M:%S}
2614 {utcnow}
2616 The current UTC date and time, by default in ISO-8601 format.
2617 You can also supply your own format string, e.g. {utc‐
2618 now:%Y-%m-%d_%H:%M:%S}
2619 {user} The user name (or UID, if no name is available) of the user run‐
2621 ning borg.
2622 {pid} The current process ID.
2624 {borgversion}
2626 The version of borg, e.g.: 1.0.8rc1
2627 {borgmajor}
2629 The version of borg, only the major version, e.g.: 1
2630 {borgminor}
2632 The version of borg, only major and minor version, e.g.: 1.0
2633 {borgpatch}
2635 The version of borg, only major, minor and patch version, e.g.:
2636 1.0.8
2637 If literal curly braces need to be used, double them for escaping:
2639 borg create /path/to/repo::{{literal_text}}
2641 Examples:
2643 borg create /path/to/repo::{hostname}-{user}-{utcnow} ...
2645 borg create /path/to/repo::{hostname}-{now:%Y-%m-%d_%H:%M:%S} ...
2646 borg prune --prefix '{hostname}-' ...
2647 NOTE:
2649 systemd uses a difficult, non-standard syntax for command lines in
2650 unit files (refer to the systemd.unit(5) manual page).
2651 When invoking borg from unit files, pay particular attention to
2653 escaping, especially when using the now/utcnow placeholders, since
2654 systemd performs its own %-based variable replacement even in quoted
2655 text. To avoid interference from systemd, double all percent signs
2656 ({hostname}-{now:%Y-%m-%d_%H:%M:%S} becomes {host‐
2657 name}-{now:%%Y-%%m-%%d_%%H:%%M:%%S}).
2658 borg help compression
2660 It is no problem to mix different compression methods in one repo,
2661 deduplication is done on the source data chunks (not on the compressed
2662 or encrypted data).
2663 If some specific chunk was once compressed and stored into the repo,
2665 creating another backup that also uses this chunk will not change the
2666 stored chunk. So if you use different compression specs for the back‐
2667 ups, whichever stores a chunk first determines its compression. See
2668 also borg recreate.
2669 Compression is lz4 by default. If you want something else, you have to
2671 specify what you want.
2672 Valid compression specifiers are:
2674 none Do not compress.
2676 lz4 Use lz4 compression. Very high speed, very low compression.
2678 (default)
2679 zstd[,L]
2681 Use zstd ("zstandard") compression, a modern wide-range algo‐
2682 rithm. If you do not explicitely give the compression level L
2683 (ranging from 1 to 22), it will use level 3. Archives com‐
2684 pressed with zstd are not compatible with borg < 1.1.4.
2685 zlib[,L]
2687 Use zlib ("gz") compression. Medium speed, medium compression.
2688 If you do not explicitely give the compression level L (ranging
2689 from 0 to 9), it will use level 6. Giving level 0 (means "no
2690 compression", but still has zlib protocol overhead) is usually
2691 pointless, you better use "none" compression.
2692 lzma[,L]
2694 Use lzma ("xz") compression. Low speed, high compression. If
2695 you do not explicitely give the compression level L (ranging
2696 from 0 to 9), it will use level 6. Giving levels above 6 is
2697 pointless and counterproductive because it does not compress
2698 better due to the buffer size used by borg - but it wastes lots
2699 of CPU cycles and RAM.
2700 auto,C[,L]
2702 Use a built-in heuristic to decide per chunk whether to compress
2703 or not. The heuristic tries with lz4 whether the data is com‐
2704 pressible. For incompressible data, it will not use compression
2705 (uses "none"). For compressible data, it uses the given C[,L]
2706 compression - with C[,L] being any valid compression specifier.
2707 Examples:
2709 borg create --compression lz4 REPO::ARCHIVE data
2711 borg create --compression zstd REPO::ARCHIVE data
2712 borg create --compression zstd,10 REPO::ARCHIVE data
2713 borg create --compression zlib REPO::ARCHIVE data
2714 borg create --compression zlib,1 REPO::ARCHIVE data
2715 borg create --compression auto,lzma,6 REPO::ARCHIVE data
2716 borg create --compression auto,lzma ...
