1 mailto(rsync-bugs@samba.org)
2 manpage(rsync)(1)(30 Mar 2005)()()
3 manpagename(rsync)(faster, flexible replacement for rcp)
6 rsync [OPTION]... SRC [SRC]... [USER@]HOST:DEST
8 rsync [OPTION]... [USER@]HOST:SRC DEST
10 rsync [OPTION]... SRC [SRC]... DEST
12 rsync [OPTION]... [USER@]HOST::SRC [DEST]
14 rsync [OPTION]... SRC [SRC]... [USER@]HOST::DEST
16 rsync [OPTION]... rsync://[USER@]HOST[:PORT]/SRC [DEST]
18 rsync [OPTION]... SRC [SRC]... rsync://[USER@]HOST[:PORT]/DEST
22 rsync is a program that behaves in much the same way that rcp does,
23 but has many more options and uses the rsync remote-update protocol to
24 greatly speed up file transfers when the destination file is being
27 The rsync remote-update protocol allows rsync to transfer just the
28 differences between two sets of files across the network connection, using
29 an efficient checksum-search algorithm described in the technical
30 report that accompanies this package.
32 Some of the additional features of rsync are:
35 it() support for copying links, devices, owners, groups, and permissions
36 it() exclude and exclude-from options similar to GNU tar
37 it() a CVS exclude mode for ignoring the same files that CVS would ignore
38 it() can use any transparent remote shell, including ssh or rsh
39 it() does not require root privileges
40 it() pipelining of file transfers to minimize latency costs
41 it() support for anonymous or authenticated rsync servers (ideal for
45 manpagesection(GENERAL)
47 There are eight different ways of using rsync. They are:
50 it() for copying local files. This is invoked when neither
51 source nor destination path contains a : separator
52 it() for copying from the local machine to a remote machine using
53 a remote shell program as the transport (such as ssh or
54 rsh). This is invoked when the destination path contains a
56 it() for copying from a remote machine to the local machine
57 using a remote shell program. This is invoked when the source
58 contains a : separator.
59 it() for copying from a remote rsync server to the local
60 machine. This is invoked when the source path contains a ::
61 separator or an rsync:// URL.
62 it() for copying from the local machine to a remote rsync
63 server. This is invoked when the destination path contains a ::
64 separator or an rsync:// URL.
65 it() for copying from a remote machine using a remote shell
66 program as the transport, using rsync server on the remote
67 machine. This is invoked when the source path contains a ::
68 separator and the bf(--rsh=COMMAND) (aka "bf(-e COMMAND)") option is
70 it() for copying from the local machine to a remote machine
71 using a remote shell program as the transport, using rsync
72 server on the remote machine. This is invoked when the
73 destination path contains a :: separator and the
74 bf(--rsh=COMMAND) option is also provided.
75 it() for listing files on a remote machine. This is done the
76 same way as rsync transfers except that you leave off the
80 Note that in all cases (other than listing) at least one of the source
81 and destination paths must be local.
85 See the file README for installation instructions.
87 Once installed, you can use rsync to any machine that you can access via
88 a remote shell (as well as some that you can access using the rsync
89 daemon-mode protocol). For remote transfers, a modern rsync uses ssh
90 for its communications, but it may have been configured to use a
91 different remote shell by default, such as rsh or remsh.
93 You can also specify any remote shell you like, either by using the bf(-e)
94 command line option, or by setting the RSYNC_RSH environment variable.
96 One common substitute is to use ssh, which offers a high degree of
99 Note that rsync must be installed on both the source and destination
102 manpagesection(USAGE)
104 You use rsync in the same way you use rcp. You must specify a source
105 and a destination, one of which may be remote.
107 Perhaps the best way to explain the syntax is with some examples:
109 quote(tt(rsync -t *.c foo:src/))
111 This would transfer all files matching the pattern *.c from the
112 current directory to the directory src on the machine foo. If any of
113 the files already exist on the remote system then the rsync
114 remote-update protocol is used to update the file by sending only the
115 differences. See the tech report for details.
117 quote(tt(rsync -avz foo:src/bar /data/tmp))
119 This would recursively transfer all files from the directory src/bar on the
120 machine foo into the /data/tmp/bar directory on the local machine. The
121 files are transferred in "archive" mode, which ensures that symbolic
122 links, devices, attributes, permissions, ownerships, etc. are preserved
123 in the transfer. Additionally, compression will be used to reduce the
124 size of data portions of the transfer.
