borg/docs/man/borg.1

'\" t
.\" Man page generated from reStructuredText
.\" by the Docutils 0.22.4 manpage writer.
.
.
.nr rst2man-indent-level 0
.
.de1 rstReportMargin
\\$1 \\n[an-margin]
level \\n[rst2man-indent-level]
level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
-
\\n[rst2man-indent0]
\\n[rst2man-indent1]
\\n[rst2man-indent2]
..
.de1 INDENT
.\" .rstReportMargin pre:
. RS \\$1
. nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin]
. nr rst2man-indent-level +1
.\" .rstReportMargin post:
..
.de UNINDENT
. RE
.\" indent \\n[an-margin]
.\" old: \\n[rst2man-indent\\n[rst2man-indent-level]]
.nr rst2man-indent-level -1
.\" new: \\n[rst2man-indent\\n[rst2man-indent-level]]
.in \\n[rst2man-indent\\n[rst2man-indent-level]]u
..
.TH "borg" "1" "2026-08-29" "" "borg backup tool"
.SH Name
borg \- deduplicating and encrypting backup tool
.SH SYNOPSIS
.sp
borg [common options] <command> [options] [arguments]
.SH DESCRIPTION
.\" we don't include the README.rst here since we want to keep this terse.
.
.sp
BorgBackup (short: Borg) is a deduplicating backup program.
Optionally, it supports compression and authenticated encryption.
.sp
The main goal of Borg is to provide an efficient and secure way to back up data.
The data deduplication technique used makes Borg suitable for daily backups
since only changes are stored.
The authenticated encryption technique makes it suitable for backups to targets not
fully trusted.
.sp
Borg stores a set of files in an \fIarchive\fP\&. A \fIrepository\fP is a collection
of \fIarchives\fP\&. The format of repositories is Borg\-specific. Borg does not
distinguish archives from each other in any way other than their name,
it does not matter when or where archives were created (e.g., different hosts).
.SH EXAMPLES
.SS A step\-by\-step example
.INDENT 0.0
.IP 1. 3
Before a backup can be made, a repository has to be initialized:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo repo\-create \-\-encryption=aes256\-ocb
.EE
.UNINDENT
.UNINDENT
.IP 2. 3
Back up the \fB~/src\fP and \fB~/Documents\fP directories into an archive called
\fIdocs\fP:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo create docs ~/src ~/Documents
.EE
.UNINDENT
.UNINDENT
.IP 3. 3
The next day, create a new archive using the same archive name:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo create \-\-stats docs ~/src ~/Documents
.EE
.UNINDENT
.UNINDENT
.sp
This backup will be much quicker and much smaller, since only new,
never\-before\-seen data is stored. The \fB\-\-stats\fP option causes Borg to
output statistics about the newly created archive such as the deduplicated
size (the amount of unique data not shared with other archives):
.INDENT 3.0
.INDENT 3.5
.sp
.EX
Repository: /path/to/repo
Archive name: docs
Archive fingerprint: 74efb3f0c9c05f8b7c822ba859919b480578a3457fd4965c2cee64fbdb631262
Time (nominal): Fri, 2026\-08\-28 12:24:24 +0200
Time (start):   Fri, 2026\-08\-28 12:24:24 +0200
Time (end):     Fri, 2026\-08\-28 12:24:24 +0200
Duration: 0.008 seconds
Number of files: 100
Original size: 1.99 MB
Deduplicated size: 703 B
Time spent in hashing: 0.000 seconds
Time spent in chunking: 0.000 seconds
Added files: 1
Unchanged files: 99
Modified files: 0
Error files: 0
Files changed while reading: 0
\&...
.EE
.UNINDENT
.UNINDENT
.sp
(some more lines with statistics about the repository store accesses follow)
.IP 4. 3
List all archives in the repository (the first column is the beginning of the
archive ID):
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo repo\-list
3affe017  Fri, 2026\-08\-28 12:24:10 +0200  docs                         user        machine
74efb3f0  Fri, 2026\-08\-28 12:24:24 +0200  docs                         user        machine
.EE
.UNINDENT
.UNINDENT
.IP 5. 3
List the contents of the first archive:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo list aid:3affe017
drwxr\-xr\-x user   group         0 Fri, 2026\-08\-28 12:22:30 +0200 home/user/Documents
\-rw\-r\-\-r\-\- user   group      7961 Fri, 2026\-08\-28 12:22:30 +0200 home/user/Documents/Important.doc
\&...
.EE
.UNINDENT
.UNINDENT
.IP 6. 3
Restore the first archive by extracting the files relative to the current directory:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo extract aid:3affe017
.EE
.UNINDENT
.UNINDENT
.IP 7. 3
Delete the first archive (please note that this does \fBnot\fP free repository disk space):
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo delete aid:3affe017
.EE
.UNINDENT
.UNINDENT
.sp
If you use an archive NAME (and not an archive ID), Borg will abort if the name matches multiple
archives (as with the two \fIdocs\fP archives here); use \fBaid:<archive\-id>\fP to delete one specific
archive, or \fB\-a PATTERN\fP to delete multiple archives. Always use \fB\-\-dry\-run\fP and \fB\-\-list\fP first!
.IP 8. 3
Recover disk space by removing objects that are not referenced by any
archive any more:
.INDENT 3.0
.INDENT 3.5
.sp
.EX
$ borg \-r /path/to/repo compact \-v
.EE
.UNINDENT
.UNINDENT
.UNINDENT
.sp
\fBNote:\fP
.INDENT 0.0
.INDENT 3.5
Borg is quiet by default (it defaults to WARNING log level).
You can use options like \fB\-\-progress\fP or \fB\-\-list\fP to get specific
reports during command execution.  You can also add the \fB\-v\fP (or
\fB\-\-verbose\fP or \fB\-\-info\fP) option to adjust the log level to INFO to
get other informational messages.
.UNINDENT
.UNINDENT
.SH NOTES
.SS Positional Arguments and Options: Order matters
.sp
Borg only supports taking options (\fB\-s\fP and \fB\-\-progress\fP in the example)
either to the left or to the right of all positional arguments (\fBarchive\fP and \fBpath\fP
in the example), but not in between them:
.INDENT 0.0
.INDENT 3.5
.sp
.EX
borg create \-s \-\-progress archive path  # good and preferred
borg create archive path \-s \-\-progress  # also works
borg create \-s archive path \-\-progress  # works, but ugly
borg create archive \-s \-\-progress path  # BAD
.EE
.UNINDENT
.UNINDENT
.sp
This is due to a problem in the argparse module: \%<https://\:bugs\:.python\:.org/\:issue15112>
.SS Repository URLs
.sp
\fBLocal filesystem\fP (or locally mounted network filesystem):
.sp
\fB/path/to/repo\fP — filesystem path to the repository directory (absolute path)
.sp
\fBpath/to/repo\fP — filesystem path to the repository directory (relative path)
.sp
Also, paths like \fB~/path/to/repo\fP or \fB~other/path/to/repo\fP work (this is
expanded by your shell).
.sp
Note: You may also prepend \fBfile://\fP to an absolute filesystem path to use URL
style, e.g. \fBfile:///abs/path/to/repo\fP\&. This only works for absolute paths —
\fBfile://rel/path\fP is rejected; use a plain relative path (see above) instead.
.sp
Note: UNC paths (\fB//server/share/path\fP, \fB\e\eserver\eshare\epath\fP) are not
supported — mount the share (on Windows: map it to a drive letter, e.g.
\fBnet use X: \e\eserver\eshare\fP) and use the mounted path instead.
.sp
\fBRemote repositories\fP accessed via SSH \%<user@\:host> (REST http over stdio):
.sp
\fBrest://user@host:port//abs/path/to/repo\fP — absolute path
.sp
\fBrest://user@host:port/rel/path/to/repo\fP — path relative to the remote login directory
.sp
\fBRemote repositories\fP accessed via SSH \%<user@\:host> (legacy borg RPC protocol):
.sp
\fBssh://user@host:port//abs/path/to/repo\fP — absolute path
.sp
\fBssh://user@host:port/rel/path/to/repo\fP — path relative to the remote login directory
.sp
For current (non\-legacy) repositories, \fBssh://\fP is rejected; use \fBrest://\fP
instead, which can also tunnel over ssh (see above). \fBssh://\fP remains available
only for legacy borg 1.x repositories, e.g. via
\fBborg transfer \-\-from\-borg1 \-\-other\-repo ssh://...\fP\&.