2717
2719 There is a borg debug command that has some subcommands which are all
2720 not intended for normal use and potentially very dangerous if used
2721 incorrectly.
2722 For example, borg debug put-obj and borg debug delete-obj will only do
2724 what their name suggests: put objects into repo / delete objects from
2725 repo.
2726 Please note:
2728 · they will not update the chunks cache (chunks index) about the object
2730 · they will not update the manifest (so no automatic chunks index
2732 resync is triggered)
2733 · they will not check whether the object is in use (e.g. before
2735 delete-obj)
2736 · they will not update any metadata which may point to the object
2738 They exist to improve debugging capabilities without direct system
2740 access, e.g. in case you ever run into some severe malfunction. Use
2741 them only if you know what you are doing or if a trusted Borg developer
2742 tells you what to do.
2743 Borg has a --debug-topic TOPIC option to enable specific debugging mes‐
2745 sages. Topics are generally not documented.
2746 A --debug-profile FILE option exists which writes a profile of the main
2748 program's execution to a file. The format of these files is not
2749 directly compatible with the Python profiling tools, since these use
2750 the "marshal" format, which is not intended to be secure (quoting the
2751 Python docs: "Never unmarshal data received from an untrusted or unau‐
2752 thenticated source.").
2753 The borg debug profile-convert command can be used to take a Borg pro‐
2755 file and convert it to a profile file that is compatible with the
2756 Python tools.
2757 Additionally, if the filename specified for --debug-profile ends with
2759 ".pyprof" a Python compatible profile is generated. This is only
2760 intended for local use by developers.
2761
2763 Here are misc. notes about topics that are maybe not covered in enough
2764 detail in the usage section.
2765 --chunker-params
2767 The chunker params influence how input files are cut into pieces
2768 (chunks) which are then considered for deduplication. They also have a
2769 big impact on resource usage (RAM and disk space) as the amount of
2770 resources needed is (also) determined by the total amount of chunks in
2771 the repository (see cache-memory-usage for details).
2772 --chunker-params=10,23,16,4095 results in a fine-grained deduplication|
2774 and creates a big amount of chunks and thus uses a lot of resources to
2775 manage them. This is good for relatively small data volumes and if the
2776 machine has a good amount of free RAM and disk space.
2777 --chunker-params=19,23,21,4095 (default) results in a coarse-grained
2779 deduplication and creates a much smaller amount of chunks and thus uses
2780 less resources. This is good for relatively big data volumes and if the
2781 machine has a relatively low amount of free RAM and disk space.
2782 If you already have made some archives in a repository and you then
2784 change chunker params, this of course impacts deduplication as the
2785 chunks will be cut differently.
2786 In the worst case (all files are big and were touched in between back‐
2788 ups), this will store all content into the repository again.
2789 Usually, it is not that bad though:
2791 · usually most files are not touched, so it will just re-use the old
2793 chunks it already has in the repo
2794 · files smaller than the (both old and new) minimum chunksize result in
2796 only one chunk anyway, so the resulting chunks are same and dedupli‐
2797 cation will apply
2798 If you switch chunker params to save resources for an existing repo
2800 that already has some backup archives, you will see an increasing
2801 effect over time, when more and more files have been touched and stored
2802 again using the bigger chunksize and all references to the smaller
2803 older chunks have been removed (by deleting / pruning archives).
2804 If you want to see an immediate big effect on resource usage, you bet‐
2806 ter start a new repository when changing chunker params.
2807 For more details, see chunker_details.
2809 --noatime / --noctime
2811 You can use these borg create options to not store the respective time‐
2812 stamp into the archive, in case you do not really need it.
2813 Besides saving a little space for the not archived timestamp, it might
2815 also affect metadata stream deduplication: if only this timestamp
2816 changes between backups and is stored into the metadata stream, the
2817 metadata stream chunks won't deduplicate just because of that.
2818 --nobsdflags
2820 You can use this to not query and store (or not extract and set) bsd‐
2821 flags - in case you don't need them or if they are broken somehow for
2822 your fs.
2823 On Linux, dealing with the bsflags needs some additional syscalls.
2825 Especially when dealing with lots of small files, this causes a notica‐
2826 ble overhead, so you can use this option also for speeding up opera‐
2827 tions.