126 quote(tt(rsync -avz foo:src/bar/ /data/tmp))
128 A trailing slash on the source changes this behavior to avoid creating an
129 additional directory level at the destination. You can think of a trailing
130 / on a source as meaning "copy the contents of this directory" as opposed
131 to "copy the directory by name", but in both cases the attributes of the
132 containing directory are transferred to the containing directory on the
133 destination. In other words, each of the following commands copies the
134 files in the same way, including their setting of the attributes of
138 tt(rsync -av /src/foo /dest)nl()
139 tt(rsync -av /src/foo/ /dest/foo)nl()
142 Note also that host and module references don't require a trailing slash to
143 copy the contents of the default directory. For example, both of these
144 copy the remote directory's contents into "/dest":
147 tt(rsync -av host: /dest)nl()
148 tt(rsync -av host::module /dest)nl()
151 You can also use rsync in local-only mode, where both the source and
152 destination don't have a ':' in the name. In this case it behaves like
153 an improved copy command.
155 quote(tt(rsync somehost.mydomain.com::))
157 This would list all the anonymous rsync modules available on the host
158 somehost.mydomain.com. (See the following section for more details.)
160 manpagesection(ADVANCED USAGE)
162 The syntax for requesting multiple files from a remote host involves using
163 quoted spaces in the SRC. Some examples:
165 quote(tt(rsync host::'modname/dir1/file1 modname/dir2/file2' /dest))
167 This would copy file1 and file2 into /dest from an rsync daemon. Each
168 additional arg must include the same "modname/" prefix as the first one,
169 and must be preceded by a single space. All other spaces are assumed
170 to be a part of the filenames.
172 quote(tt(rsync -av host:'dir1/file1 dir2/file2' /dest))
174 This would copy file1 and file2 into /dest using a remote shell. This
175 word-splitting is done by the remote shell, so if it doesn't work it means
176 that the remote shell isn't configured to split its args based on
177 whitespace (a very rare setting, but not unknown). If you need to transfer
178 a filename that contains whitespace, you'll need to either escape the
179 whitespace in a way that the remote shell will understand, or use wildcards
180 in place of the spaces. Two examples of this are:
183 tt(rsync -av host:'file\ name\ with\ spaces' /dest)nl()
184 tt(rsync -av host:file?name?with?spaces /dest)nl()
187 This latter example assumes that your shell passes through unmatched
188 wildcards. If it complains about "no match", put the name in quotes.
190 manpagesection(CONNECTING TO AN RSYNC SERVER)
192 It is also possible to use rsync without a remote shell as the
193 transport. In this case you will connect to a remote rsync server
194 running on TCP port 873.
196 You may establish the connection via a web proxy by setting the
197 environment variable RSYNC_PROXY to a hostname:port pair pointing to
198 your web proxy. Note that your web proxy's configuration must support
199 proxy connections to port 873.
201 Using rsync in this way is the same as using it with a remote shell except
205 it() you either use a double colon :: instead of a single colon to
206 separate the hostname from the path, or you use an rsync:// URL.
207 it() the remote server may print a message of the day when you
209 it() if you specify no path name on the remote server then the
210 list of accessible paths on the server will be shown.
211 it() if you specify no local destination then a listing of the
212 specified files on the remote server is provided.
215 Some paths on the remote server may require authentication. If so then
216 you will receive a password prompt when you connect. You can avoid the
217 password prompt by setting the environment variable RSYNC_PASSWORD to
218 the password you want to use or using the bf(--password-file) option. This
219 may be useful when scripting rsync.
221 WARNING: On some systems environment variables are visible to all
222 users. On those systems using bf(--password-file) is recommended.