.sp
\fBRemote repositories\fP accessed via SFTP:
.sp
\fBsftp://user@host:port//abs/path/to/repo\fP — absolute path
.sp
\fBsftp://user@host:port/rel/path/to/repo\fP — path relative to the remote login directory
.sp
For REST, SSH and SFTP URLs, the \fBuser@\fP and \fB:port\fP parts are optional, but the
path is required: a URL without one, e.g. \fBrest://host\fP or \fBrest://host/\fP, is
rejected. Mind the difference between one and two slashes after the host: one slash
means a path relative to the directory the remote login lands in (usually the remote
user\(aqs home directory), two slashes mean an absolute path.
.sp
\fBRemote repositories\fP accessed directly via HTTP(S), talking to a borgstore REST
server:
.sp
\fBhttp://host:port\fP — plain HTTP
.sp
\fBhttps://user:password@host:port\fP — HTTPS, optionally with credentials embedded in the URL
.sp
For \fBhttp://\fP and \fBhttps://\fP URLs, \fBuser:password@\fP and \fB:port\fP are optional.
Authentication is HTTP Basic auth: credentials come from the URL if given, otherwise
from the \fBBORGSTORE_REST_USERNAME\fP / \fBBORGSTORE_REST_PASSWORD\fP environment
variables; prefer \fBhttps://\fP over \fBhttp://\fP whenever credentials are used, since
Basic auth sends them on every request. A URL path after the host is optional and,
unlike \fBrest://\fP, is not a remote filesystem path — it does not follow the
one\-vs\-two\-slash rule above and is only needed to reach the server through a reverse
proxy mounted below a sub\-path.
.sp
\fBRemote repositories\fP accessed via rclone:
.sp
\fBrclone:remote:path\fP — see the rclone docs for more details about \fBremote:path\fP\&.
.sp
\fBRemote repositories\fP accessed via S3:
.sp
\fB(s3|b2):[(profile|(access_key_id:access_key_secret))@][scheme://hostname[:port]]/bucket/path\fP — see the boto3 docs for more details about credentials.
.sp
If you are connecting to AWS S3, \fB[schema://hostname[:port]]\fP is optional, but \fBbucket\fP and \fBpath\fP are always required.
\fIscheme\fP is usually \fIhttps\fP here, hostname and optional port refer to your S3/B2 server, if that is not Amazon\(aqs.
.sp
Note: There is a known issue with some S3\-compatible services, e.g., Backblaze B2. If you encounter problems, try using \fBb2:\fP instead of \fBs3:\fP in the URL.
.sp
If you frequently need the same repository URL, it is a good idea to set the
\fBBORG_REPO\fP environment variable to set a default repository URL:
.INDENT 0.0
.INDENT 3.5
.sp
.EX
export BORG_REPO=\(aqrest://user@host:port/rel/path/to/repo\(aq
.EE
.UNINDENT
.UNINDENT
.sp
Then simply omit the \fB\-\-repo\fP option when you want
to use the default — it will be read from BORG_REPO.
.SS Repository Locations / Archive Names
.sp
Many commands need to know the repository location; specify it via \fB\-r\fP/\fB\-\-repo\fP
or use the \fBBORG_REPO\fP environment variable.
.sp
Commands that need one or two archive names usually take them as positional arguments.
.sp
Commands that work with an arbitrary number of archives usually accept \fB\-a ARCH_GLOB\fP\&.
.sp
Archive names must not contain the \fB/\fP (slash) character. For simplicity,
also avoid spaces or other characters that have special meaning to the
shell or in a filesystem (\fBborg mount\fP uses the archive name as a directory
name).
.\" This file is the man page variant of docs/usage/usage_general.rst.inc and must
.\" not define the labels that variant defines - otherwise Sphinx warns about
.\" duplicate labels. archive-specification.rst.inc carries its label inside the
.\" file, so skip that line here.
.
.SS Specifying an archive
.sp
How to refer to an archive depends on whether you use archive series or not.
.sp
\fBBy ID:\fP if you use archive series, many or all archives will have the same name, thus
you need to refer to a single archive by its archive ID (see \fBborg repo\-list\fP
output):
.INDENT 0.0
.INDENT 3.5
.sp
.EX
borg info aid:f7dea078
.EE
.UNINDENT
.UNINDENT
.sp
The \fBaid:\fP prefix does a prefix match on the archive ID (the hex representation
of the archive fingerprint). You only need to give enough hex digits to uniquely
identify the archive. This is useful when archive names are ambiguous or when
you want to refer to an archive by its immutable ID.
.sp
\fBBy name:\fP if you don\(aqt use archive series, but do it old\-style by giving every archive
a unique name, you can refer to an archive by its name:
.INDENT 0.0
.INDENT 3.5
.sp
.EX
borg info my\-backup\-202512312359
.EE
.UNINDENT
.UNINDENT
.sp
For more details on archive matching patterns (including shell\-style globs,
regular expressions, and matching by user/host/tags), see \fIborg_match\-archives\fP\&.
.SS Logging
.sp
Borg writes all log output to stderr by default. However, output on stderr does
not necessarily indicate an error. Check the log levels of the messages and the
return code of borg to determine error, warning, or success conditions.
.sp
If you want to capture the log output to a file, just redirect it:
.INDENT 0.0
.INDENT 3.5
.sp
.EX
borg create \-\-repo repo archive myfiles 2>> logfile
.EE
.UNINDENT
.UNINDENT
.sp
Custom logging configurations can be implemented via BORG_LOGGING_CONF.
.sp
The log level of the built\-in logging configuration defaults to WARNING.
This is because we want Borg to be mostly silent and only output
warnings, errors, and critical messages unless output has been requested
by supplying an option that implies output (e.g., \fB\-\-list\fP or \fB\-\-progress\fP).
.sp
Log levels: DEBUG < INFO < WARNING < ERROR < CRITICAL
.sp
Use \fB\-\-debug\fP to set the DEBUG log level —
this prints debug, info, warning, error, and critical messages.
.sp
Use \fB\-\-info\fP (or \fB\-v\fP or \fB\-\-verbose\fP) to set the INFO log level —
this prints info, warning, error, and critical messages.
.sp
Use \fB\-\-warning\fP (default) to set the WARNING log level —
this prints warning, error, and critical messages.
.sp
Use \fB\-\-error\fP to set the ERROR log level —
this prints error and critical messages.
.sp
Use \fB\-\-critical\fP to set the CRITICAL log level —
this prints only critical messages.
.sp
While you can set miscellaneous log levels, do not expect every command to
produce different output at different log levels — it\(aqs merely a possibility.
.sp
\fBWarning:\fP
.INDENT 0.0
.INDENT 3.5
Options \fB\-\-critical\fP and \fB\-\-error\fP are provided for completeness,
their usage is not recommended as you might miss important information.
.UNINDENT
.UNINDENT
.SS Return codes
.sp
Borg can exit with the following return codes (rc):
.TS
box center;
l|l.
T{
Return code
T}	T{
Meaning
T}
_
T{
0
T}	T{
success (logged as INFO)
T}
_
T{
1
T}	T{
generic warning (operation reached its normal end, but there
were warnings \- you should check the log; logged as WARNING)
T}
_
T{
2
T}	T{
generic error (such as a fatal error or a local/remote exception;
the operation did not reach its normal end; logged as ERROR)
T}
_
T{
3..99
T}	T{
specific error (see below; logged as ERROR)
T}
_
T{
100..127
T}	T{
specific warning (see below; logged as WARNING)
T}
_
T{
128+N
T}	T{
terminated by signal N (e.g. 130 == SIGINT, Ctrl+C, or kill \-2;
logged as ERROR)
T}
.TE
.sp
If you use \fB\-\-show\-rc\fP, the return code is also logged at the indicated
level as the last log entry.
.sp
Borg categorizes return codes into groups and exits with the more severe
group: signals (rc 128+N) are more severe than errors (rc 2 and 3..99),
which take precedence over warnings (rc 1 and 100..127), and lastly
success (rc 0).
.sp
Within the signal and error groups, the first signal or error determines
the final return code. Within the warning group, Borg returns the specific
warning code (rc 100..127) if there were one or more warnings of the same
kind. If warnings of different kinds occurred, Borg returns the generic
warning code (rc 1) instead. All errors and warnings are still logged
individually.
.sp
Borg 2 exits with specific error (rc 3..99) and warning (rc 100..127) codes
by default. If you want Borg 2 to always exit with the generic error (rc 2)
or generic warning (rc 1) code instead (like Borg 1 did), set the
\fBBORG_EXIT_CODES=legacy\fP environment variable.