2828 --umask
2830 If you use --umask, make sure that all repository-modifying borg com‐
2831 mands (create, delete, prune) that access the repository in question
2832 use the same --umask value.
2833 If multiple machines access the same repository, this should hold true
2835 for all of them.
2836 --read-special
2838 The --read-special option is special - you do not want to use it for
2839 normal full-filesystem backups, but rather after carefully picking some
2840 targets for it.
2841 The option --read-special triggers special treatment for block and char
2843 device files as well as FIFOs. Instead of storing them as such a device
2844 (or FIFO), they will get opened, their content will be read and in the
2845 backup archive they will show up like a regular file.
2846 Symlinks will also get special treatment if (and only if) they point to
2848 such a special file: instead of storing them as a symlink, the target
2849 special file will get processed as described above.
2850 One intended use case of this is backing up the contents of one or mul‐
2852 tiple block devices, like e.g. LVM snapshots or inactive LVs or disk
2853 partitions.
2854 You need to be careful about what you include when using --read-spe‐
2856 cial, e.g. if you include /dev/zero, your backup will never terminate.
2857 Restoring such files' content is currently only supported one at a time
2859 via --stdout option (and you have to redirect stdout to where ever it
2860 shall go, maybe directly into an existing device file of your choice or
2861 indirectly via dd).
2862 To some extent, mounting a backup archive with the backups of special
2864 files via borg mount and then loop-mounting the image files from inside
2865 the mount point will work. If you plan to access a lot of data in
2866 there, it likely will scale and perform better if you do not work via
2867 the FUSE mount.
2868 Example
2870 Imagine you have made some snapshots of logical volumes (LVs) you want
2871 to backup.
2872 NOTE:
2874 For some scenarios, this is a good method to get "crash-like" con‐
2875 sistency (I call it crash-like because it is the same as you would
2876 get if you just hit the reset button or your machine would abruptly
2877 and completely crash). This is better than no consistency at all
2878 and a good method for some use cases, but likely not good enough if
2879 you have databases running.
2880 Then you create a backup archive of all these snapshots. The backup
2882 process will see a "frozen" state of the logical volumes, while the
2883 processes working in the original volumes continue changing the data
2884 stored there.
2885 You also add the output of lvdisplay to your backup, so you can see the
2887 LV sizes in case you ever need to recreate and restore them.
2888 After the backup has completed, you remove the snapshots again.
2890 $ # create snapshots here
2892 $ lvdisplay > lvdisplay.txt
2893 $ borg create --read-special /path/to/repo::arch lvdisplay.txt /dev/vg0/*-snapshot
2894 $ # remove snapshots here
2895 Now, let's see how to restore some LVs from such a backup.
2897 $ borg extract /path/to/repo::arch lvdisplay.txt
2899 $ # create empty LVs with correct sizes here (look into lvdisplay.txt).
2900 $ # we assume that you created an empty root and home LV and overwrite it now:
2901 $ borg extract --stdout /path/to/repo::arch dev/vg0/root-snapshot > /dev/vg0/root
2902 $ borg extract --stdout /path/to/repo::arch dev/vg0/home-snapshot > /dev/vg0/home
2903 Append-only mode
2905 A repository can be made "append-only", which means that Borg will
2906 never overwrite or delete committed data (append-only refers to the
2907 segment files, but borg will also reject to delete the repository com‐
2908 pletely). This is useful for scenarios where a backup client machine
2909 backups remotely to a backup server using borg serve, since a hacked
2910 client machine cannot delete backups on the server permanently.
2911 To activate append-only mode, set append_only to 1 in the repository
2913 config:
2914 borg config /path/to/repo append_only 1
2916 Note that you can go back-and-forth between normal and append-only
2918 operation with borg config; it's not a "one way trip."
2919 In append-only mode Borg will create a transaction log in the transac‐
2921 tions file, where each line is a transaction and a UTC timestamp.
2922 In addition, borg serve can act as if a repository is in append-only
2924 mode with its option --append-only. This can be very useful for
2925 fine-tuning access control in .ssh/authorized_keys:
2926 command="borg serve --append-only ..." ssh-rsa <key used for not-always-trustable backup clients>
2928 command="borg serve ..." ssh-rsa <key used for backup management>
2929 Running borg init via a borg serve --append-only server will not create
2931 an append-only repository. Running borg init --append-only creates an
2932 append-only repository regardless of server settings.