224 manpagesection(CONNECTING TO AN RSYNC SERVER OVER A REMOTE SHELL PROGRAM)
226 It is sometimes useful to be able to set up file transfers using rsync
227 server capabilities on the remote machine, while still using ssh or
228 rsh for transport. This is especially useful when you want to connect
229 to a remote machine via ssh (for encryption or to get through a
230 firewall), but you still want to have access to the rsync server
231 features (see RUNNING AN RSYNC SERVER OVER A REMOTE SHELL PROGRAM,
234 From the user's perspective, using rsync in this way is the same as
235 using it to connect to an rsync server, except that you must
236 explicitly set the remote shell program on the command line with
237 bf(--rsh=COMMAND). (Setting RSYNC_RSH in the environment will not turn on
240 In order to distinguish between the remote-shell user and the rsync
241 server user, you can use '-l user' on your remote-shell command:
243 verb( rsync -av --rsh="ssh -l ssh-user" \
244 rsync-user@host::module[/path] local-path)
246 The "ssh-user" will be used at the ssh level; the "rsync-user" will be
247 used to check against the rsyncd.conf on the remote host.
249 manpagesection(RUNNING AN RSYNC SERVER)
251 An rsync server is configured using a configuration file. Please see the
252 rsyncd.conf(5) man page for more information. By default the configuration
253 file is called /etc/rsyncd.conf, unless rsync is running over a remote
254 shell program and is not running as root; in that case, the default name
255 is rsyncd.conf in the current directory on the remote computer
258 manpagesection(RUNNING AN RSYNC SERVER OVER A REMOTE SHELL PROGRAM)
260 See the rsyncd.conf(5) man page for full information on the rsync
261 server configuration file.
263 Several configuration options will not be available unless the remote
264 user is root (e.g. chroot, setuid/setgid, etc.). There is no need to
265 configure inetd or the services map to include the rsync server port
266 if you run an rsync server only via a remote shell program.
268 To run an rsync server out of a single-use ssh key, see this section
269 in the rsyncd.conf(5) man page.
271 manpagesection(EXAMPLES)
273 Here are some examples of how I use rsync.
275 To backup my wife's home directory, which consists of large MS Word
276 files and mail folders, I use a cron job that runs
278 quote(tt(rsync -Cavz . arvidsjaur:backup))
280 each night over a PPP connection to a duplicate directory on my machine
283 To synchronize my samba source trees I use the following Makefile
287 rsync -avuzb --exclude '*~' samba:samba/ .
289 rsync -Cavuzb . samba:samba/
292 this allows me to sync with a CVS directory at the other end of the
293 connection. I then do CVS operations on the remote machine, which saves a
294 lot of time as the remote CVS protocol isn't very efficient.
296 I mirror a directory between my "old" and "new" ftp sites with the
299 tt(rsync -az -e ssh --delete ~ftp/pub/samba nimbus:"~ftp/pub/tridge")
301 This is launched from cron every few hours.
303 manpagesection(OPTIONS SUMMARY)
305 Here is a short summary of the options available in rsync. Please refer
306 to the detailed description below for a complete description. verb(
307 -v, --verbose increase verbosity
308 -q, --quiet suppress non-error messages
309 -c, --checksum skip based on checksum, not mod-time & size
310 -a, --archive archive mode; same as -rlptgoD (no -H)
311 -r, --recursive recurse into directories
312 -R, --relative use relative path names
313 --no-relative turn off --relative
314 --no-implied-dirs don't send implied dirs with -R
315 -b, --backup make backups (see --suffix & --backup-dir)
316 --backup-dir=DIR make backups into hierarchy based in DIR
317 --suffix=SUFFIX backup suffix (default ~ w/o --backup-dir)
318 -u, --update skip files that are newer on the receiver
319 --inplace update destination files in-place
320 -d, --dirs transfer directories without recursing
321 -l, --links copy symlinks as symlinks
322 -L, --copy-links transform symlink into referent file/dir
323 --copy-unsafe-links only "unsafe" symlinks are transformed
324 --safe-links ignore symlinks that point outside the tree
325 -H, --hard-links preserve hard links
326 -K, --keep-dirlinks treat symlinked dir on receiver as dir
327 -p, --perms preserve permissions
328 -o, --owner preserve owner (root only)
329 -g, --group preserve group
330 -D, --devices preserve devices (root only)
331 -t, --times preserve times
332 -O, --omit-dir-times omit directories when preserving times
333 -S, --sparse handle sparse files efficiently
334 -n, --dry-run show what would have been transferred
335 -W, --whole-file copy files whole (without rsync algorithm)
336 --no-whole-file always use incremental rsync algorithm