.sp
For a list of all specific error and warning codes, see \fImsgid\fP\&.
.SS Configuration Precedence
.INDENT 0.0
.INDENT 3.5
From lowest to highest:
.INDENT 0.0
.INDENT 3.5
.INDENT 0.0
.IP 1. 3
Defaults defined in the source code.
.IP 2. 3
Default config file (\fB$BORG_CONFIG_DIR/default.yaml\fP).
.IP 3. 3
\fB\-\-config\fP file(s) (in the order given).
.IP 4. 3
Full config environment variable: (\fBBORG_CONFIG\fP).
.IP 5. 3
Environment variables (e.g. \fBBORG_LOG_LEVEL\fP).
.IP 6. 3
Command\-line arguments in order left to right (might include config files).
.UNINDENT
.UNINDENT
.UNINDENT
.UNINDENT
.UNINDENT
.SS Configuration files
.sp
Borg supports reading options from YAML configuration files.  This is
implemented via jsonargparse \%<https://\:jsonargparse\:.readthedocs\:.io/>
and works for all options that can also be set on the command line.
.INDENT 0.0
.TP
.B Default configuration file
\fB$BORG_CONFIG_DIR/default.yaml\fP is loaded automatically on every Borg
invocation if it exists. You do not need to pass \fB\-\-config\fP explicitly
for this file.
.TP
.B \fB\-\-config PATH\fP
Load additional options from the YAML file at \fIPATH\fP\&.
Options in this file take precedence over the default config file but are
overridden by explicit command\-line arguments. This option can be used
multiple times, with later files overriding earlier ones.
.TP
.B \fB\-\-print_config\fP
Print the current effective configuration (all options in YAML format) to
stdout and exit. This reflects the merged result of the default config
file, any \fB\-\-config\fP file, environment variables, and command\-line
arguments. The output can be used as a starting point for a config file.
.sp
\fB\-\-print_config\fP is a common option, so it must be given \fIbefore\fP the
subcommand name — the subcommand\(aqs own options still show up in the output.
.TP
.B File format
Config files are YAML documents. Top\-level keys are option names
(without leading \fB\-\-\fP and with \fB\-\fP replaced by \fB_\fP).
Nested keys correspond to subcommands.
.sp
Example \fBdefault.yaml\fP:
.INDENT 7.0
.INDENT 3.5
.sp
.EX
# apply to all borg commands:
log_level: info
show_rc: true

# options specific to \(dqborg create\(dq:
create:
    compression: zstd,3
    stats: true
.EE
.UNINDENT
.UNINDENT
.sp
The top\-level keys set options that are common to all commands (equivalent
to placing them before the subcommand on the command line).  Keys nested
under a subcommand name (e.g. \fBcreate:\fP) are only applied when that
subcommand is invoked.
.sp
\fBborgfs\fP reads the same config files, but as it has no subcommands, it uses
the top\-level keys and the keys of the \fBmount:\fP section (\fBborgfs\fP is the
\fBborg mount\fP command, and its keys win over the top\-level ones); all other
subcommand sections are ignored.
.UNINDENT
.sp
\fBNote:\fP
.INDENT 0.0
.INDENT 3.5
\fB\-\-print_config\fP shows the merged effective configuration and is a
convenient way to check what values Borg will actually use, and to
generate contents for your borg config file(s):
.INDENT 0.0
.INDENT 3.5
.sp
.EX
borg \-\-repo /backup/main \-\-print_config create \-\-compression zstd,3
.EE
.UNINDENT
.UNINDENT
.UNINDENT
.UNINDENT
.SS Environment Variables
.\" IMPORTANT: this file is auto-generated from borg's built-in help, do not edit!
.
.sp
Borg uses some environment variables for automation:
.INDENT 0.0
.TP
.B General:
.INDENT 7.0
.TP
.B BORG_REPO
When set, use the value to give the default repository location.
Use this so you do not need to type \fB\-\-repo /path/to/my/repo\fP all the time.
.TP
.B BORG_OTHER_REPO
Similar to BORG_REPO, but gives the default for \fB\-\-other\-repo\fP\&.
.TP
.B BORG_PASSPHRASE (and BORG_OTHER_PASSPHRASE)
When set, use the value to answer the passphrase question for encrypted repositories.
It is used when a passphrase is needed to access an encrypted repo as well as when a new
passphrase should be initially set when initializing an encrypted repo.
BORG_PASSPHRASE, BORG_PASSCOMMAND and BORG_PASSPHRASE_FD are mutually exclusive:
if more than one of them is set, borg refuses to guess and aborts with
\(dqMore than one passphrase environment variable is set\(dq. The same applies to the
\fBBORG_OTHER_*\fP variants (which are a separate, independent group).
See also BORG_NEW_PASSPHRASE.
.TP
.B BORG_PASSCOMMAND (and BORG_OTHER_PASSCOMMAND)
When set, use the standard output of the command (trailing newlines are stripped) to answer the
passphrase question for encrypted repositories.
It is used when a passphrase is needed to access an encrypted repo as well as when a new
passphrase should be initially set when initializing an encrypted repo. Note that the command
is executed without a shell. So variables, like \fB$HOME\fP will work, but \fB~\fP won\(aqt.
Mutually exclusive with BORG_PASSPHRASE and BORG_PASSPHRASE_FD, see there.
See also BORG_NEW_PASSPHRASE.
.TP
.B BORG_PASSPHRASE_FD (and BORG_OTHER_PASSPHRASE_FD)
When set, specifies a file descriptor to read a passphrase
from. Programs starting borg may choose to open an anonymous pipe
and use it to pass a passphrase. This is safer than passing via
BORG_PASSPHRASE, because on some systems (e.g. Linux) environment
can be examined by other processes.
Mutually exclusive with BORG_PASSPHRASE and BORG_PASSCOMMAND, see there.
.TP
.B BORG_NEW_PASSPHRASE
When set, use the value to answer the passphrase question when a \fBnew\fP passphrase is asked for.
This variable is checked first. If it is not set, BORG_PASSPHRASE, BORG_PASSCOMMAND and
BORG_PASSPHRASE_FD are checked (in that order).
Main use case for this is to fully automate \fBborg key change\-passphrase\fP\&.
.TP
.B BORG_DISPLAY_PASSPHRASE
When set, use the value to answer the \(dqdisplay the passphrase for verification\(dq question when defining a new passphrase for encrypted repositories.
.TP
.B BORG_DEBUG_PASSPHRASE
When set to YES, display debugging information that includes passphrases used and passphrase related env vars set.
.TP
.B BORG_EXIT_CODES
When set to \(dqmodern\(dq, the borg process will return more specific exit codes (rc).
When set to \(dqlegacy\(dq, the borg process will return rc 2 for all errors, 1 for all warnings, 0 for success.
Default is \(dqmodern\(dq.
.TP
.B BORG_HOST_ID
Borg usually computes a host id from the FQDN plus the results of \fBuuid.getnode()\fP (which usually returns
a unique id based on the MAC address of the network interface. Except if that MAC happens to be all\-zero \- in
that case it returns a random value, which is not what we want (because it kills automatic stale lock removal).
So, if you have an all\-zero MAC address or other reasons to better control the host id externally, just set this
environment variable to a unique value. If all your FQDNs are unique, you can just use the FQDN. If not,
use \%<FQDN@\:uniqueid>\&.
.TP
.B BORG_HOSTNAME
When set, use this value as the hostname (instead of the auto\-detected one), e.g. to run borg
on one host, but impersonate another host. This affects the hostname stored in newly created
archives as well as the \fB{hostname}\fP placeholder.
.TP
.B BORG_USERNAME
When set, use this value as the username (instead of the auto\-detected one), e.g. to run borg
as one user, but impersonate another user. This affects the username stored in newly created
archives as well as the \fB{user}\fP placeholder.
.TP
.B BORG_LOCK_WAIT
You can set the default value for the \fB\-\-lock\-wait\fP option with this, so
you do not need to give it as a command line option.
.TP
.B BORG_LOGGING_CONF
When set, use the given filename as INI\-style logging configuration (see
\%<https://\:docs\:.python\:.org/\:3/\:library/\:logging\:.config\:.html#\:configuration-file-format>).
A basic example conf can be found at \fBdocs/misc/logging.conf\fP\&.