2933 Example
2935 Suppose an attacker remotely deleted all backups, but your repository
2936 was in append-only mode. A transaction log in this situation might look
2937 like this:
2938 transaction 1, UTC time 2016-03-31T15:53:27.383532
2940 transaction 5, UTC time 2016-03-31T15:53:52.588922
2941 transaction 11, UTC time 2016-03-31T15:54:23.887256
2942 transaction 12, UTC time 2016-03-31T15:55:54.022540
2943 transaction 13, UTC time 2016-03-31T15:55:55.472564
2944 From your security logs you conclude the attacker gained access at
2946 15:54:00 and all the backups where deleted or replaced by compromised
2947 backups. From the log you know that transactions 11 and later are com‐
2948 promised. Note that the transaction ID is the name of the last file in
2949 the transaction. For example, transaction 11 spans files 6 to 11.
2950 In a real attack you'll likely want to keep the compromised repository
2952 intact to analyze what the attacker tried to achieve. It's also a good
2953 idea to make this copy just in case something goes wrong during the
2954 recovery. Since recovery is done by deleting some files, a hard link
2955 copy (cp -al) is sufficient.
2956 The first step to reset the repository to transaction 5, the last
2958 uncompromised transaction, is to remove the hints.N, index.N and
2959 integrity.N files in the repository (these files are always expend‐
2960 able). In this example N is 13.
2961 Then remove or move all segment files from the segment directories in
2963 data/ starting with file 6:
2964 rm data/**/{6..13}
2966 That's all to do in the repository.
2968 If you want to access this rollbacked repository from a client that
2970 already has a cache for this repository, the cache will reflect a newer
2971 repository state than what you actually have in the repository now,
2972 after the rollback.
2973 Thus, you need to clear the cache:
2975 borg delete --cache-only repo
2977 The cache will get rebuilt automatically. Depending on repo size and
2979 archive count, it may take a while.
2980 You also will need to remove ~/.config/borg/security/REPOID/mani‐
2982 fest-timestamp.
2983 Drawbacks
2985 As data is only appended, and nothing removed, commands like prune or
2986 delete won't free disk space, they merely tag data as deleted in a new
2987 transaction.
2988 Be aware that as soon as you write to the repo in non-append-only mode
2990 (e.g. prune, delete or create archives from an admin machine), it will
2991 remove the deleted objects permanently (including the ones that were
2992 already marked as deleted, but not removed, in append-only mode). Auto‐
2993 mated edits to the repository (such as a cron job running borg prune)
2994 will render append-only mode moot if data is deleted.
2995 Even if an archive appears to be available, it is possible an attacker
2997 could delete just a few chunks from an archive and silently corrupt its
2998 data. While in append-only mode, this is reversible, but borg check
2999 should be run before a writing/pruning operation on an append-only
3000 repository to catch accidental or malicious corruption:
3001 # run without append-only mode
3003 borg check --verify-data repo
3004 Aside from checking repository & archive integrity you may want to also
3006 manually check backups to ensure their content seems correct.
3007 Further considerations
3009 Append-only mode is not respected by tools other than Borg. rm still
3010 works on the repository. Make sure that backup client machines only get
3011 to access the repository via borg serve.
3012 Ensure that no remote access is possible if the repository is temporar‐
3014 ily set to normal mode for e.g. regular pruning.
3015 Further protections can be implemented, but are outside of Borg's
3017 scope. For example, file system snapshots or wrapping borg serve to set
3018 special permissions or ACLs on new data files.
3019 SSH batch mode
3021 When running Borg using an automated script, ssh might still ask for a
3022 password, even if there is an SSH key for the target server. Use this
3023 to make scripts more robust:
3024 export BORG_RSH='ssh -oBatchMode=yes'
3026
3028 The Borg Collective (see AUTHORS file)
3029
3031 2010-2021 Jonas Borgström, 2015-2020 The Borg Collective (see AUTHORS
3032 file)
30331.1.16 2021-03-23 BORG(1)