.TP
.B BORG_RSH
When set, use this command instead of \fBssh\fP\&. This can be used to specify ssh options, such as
a custom identity file \fBssh \-i /path/to/private/key\fP\&. See \fBman ssh\fP for other options.
This is the replacement for the removed \fB\-\-rsh CMD\fP command line option.
borg also gives this to borgstore as \fBBORGSTORE_RSH\fP, except if that is already set.
.TP
.B BORG_REMOTE_PATH
When set, use the given path as borg executable on the remote (defaults to \(dqborg\(dq if unset).
This is the replacement for the removed \fB\-\-remote\-path PATH\fP command line option.
.TP
.B BORG_UNITS
Determines how borg formats sizes in its human\-readable output:
.INDENT 7.0
.IP \(bu 2
\fBsi\fP (default): decimal units, e.g. \fB1.23 MB\fP (1kB = 1000B)
.IP \(bu 2
\fBiec\fP: binary units, e.g. \fB1.18 MiB\fP (1KiB = 1024B)
.IP \(bu 2
\fBraw\fP: exact byte counts, e.g. \fB1234567 B\fP
.UNINDENT
.sp
Use \fBraw\fP if you want to parse sizes with scripts (e.g. for monitoring),
so you do not have to deal with scaled values and different units.
Alternatively, use a command\(aqs \fB\-\-json\fP output or, for the commands
supporting \fB\-\-format\fP, the size related format keys \- sizes are given
as byte counts there anyway.
.sp
\fBBORG_UNITS=iec\fP is the replacement for the removed \fBBORG_IEC\fP environment
variable (and for the \fB\-\-iec\fP command line option removed before that).
.TP
.B BORG_PROGRESS_FPS
How often the \fB\-\-progress\fP output is updated at most, in updates per
second (default: 5). Fractional values are allowed, e.g.
\fBBORG_PROGRESS_FPS=0.1\fP limits it to one update every 10 seconds.
Lower values are useful when the output goes into a logfile rather than
to an interactive terminal.
.TP
.B BORG_SPINNER
Controls the spinner borg animates on a terminal while doing work of unknown
duration:
.INDENT 7.0
.IP \(bu 2
unset (default): animate, using Unicode frames if the terminal can display them
.IP \(bu 2
\fBascii\fP: animate, but only use ASCII frames (\fB|/\-\e\fP)
.IP \(bu 2
\fBoff\fP: do not animate, only output the messages next to the spinner
.UNINDENT
.sp
The spinner is animated only on an interactive terminal anyway (and never
with \fB\-\-log\-json\fP), and its colour follows the usual \fBNO_COLOR\fP and
\fBCOLORTERM\fP conventions. See also \fBBORG_PROGRESS_FPS\fP: it also gives
the spinner its frame rate.
.TP
.B BORG_DEBUG_PROFILE
When set to a filename, write an execution profile in Borg format into that file
(see \fIdebugging\fP). If the filename ends with \fB\&.pyprof\fP, a Python\-compatible
profile is written instead.
This is the replacement for the removed \fB\-\-debug\-profile\fP command line option.
Note: every borg invocation writes the profile, so unset it again when you are done.
.TP
.B BORG_REPO_PERMISSIONS
Set repository permissions, see also: \fIborg_serve\fP
.TP
.B BORG_FILES_CACHE_SUFFIX
When set to a value at least one character long, instructs borg to use a specifically named
(based on the suffix) alternative files cache. This can be used to avoid loading and saving
cache entries for backup sources other than the current sources.
.TP
.B BORG_FILES_CACHE_TTL
When set to a numeric value, this determines the maximum \(dqtime to live\(dq for the files cache
entries (default: 2). The files cache is used to determine quickly whether a file is unchanged.
.TP
.B BORG_STORE_CACHE
When set, borg keeps a local writethrough cache of the repository\(aqs \fBpacks/\fP
namespace: on a cache miss the whole pack is fetched once and later reads of the
objects inside that pack are served from the cache. Use this for slow or
high\-latency repositories.
Set it to \fB1\fP to use \fB$BORG_CACHE_DIR/storecache\fP, or to a directory path to
use that directory (it is created if it does not exist). Packs are named by
content hash, so one cache directory can safely hold packs of multiple repositories.
If it is not set, no such caching happens.
.TP
.B BORG_PACK_CACHE_SIZE
When set to a numeric value, limit the pack cache to that many bytes.
Only has an effect if BORG_STORE_CACHE is set.
.TP
.B BORG_PACK_MAX_SIZE
When set to a numeric value, cap packs (the repository objects that batch up many
chunks, see the internals documentation about pack files) at that many bytes
instead of the default of 50000000.
Smaller packs mean more (but smaller) repository objects and more
fine\-grained uploads; bigger packs mean fewer objects and fewer stores.
.TP
.B BORG_PACK_MAX_COUNT
When set to a numeric value, cap packs at that many objects per pack.
If BORG_PACK_MAX_SIZE is not also set, packs are then bound by count only.
.TP
.B BORG_PACK_ASYNC
When set to \fBno\fP, disable the background thread that stores a finished pack
while the next one is being assembled, and store packs synchronously instead.
This is mainly a debugging aid.
.TP
.B BORG_PACK_TRACE
When set to \fByes\fP, print one\-character lifecycle markers of the background
pack store\-thread to stderr (\fB<\fP thread started, \fBH\fP hashing starts,
\fBS\fP storing starts, \fB>\fP thread finished). This is a debugging aid to
visualize how pack stores overlap with the assembly of the next pack.
.TP
.B BORG_ASSERT_ID
Comma\-separated list of the places where borg shall verify that a chunk\(aqs content matches
its chunk id (\fBchunkid == id_hash(content)\fP) after decrypting and decompressing it.
Verifying costs a full hash pass over everything that is read at such a place.
.sp
Default (variable not set):
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_ASSERT_ID=repair,transfer,rechunk
.EE
.UNINDENT
.UNINDENT
.sp
These are the place names that can be listed:
.INDENT 7.0
.TP
.B read
Every read that decompresses a chunk: \fBborg extract\fP, \fBborg mount\fP,
\fBborg export\-tar\fP, \fBborg diff\fP, ... This is by far the most data borg reads, so
this place is \fBnot\fP in the default, see the explanation below.
.TP
.B repair
\fBborg check \-\-repair\fP\&. It rebuilds archives from the item metadata stream it reads,
re\-packing it into new chunks with freshly computed ids, and it recreates manifest and
archives directory entries from what it reads.
.TP
.B transfer
\fBborg transfer\fP, for everything it reads from the source repository. Transferring
re\-anchors the content in another repository, which is a trust boundary.
.TP
.B rechunk
\fBborg recreate \-\-chunker\-params ...\fP, i.e. re\-chunking reads. Re\-chunking computes
new chunk ids from the content it reads, so a violation would not be noticeable any
more afterwards. (Re\-chunking in \fBborg transfer\fP is covered by \fBtransfer\fP\&.)
.UNINDENT
.sp
An unknown place name is an error. An empty value (\fBBORG_ASSERT_ID=\fP) verifies at none of
these places, but still where borg always verifies (see below).
.sp
Why \fBread\fP is not in the default: in the keyed modes, the envelope already authenticates
every read \- the AEAD tag for the encrypted ciphersuites, the MAC for the (unencrypted)
\fBauthenticated\-*\fP modes \- and the chunk id is part of what that tag is computed over. So
a successful decryption resp. tag check already proves that a holder of the borg key
deliberately stored exactly this payload for exactly this chunk id, and a malicious or
buggy \fBrepository\fP can not swap, splice or substitute objects, whether the id is
verified on read or not. What the id check adds is the detection of chunks whose content
does not match their id, which only a malicious or compromised \fBborg client that had
your borg key\fP could have written (e.g. to poison future deduplication). If that is in
your threat model \- e.g. because some machines writing into the repository are not fully
trusted \- add \fBread\fP to the list:
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_ASSERT_ID=read,repair,transfer,rechunk
.EE
.UNINDENT
.UNINDENT
.sp
Otherwise, running \fBborg check \-\-verify\-data\fP periodically is recommended: it is the
audit that re\-certifies the invariant for all chunks in the background, instead of on
every read.
.sp
Independent of this variable, borg always verifies the chunk id:
.INDENT 7.0
.IP \(bu 2
in \fBborg check \-\-verify\-data\fP\&. That audit is what makes not verifying elsewhere
defensible, so it is not configurable (there is no \fBverify_data\fP place name).
.IP \(bu 2
for \fBnone\-*\fP mode repositories: they have no key, so nothing authenticates a read
there and their unkeyed checksums only detect accidental corruption. The id check is
therefore not optional there: it happens at every place, whatever this variable says.
Same for reading borg 1.x repositories (\fBborg transfer\fP).
.UNINDENT
.TP
.B BORG_BLAKE3_MT_THRESHOLD
When set to a numeric value, chunks of at least that many KiB get their id computed by
multi\-threaded BLAKE3, smaller ones single\-threaded (default: 256, i.e. 256KiB).
Only relevant for repositories using \fB\-\-id\-hash blake3\fP\&.
Multi\-threading only pays off for big enough chunks and the break\-even point depends on
the machine\(aqs core count, so the default is deliberately conservative.
Run \fBscripts/blake3\-optimize\-mt\-threshold.py\fP to measure the best value for your
machine \- it sweeps input sizes, prints the recommended threshold and the command to
set it, and can optionally show a chart of the measurements in your browser
(\fB\-\-html \-\-open\fP).
0 means \(dqalways multi\-threaded\(dq, a very large value effectively disables multi\-threading.
.TP
.B BORG_ZSTD_MT_WORKERS
When set to a numeric value, use that many threads to zstd\-compress a single chunk
(default: the cpu count, but at most 4). 0 or 1 means single\-threaded compression.
Only relevant when compressing with \fBzstd\fP\&.
Chunks below 768KiB are always compressed single\-threaded: libzstd will not use a
compression job smaller than 512KiB, so a small chunk gets split very unevenly and
multi\-threading it would be slower than not doing it at all.
The default is capped at 4 because a chunk of the size the default chunker aims at
(2MiB) splits into just 4 such jobs: threads beyond that get (nearly) no work, but
the whole thread pool is created again for every chunk. Measured on a 12\-core
machine, 4 threads beat 12 on every test corpus at the default \fBzstd,\-4\fP
(+13% .. +37%). Raising the value only pays off if you configured the chunker
for much bigger chunks. \fBborg export\-tar\fP compresses one long stream instead of
separate chunks and always defaults to the cpu count.
Multi\-threading trades a little compression ratio for speed (measured at \fBzstd,3\fP:
+0.05% archive size for 1MiB chunks, +0.64% for 8MiB ones, more at higher levels), and
it uses more cpu time in total to reduce the wallclock time. Set it to 1 if you would
rather have the smaller archive, or if borg has to share the cpu with other work.
Single\-threaded can even be faster on data zstd races through anyway, e.g.
already\-compressed/incompressible data or long\-repeat data like VM images.
.TP
.B BORG_FASTCDC_KERNEL / BORG_BUZHASH64_KERNEL
Select the scan kernel the \fBfastcdc\fP / \fBbuzhash64\fP chunker uses. Accepted values
are \fBavx512\fP, \fBavx2\fP, \fBneon\fP, \fBblockwise\fP and \fBscalar\fP\&.
The default is whichever benchmarked fastest for the architecture: \fBneon\fP on
aarch64, and \fBscalar\fP (the plain sequential loop) on x86\-64, where the compiler
folds the rolling hash update into a single instruction and thereby beats the vector
kernels. Other architectures get \fBblockwise\fP, the portable multi\-lane C kernel.
All kernels chunk identically \- same cut points, same chunk ids \- and differ only in
speed, so this is safe to change at any time, also for an existing repository.
Which kernel is fastest is not predictable from the instruction set: it depends on the
cpu and on the compiler that built borg, and the sequential loop wins on some machines.
Measure on your own hardware with \fBborg benchmark cpu \-\-chunking\fP before overriding
the default.
\fBavx512\fP and \fBavx2\fP exist only on x86\-64, \fBneon\fP only on aarch64, and only if the
compiler that built borg supported them; \fBscalar\fP and \fBblockwise\fP are portable C
and always available.
Requesting a kernel that this build or this cpu cannot run is an error rather than a
silent fallback, so a benchmark can not accidentally measure a different kernel.
\fBborg create \-\-debug\fP logs the chunker and the kernel it was created with.
.TP
.B BORG_AES_CHUNKER_KERNEL
Select the scan kernel used by the AES based chunkers \- one variable for all three of
\fBtoeplitz\-aes\fP, \fBrabin\-aes\fP and \fBgoldilocks\-aes\fP\&. Accepted values are \fBvaes\fP,
\fBaes\-ni\fP, \fBaes\-arm64\fP and \fBevp\fP\&.
Unlike the chunker kernels above, wider is simply faster here, so the default is the
best path this build and cpu offer: \fBvaes\fP, else \fBaes\-ni\fP on x86\-64, \fBaes\-arm64\fP
on aarch64, and \fBevp\fP (the portable OpenSSL path) where there is no AES hardware
path.
As with the chunker kernels above, all of them chunk identically and differ only in
speed, and a kernel that can not run here is an error rather than a silent fallback.
\fBvaes\fP and \fBaes\-ni\fP exist only on x86\-64, \fBaes\-arm64\fP only on aarch64.
\fBvaes\fP additionally needs a compiler that knows it (gcc >= 11 / clang >= 14), so a
cpu supporting VAES is not by itself enough to have that kernel available.
.TP
.B BORG_SHOW_SYSINFO
When set to no (default: yes), system information (like OS, Python version, ...) in
exceptions is not shown.
Please only use for good reasons as it makes issues harder to analyze.
.TP
.B BORG_MSGPACK_VERSION_CHECK
Controls whether Borg checks the \fBmsgpack\fP version.
The default is \fByes\fP (strict check). Set to \fBno\fP to disable the version check and
allow any installed \fBmsgpack\fP version. Use this at your own risk; malfunctioning or
incompatible \fBmsgpack\fP versions may cause subtle bugs or repository data corruption.
.TP
.B BORG_FUSE_IMPL
Choose the low\-level FUSE implementation borg shall use for \fBborg mount\fP\&.
This is a comma\-separated list of implementation names, they are tried in the
given order, e.g.:
.INDENT 7.0
.IP \(bu 2
\fBmfusepy,pyfuse3,llfuse\fP: default, first try to load mfusepy, then pyfuse3, then llfuse.
.IP \(bu 2
\fBllfuse,pyfuse3\fP: first try to load llfuse, then try to load pyfuse3.
.IP \(bu 2
\fBmfusepy\fP: only try to load mfusepy
.IP \(bu 2
\fBpyfuse3\fP: only try to load pyfuse3
.IP \(bu 2
\fBllfuse\fP: only try to load llfuse
.IP \(bu 2
\fBnone\fP: do not try to load an implementation
.UNINDENT
.TP
.B BORG_MOUNT_DATA_CACHE_ENTRIES
Number of decrypted file content chunks \fBborg mount\fP and \fBborg webdav\fP keep
in an in\-memory cache, so that the many small, sequential reads a mounted file
system does for a big file do not re\-fetch and re\-decrypt the same chunk over and
over (default: the cpu count). Additional memory usage can be up to the chunk size
times this number.
.TP
.B BORG_SELFTEST
This can be used to influence borg\(aqs built\-in self\-tests. The default is to execute the tests
at the beginning of each borg command invocation.
.sp
BORG_SELFTEST=disabled can be used to switch off the tests and rather save some time.
Disabling is not recommended for normal borg users, but large scale borg storage providers can
use this to optimize production servers after at least doing a one\-time test borg (with
self\-tests not disabled) when installing or upgrading machines/OS/Borg.
.TP
.B BORG_WORKAROUNDS
A list of comma\-separated strings that trigger workarounds in borg,
e.g. to work around bugs in other software.
.sp
Currently known strings are:
.INDENT 7.0
.TP
.B basesyncfile
Use the more simple BaseSyncFile code to avoid issues with sync_file_range.
You might need this to run borg on WSL (Windows Subsystem for Linux) or
in systemd.nspawn containers on some architectures (e.g. ARM).
Using this does not affect data safety, but might result in a more bursty
write\-to\-disk behavior (not continuously streaming to disk).
.TP
.B retry_erofs
Retry opening a file without O_NOATIME if opening a file with O_NOATIME
caused EROFS. You will need this to make archives from volume shadow copies
in WSL1 (Windows Subsystem for Linux 1).
.TP
.B authenticated_no_key
Work around a lost passphrase or a lost borg key for an \fBauthenticated\-*\fP
mode repository (these are only authenticated, but not encrypted).
If a borg key is found \- an object below \fBkeys/\fP in the repository (repokey)
resp. a key file in the keys directory (keyfile) \- it is not unlocked, so the
passphrase does not matter. If no borg key is found at all, borg proceeds
anyway, without any key material.
.sp
Without the key, borg can not verify anything that needs it: neither the
authentication tag of the repository objects nor the chunk ids. It therefore
reads the repository \fBunverified\fP \- a corrupted or tampered repository will
not be detected. (This only concerns the \fBauthenticated\-*\fP modes; the
\fBnone\-*\fP modes need no key and keep verifying their checksums.)
.sp
This workaround is \fBonly\fP for emergencies and \fBonly\fP to extract data
from an affected repository (read\-only access):
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_WORKAROUNDS=authenticated_no_key borg extract \-\-repo repo archive
.EE
.UNINDENT
.UNINDENT
.sp
After you have extracted all data you need, you MUST delete the repository:
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_WORKAROUNDS=authenticated_no_key borg repo\-delete \-\-repo repo
.EE
.UNINDENT
.UNINDENT
.sp
Now you can create a fresh repository with \fBborg repo\-create\fP\&. Make sure you
do not use the workaround any more.
.UNINDENT
.UNINDENT
.TP
.B Output formatting:
.INDENT 7.0
.TP
.B BORG_CHECK_FORMAT
Giving the default value for \fBborg check \-\-format=X\fP\&.
.TP
.B BORG_DIFF_FORMAT
Giving the default value for \fBborg diff \-\-format=X\fP\&.
Note: \fBborg diff \-\-content\-only\fP uses its own format and ignores this.
.TP
.B BORG_FIND_FORMAT
Giving the default value for \fBborg find \-\-format=X\fP\&.
.TP
.B BORG_LIST_FORMAT
Giving the default value for \fBborg list \-\-format=X\fP\&.
.TP
.B BORG_REPO_LIST_FORMAT
Giving the default value for \fBborg repo\-list \-\-format=X\fP\&.
.TP
.B BORG_PRUNE_FORMAT
Giving the default value for \fBborg prune \-\-format=X\fP\&.
.TP
.B BORG_MOUNT_ARCHIVE_DIR_FORMAT
Giving the format of the archive directory names when \fBborg mount\fP or
\fBborg webdav\fP show a whole repository, default: \fB{name}\fP\&. The placeholders
are the ones of \fBborg repo\-list \-\-format\fP; names that are not unique get
\fB\-{id:.8}\fP appended. See \fBborg mount \-\-help\fP\&.
.TP
.B BORG_JSON_INDENT
Indentation of the \fB\-\-json\fP output (default: \fB4\fP).
A number gives that many spaces per nesting level (\fB0\fP still puts every item on
its own line), \fBnone\fP gives compact single\-line JSON, and any other value is used
as the literal indent string (e.g. a tab or the empty string).
.UNINDENT
.TP
.B Some automatic \(dqanswerers\(dq (if set, they automatically answer confirmation questions):
.INDENT 7.0
.TP
.B BORG_UNKNOWN_UNENCRYPTED_REPO_ACCESS_IS_OK=no (or =yes)
For \(dqWarning: Attempting to access a previously unknown unencrypted repository\(dq
.TP
.B BORG_RELOCATED_REPO_ACCESS_IS_OK=no (or =yes)
For \(dqWarning: The repository at location ... was previously located at ...\(dq
.TP
.B BORG_CHECK_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
For \(dqThis is a potentially dangerous function...\(dq (check \-\-repair)
.TP
.B BORG_DELETE_I_KNOW_WHAT_I_AM_DOING=NO (or =YES)
For \(dqYou requested to DELETE the repository completely \fIincluding\fP all archives it contains:\(dq
.UNINDENT
.sp
Note: answers are case sensitive. setting an invalid answer value might either give the default
answer or ask you interactively, depending on whether retries are allowed (they by default are
allowed). So please test your scripts interactively before making them a non\-interactive script.
.TP
.B Directories and files:
Borg 2 uses the platformdirs library (\%<https://\:pypi\:.org/\:project/\:platformdirs/>) to determine
default directory locations. This means that default paths are \fBplatform\-specific\fP:
.INDENT 7.0
.IP \(bu 2
Linux: XDG Base Directory Specification paths are used (e.g. \fB~/.config/borg\fP,
\fB~/.cache/borg\fP, \fB~/.local/share/borg\fP). \fBXDG_*\fP environment variables are
honoured (see \%<https://\:specifications\:.freedesktop\:.org/\:basedir/\:latest/>).
.IP \(bu 2
macOS: native macOS directories are used by default (e.g. \fB~/Library/Application Support/borg\fP,
\fB~/Library/Caches/borg\fP). \fBXDG_*\fP environment variables are honoured if set.
.IP \(bu 2
Windows: local (not roaming) Windows AppData directories are used, e.g.
\fBC:\eUsers\e<user>\eAppData\eLocal\eborg\eborg\fP\&. \fBborg\fP appears twice in that path because
borg does not give platformdirs a separate \(dqapp author\(dq name, so it defaults to the app name.
\fBXDG_*\fP environment variables are \fBnot\fP honoured.
.UNINDENT
.sp
On all platforms, you can override each directory individually using the specific environment
variables described below. You can also set \fBBORG_BASE_DIR\fP to force borg to use
\fBBORG_BASE_DIR/.config/borg\fP, \fBBORG_BASE_DIR/.cache/borg\fP, etc., regardless of the platform.
.sp
Default directory locations by platform (when no \fBBORG_*\fP environment variables are set):
.INDENT 7.0
.INDENT 3.5
.sp
.EX
Directory  Linux                 macOS                                 Windows
Config     ~/.config/borg        ~/Library/Application Support/borg    %LOCALAPPDATA%\eborg\eborg
Cache      ~/.cache/borg         ~/Library/Caches/borg                 %LOCALAPPDATA%\eborg\eborg\eCache
Data       ~/.local/share/borg   ~/Library/Application Support/borg    %LOCALAPPDATA%\eborg\eborg
Runtime    /run/user/<uid>/borg  ~/Library/Caches/TemporaryItems/borg  %LOCALAPPDATA%\eTemp\eborg\eborg
Keys       <config_dir>/keys     <config_dir>/keys                     <config_dir>\ekeys
Security   <data_dir>/security   <data_dir>/security                   <data_dir>\esecurity
.EE
.UNINDENT
.UNINDENT
.INDENT 7.0
.TP
.B BORG_BASE_DIR
Not set by default \- then the platform\-specific directories shown in the table above
are used.
If you want to move all borg\-specific folders to a custom path at once, all you need to do is
to modify \fBBORG_BASE_DIR\fP: the other paths for cache, config etc. will adapt accordingly
(assuming you didn\(aqt set them to a different custom value).
.TP
.B BORG_CACHE_DIR
Defaults to the platform\-specific cache directory (see table above).
If \fBBORG_BASE_DIR\fP is set, defaults to \fB$BORG_BASE_DIR/.cache/borg\fP\&.
On Linux and macOS, \fBXDG_CACHE_HOME\fP is also honoured if \fBBORG_BASE_DIR\fP is not set.
This directory contains the local cache and might need a lot
of space for dealing with big repositories. Make sure you\(aqre aware of the associated
security aspects of the cache location: \fIcache_security\fP
.TP
.B BORG_CONFIG_DIR
Defaults to the platform\-specific config directory (see table above).
If \fBBORG_BASE_DIR\fP is set, defaults to \fB$BORG_BASE_DIR/.config/borg\fP\&.
On Linux and macOS, \fBXDG_CONFIG_HOME\fP is also honoured if \fBBORG_BASE_DIR\fP is not set.
This directory contains all borg configuration directories, see the FAQ
for a security advisory about the data in this directory: \fIhome_config_borg\fP
.TP
.B BORG_DATA_DIR
Defaults to the platform\-specific data directory (see table above).
If \fBBORG_BASE_DIR\fP is set, defaults to \fB$BORG_BASE_DIR/.local/share/borg\fP\&.
On Linux and macOS, \fBXDG_DATA_HOME\fP is also honoured if \fBBORG_BASE_DIR\fP is not set.
This directory contains all borg data directories, see the FAQ
for a security advisory about the data in this directory: \fIhome_data_borg\fP
.TP
.B BORG_RUNTIME_DIR
Defaults to the platform\-specific runtime directory (see table above).
If \fBBORG_BASE_DIR\fP is set, defaults to \fB$BORG_BASE_DIR/.cache/borg\fP\&.
On Linux and macOS, \fBXDG_RUNTIME_DIR\fP is also honoured if \fBBORG_BASE_DIR\fP is not set.
This directory contains borg runtime files, like e.g. the socket file.
.TP
.B BORG_SECURITY_DIR
Defaults to \fB$BORG_DATA_DIR/security\fP\&.
This directory contains security relevant data.
.TP
.B BORG_KEYS_DIR
Defaults to \fB$BORG_CONFIG_DIR/keys\fP\&.
This directory contains keys for encrypted repositories.
.TP
.B BORG_KEY_FILE
When set, use the given path as repository key file. Please note that this is only
for rather special applications that externally fully manage the key files:
.INDENT 7.0
.IP \(bu 2
this setting only applies to the keyfile modes (not to the repokey modes).
.IP \(bu 2
using a full, absolute path to the key file is recommended.
.IP \(bu 2
all directories in the given path must exist.
.IP \(bu 2
this setting forces borg to use the key file at the given location.
.IP \(bu 2
the key file must either exist (for most commands) or will be created (\fBborg repo\-create\fP).
.IP \(bu 2
you need to give a different path for different repositories.
.IP \(bu 2
you need to point to the correct key file matching the repository the command will operate on.
.UNINDENT
.TP
.B TMPDIR
This is where temporary files are stored (might need a lot of temporary space for some
operations), see \%<https://\:docs\:.python\:.org/\:3/\:library/\:tempfile\:.html#\:tempfile\:.gettempdir>
for details.
.UNINDENT
.TP
.B Building:
These are only read by \fBsetup.py\fP while building borg\(aqs C extensions. Each
\fBBORG_*_PREFIX\fP variable names the install prefix of a library that borg links
against: if it is set, \fB$PREFIX/include\fP and \fB$PREFIX/lib\fP are used unconditionally.
If it is not set, the library is located via pkg\-config, and if that does not find it
either, the build fails \- there is no bundled fallback implementation.
.INDENT 7.0
.TP
.B BORG_OPENSSL_PREFIX
Prefix of the OpenSSL installation to build libcrypto against.
On Windows, the libraries are expected in \fB$PREFIX\fP itself rather than in
\fB$PREFIX/lib\fP\&. On OpenBSD, this defaults to \fB/usr/local\fP, pkg\-config is not
used and libcrypto is linked statically (borg needs AES\-OCB via the EVP API, which
LibreSSL does not have).
.TP
.B BORG_OPENSSL_NAME
OpenBSD only: the OpenSSL flavour to use, i.e. the \fBinclude/\fP and \fBlib/\fP
subdirectory name below \fBBORG_OPENSSL_PREFIX\fP (default: \fBeopenssl35\fP).
.TP
.B BORG_LIBLZ4_PREFIX
Prefix of the liblz4 installation to build against.
.TP
.B BORG_LIBACL_PREFIX
Linux only: prefix of the libacl installation to build against.
.UNINDENT
.TP
.B Automatic option environment variables:
Borg uses jsonargparse (\%<https://\:jsonargparse\:.readthedocs\:.io/>) with \fBdefault_env=True\fP,
which means that every command\-line option can also be set via an environment variable.
.sp
The environment variable name is derived from the program name (\fBborg\fP),
the subcommand (if any), and the option name, all converted to uppercase
with dashes replaced by underscores.
.sp
For \fBtop\-level options\fP (not specific to a subcommand), the pattern is:
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_<OPTION>
.EE
.UNINDENT
.UNINDENT
.sp
For example, \fB\-\-lock\-wait\fP can be set via \fBBORG_LOCK_WAIT\fP\&.
.sp
For \fBsubcommand options\fP, the subcommand and option are separated by a
double underscore:
.INDENT 7.0
.INDENT 3.5
.sp
.EX
BORG_<SUBCOMMAND>__<OPTION>
.EE
.UNINDENT
.UNINDENT
.sp
For example, \fBborg create \-\-comment\fP can be set via \fBBORG_CREATE__COMMENT\fP\&.
.UNINDENT
.sp
Please note:
.INDENT 0.0
.IP \(bu 2
Be very careful when using the \(dqyes\(dq sayers, the warnings with prompt exist for your / your data\(aqs security/safety.
.IP \(bu 2
Also be very careful when putting your passphrase into a script, make sure it has appropriate file permissions (e.g.
mode 600, root:root).
.UNINDENT
.SS File systems
.sp
We recommend using a reliable, scalable journaling filesystem for the
repository, e.g., zfs, btrfs, ext4, apfs.
.sp
Borg now uses the \fBborgstore\fP package to implement the key/value store it
uses for the repository.
.sp
For a local repository (or a locally mounted network filesystem) it uses the
\fBfile:\fP store (posixfs backend). For a remote repository it uses the \fBrest:\fP
store, talking to a \fBborg serve \-\-rest\fP process on the remote side (which in
turn uses borgstore there). Other backends (\fBsftp:\fP, \fBrclone:\fP, \fBs3:\fP/
\fBb2:\fP) do not need a filesystem on the repository side at all.
.sp
Borg does not store each chunk as a separate store object. It groups many chunks
into \fIpack files\fP of up to about 50 MB and stores each pack as one object; reading
a single chunk is a partial read at a known offset inside its pack. Thus, the
repository holds far fewer objects than it has chunks (see \fIpacks\fP in the
Internals chapter for the details).
.sp
This has some pros and cons (compared to legacy Borg 1.x segment files):
.sp
Pros:
.INDENT 0.0
.IP \(bu 2
Simplicity and better maintainability of the Borg code.
.IP \(bu 2
The repository is just a key/value store, so it is easy to adapt to other kinds
of storage: borgstore\(aqs backends are quite simple to implement.
\fBsftp:\fP, \fBrclone:\fP and \fBs3:\fP/\fBb2:\fP backends already exist, others might
be easy to add.
.IP \(bu 2
Parallel repository access with less locking is easier to implement.
.UNINDENT
.sp
Cons:
.INDENT 0.0
.IP \(bu 2
Individual chunks cannot be deleted from a pack file; \fBborg compact\fP can only
remove a pack once none of its chunks are referenced any more, so space is not
always reclaimed immediately.
.IP \(bu 2
Greater filesystem space overhead (depends on the allocation block size — modern
filesystems like zfs are rather clever here, using a variable block size).
.IP \(bu 2
Sometimes slower, due to less sequential and more random access operations.
.UNINDENT
.SS Units
.sp
To display quantities, Borg takes care of respecting the
usual conventions of scale. Disk sizes are displayed in decimal \%<https://\:en\:.wikipedia\:.org/\:wiki/\:Decimal>, using powers of ten (so
\fBkB\fP means 1000 bytes). For memory usage, binary prefixes \%<https://\:en\:.wikipedia\:.org/\:wiki/\:Binary_prefix> are used, and are
indicated using the IEC binary prefixes \%<https://\:en\:.wikipedia\:.org/\:wiki/\:Binary_prefix#\:IEC_standard_prefixes>,
using powers of two (so \fBKiB\fP means 1024 bytes).
.SS Date and Time
.sp
We format date and time in accordance with ISO 8601, that is: YYYY\-MM\-DD and
HH:MM:SS (24\-hour clock).
.sp
For more information, see: \%<https://\:xkcd\:.com/\:1179/>
.sp
Unless otherwise noted, we display local date and time.
Internally, we store and process date and time as UTC.
TIMESPAN / INTERVAL
.sp
Some options accept a TIMESPAN or an INTERVAL parameter, which can be given as
a number of years (e.g. \fB2y\fP), months (e.g. \fB12m\fP), weeks (e.g. \fB2w\fP),
days (e.g. \fB7d\fP), hours (e.g. \fB8H\fP), minutes (e.g. \fB30M\fP), or seconds
(e.g. \fB150S\fP).
.sp
The \fBborg prune\fP \fB\-\-keep\-*\fP retention options accept either a plain count
(e.g. \fB\-\-keep\-daily 7\fP, keeping up to 7 daily archives) or a time interval
(e.g. \fB\-\-keep\-daily 7d\fP, keeping one daily archive per day within a 7\-day window).
When using interval\-based retention, \fB\-\-from\fP may be specified to set the
reference timestamp for the interval (defaults to the current time).
.sp
Please note that Borg treats months (e.g. \fB12m\fP) as fixed 31\-day periods
rather than calendar months. As a result, \fB12m\fP corresponds to
12 × 31 = 372 days. Similarly, years (e.g. \fB2y\fP) are treated as fixed
365\-day periods and do not take leap years into account.
.SS Resource Usage
.sp
Borg might use significant resources depending on the size of the data set it is dealing with.
.sp
If you use Borg in a client/server way (with an SSH repository),
the resource usage occurs partly on the client and partly on the
server.
.sp
If you use Borg as a single process (with a filesystem repository),
all resource usage occurs in that one process, so add up client and
server to get the approximate resource usage.
.INDENT 0.0
.TP
.B CPU client:
.INDENT 7.0
.IP \(bu 2
\fBborg create:\fP chunking, hashing, compression, encryption (high CPU usage)
.IP \(bu 2
\fBchunks index rebuild:\fP quite heavy on CPU, doing lots of hash table operations
.IP \(bu 2
\fBborg extract:\fP decryption, decompression (medium to high CPU usage)
.IP \(bu 2
\fBborg prune/borg delete archive:\fP quick, low CPU usage
.IP \(bu 2
\fBborg repo\-delete:\fP low CPU usage, it just removes the repository\(aqs objects
.IP \(bu 2
\fBborg compact:\fP medium CPU usage
.IP \(bu 2
\fBborg check:\fP medium CPU usage, but depends on options given
.UNINDENT
.sp
Most of Borg is single\-threaded, but some parts do use more than one CPU core:
.INDENT 7.0
.IP \(bu 2
zstd compresses a chunk multi\-threaded by default (only for chunks of at
least 768 KiB, using at most 4 threads); see \fBBORG_ZSTD_MT_WORKERS\fP\&.
.IP \(bu 2
the BLAKE3 based id\-hash / MAC modes hash multi\-threaded for inputs from
256 KiB on; see \fBBORG_BLAKE3_MT_THRESHOLD\fP\&.
.IP \(bu 2
\fBborg create\fP hands each finished pack file to a background thread, so
hashing and storing a pack overlaps with processing the next one.
.UNINDENT
.sp
Especially higher zlib and lzma compression levels use significant amounts
of CPU cycles. Crypto might be cheap on the CPU (if hardware\-accelerated) or
expensive (if not).
.TP
.B CPU server:
It usually does not need much CPU; it just deals with the key/value store
(repository).
.sp
borg check: the repository check computes the checksums of all chunks
(medium CPU usage)
borg compact: low to medium CPU usage
.TP
.B CPU (only for client/server operation):
When using Borg in a client/server way with an ssh\-type repository, the SSH
processes used for the transport layer will need some CPU on the client and
on the server due to the crypto they are doing — especially if you are pumping
large amounts of data.
.TP
.B Memory (RAM) client:
The chunks index and the files index are read into memory for performance
reasons. Might need large amounts of memory (see below).
Compression, especially with high compression levels, might need substantial
amounts of memory.
.TP
.B Memory (RAM) server:
Usually rather low memory needs, much less than the client.
.TP
.B Chunks index (client only):
Proportional to the number of data chunks in your repo. Lots of chunks
in your repo imply a big chunks index.
It is possible to tweak the chunker parameters (see create options).
.TP
.B Files index (client only):
Proportional to the number of files in your last backups. Can be switched
off (see create options), but the next backup might be much slower if you do.
The speed benefit of using the files cache is proportional to file size.
.TP
.B Temporary files (client):
TODO
.TP
.B Temporary files (server):
TODO
.TP
.B Cache files (client only):
Contains the files cache, which might become quite large depending on the
amount and size of files.
.TP
.B Network (only for client/server operation):
If your repository is remote, all deduplicated (and optionally compressed/
encrypted) data has to go over the network connection.
.UNINDENT
.SS Support for file metadata
.sp
Besides regular file and directory structures, Borg can preserve
.INDENT 0.0
.IP \(bu 2
symlinks (stored as a symlink; the symlink is not followed)
.IP \(bu 2
special files:
.INDENT 2.0
.IP \(bu 2
character and block device files (restored via mknod(2))
.IP \(bu 2
FIFOs (\(dqnamed pipes\(dq)
.IP \(bu 2
special file \fIcontents\fP can be backed up in \fB\-\-read\-special\fP mode.
By default, the metadata to create them with mknod(2), mkfifo(2), etc. is stored.
.UNINDENT
.IP \(bu 2
hard\-linked regular files, devices, symlinks, FIFOs (considering all items in the same archive)
.IP \(bu 2
timestamps with nanosecond precision: mtime, atime, ctime
.IP \(bu 2
other timestamps: birthtime (on platforms supporting it)
.IP \(bu 2
permissions:
.INDENT 2.0
.IP \(bu 2
IDs of owning user and owning group
.IP \(bu 2
names of owning user and owning group (if the IDs can be resolved)
.IP \(bu 2
Unix Mode/Permissions (u/g/o permissions, suid, sgid, sticky)
.UNINDENT
.UNINDENT
.sp
On some platforms additional features are supported:
.\" Yes/No's are grouped by reason/mechanism/reference.
.
.TS
box center;
l|l|l|l.
T{
Platform
T}	T{
ACLs
[4]
T}	T{
xattr
[5]
T}	T{
Flags
[6]
T}
_
T{
Linux
T}	T{
Yes
T}	T{
Yes
T}	T{
Yes [1]
T}
_
T{
macOS
T}	T{
Yes
T}	T{
Yes
T}	T{
Yes (all)
T}
_
T{
FreeBSD
T}	T{
Yes
T}	T{
Yes
T}	T{
Yes (all)
T}
_
T{
OpenBSD
T}	T{
n/a
T}	T{
n/a
T}	T{
Yes (all)
T}
_
T{
NetBSD
T}	T{
n/a
T}	T{
Yes
T}	T{
Yes (all)
T}
_
T{
Solaris and derivatives
T}	T{
No [2]
T}	T{
Yes
T}	T{
n/a
T}
_
T{
Windows (cygwin)
T}	T{
No [3]
T}	T{
No
T}	T{
No
T}
.TE
.sp
Other Unix\-like operating systems may work as well, but have not been tested yet.
.sp
Note that most platform\-dependent features also depend on the filesystem.
For example, ntfs\-3g on Linux is not able to convey NTFS ACLs.
.IP [1] 5
Only \(dqnodump\(dq, \(dqimmutable\(dq, \(dqcompressed\(dq and \(dqappend\(dq are supported.
Feature request #618 for more flags.
.IP [2] 5
Feature request #1337
.IP [3] 5
Cygwin tries to map NTFS ACLs to permissions with varying degrees of success.
.IP [4] 5
The native access control list mechanism of the OS. This normally limits access to
non\-native ACLs. For example, NTFS ACLs are not completely accessible on Linux with ntfs\-3g.
.IP [5] 5
Extended attributes; key\-value pairs attached to a file, mainly used by the OS.
This includes resource forks on macOS.
.IP [6] 5
Also known as \fIBSD flags\fP\&. The Linux set of flags [1] is portable across platforms.
The BSDs define additional flags.
.SH SEE ALSO
.sp
\fIborg\-common(1)\fP for common command line options
.sp
\fIborg\-repo\-create(1)\fP, \fIborg\-repo\-delete(1)\fP, \fIborg\-repo\-list(1)\fP, \fIborg\-repo\-info(1)\fP,
\fIborg\-create(1)\fP, \fIborg\-mount(1)\fP, \fIborg\-extract(1)\fP,
\fIborg\-list(1)\fP, \fIborg\-info(1)\fP,
\fIborg\-delete(1)\fP, \fIborg\-prune(1)\fP, \fIborg\-compact(1)\fP,
\fIborg\-recreate(1)\fP
.sp
\fIborg\-compression(1)\fP, \fIborg\-patterns(1)\fP, \fIborg\-placeholders(1)\fP, \fIborg\-environment(1)\fP
.INDENT 0.0
.IP \(bu 2
Main web site \%<https://\:www\:.borgbackup\:.org/>
.IP \(bu 2
Releases \%<https://\:github\:.com/\:borgbackup/\:borg/\:releases>
.IP \(bu 2
Changelog \%<https://\:github\:.com/\:borgbackup/\:borg/\:blob/\:master/\:docs/\:changes\:.rst>
.IP \(bu 2
GitHub \%<https://\:github\:.com/\:borgbackup/\:borg>
.IP \(bu 2
Security contact \%<https://\:borgbackup\:.readthedocs\:.io/\:en/\:latest/\:support\:.html#\:security-contact>
.UNINDENT
.SH Author
The Borg Collective

orphan: 
.\" End of generated man page.