337 -x, --one-file-system don't cross filesystem boundaries
338 -B, --block-size=SIZE force a fixed checksum block-size
339 -e, --rsh=COMMAND specify the remote shell to use
340 --rsync-path=PROGRAM specify the rsync to run on remote machine
341 --existing only update files that already exist
342 --ignore-existing ignore files that already exist on receiver
343 --remove-sent-files sent files/symlinks are removed from sender
344 --del an alias for --delete-during
345 --delete delete files that don't exist on sender
346 --delete-before receiver deletes before transfer (default)
347 --delete-during receiver deletes during xfer, not before
348 --delete-after receiver deletes after transfer, not before
349 --delete-excluded also delete excluded files on receiver
350 --ignore-errors delete even if there are I/O errors
351 --force force deletion of dirs even if not empty
352 --max-delete=NUM don't delete more than NUM files
353 --max-size=SIZE don't transfer any file larger than SIZE
354 --partial keep partially transferred files
355 --partial-dir=DIR put a partially transferred file into DIR
356 --delay-updates put all updated files into place at end
357 --numeric-ids don't map uid/gid values by user/group name
358 --timeout=TIME set I/O timeout in seconds
359 -I, --ignore-times don't skip files that match size and time
360 --size-only skip files that match in size
361 --modify-window=NUM compare mod-times with reduced accuracy
362 -T, --temp-dir=DIR create temporary files in directory DIR
363 -y, --fuzzy find similar file for basis if no dest file
364 --compare-dest=DIR also compare received files relative to DIR
365 --copy-dest=DIR ... and include copies of unchanged files
366 --link-dest=DIR hardlink to files in DIR when unchanged
367 -z, --compress compress file data during the transfer
368 -C, --cvs-exclude auto-ignore files in the same way CVS does
369 -f, --filter=RULE add a file-filtering RULE
370 -F same as --filter='dir-merge /.rsync-filter'
371 repeated: --filter='- .rsync-filter'
372 --exclude=PATTERN exclude files matching PATTERN
373 --exclude-from=FILE read exclude patterns from FILE
374 --include=PATTERN don't exclude files matching PATTERN
375 --include-from=FILE read include patterns from FILE
376 --files-from=FILE read list of source-file names from FILE
377 -0, --from0 all *from/filter files are delimited by 0s
378 --version print version number
379 --address=ADDRESS bind address for outgoing socket to daemon
380 --port=PORT specify double-colon alternate port number
381 --blocking-io use blocking I/O for the remote shell
382 --no-blocking-io turn off blocking I/O when it is default
383 --stats give some file-transfer stats
384 --progress show progress during transfer
385 -P same as --partial --progress
386 -i, --itemize-changes output a change-summary for all updates
387 --log-format=FORMAT log file-transfers using specified format
388 --password-file=FILE read password from FILE
389 --list-only list the files instead of copying them
390 --bwlimit=KBPS limit I/O bandwidth; KBytes per second
391 --write-batch=FILE write a batched update to FILE
392 --only-write-batch=FILE like --write-batch but w/o updating dest
393 --read-batch=FILE read a batched update from FILE
394 --protocol=NUM force an older protocol version to be used
395 --checksum-seed=NUM set block/file checksum seed (advanced)
396 -4, --ipv4 prefer IPv4
397 -6, --ipv6 prefer IPv6
398 -h, --help show this help screen)
400 Rsync can also be run as a daemon, in which case the following options are
402 --daemon run as an rsync daemon
403 --address=ADDRESS bind to the specified address
404 --bwlimit=KBPS limit I/O bandwidth; KBytes per second
405 --config=FILE specify alternate rsyncd.conf file
406 --no-detach do not detach from the parent
407 --port=PORT listen on alternate port number
408 -v, --verbose increase verbosity
409 -4, --ipv4 prefer IPv4
410 -6, --ipv6 prefer IPv6
411 -h, --help show this help screen)
415 rsync uses the GNU long options package. Many of the command line
416 options have two variants, one short and one long. These are shown
417 below, separated by commas. Some options only have a long variant.
418 The '=' for options that take a parameter is optional; whitespace
422 dit(bf(-h, --help)) Print a short help page describing the options
425 dit(bf(--version)) print the rsync version number and exit.
427 dit(bf(-v, --verbose)) This option increases the amount of information you
428 are given during the transfer. By default, rsync works silently. A
429 single bf(-v) will give you information about what files are being
430 transferred and a brief summary at the end. Two bf(-v) flags will give you
431 information on what files are being skipped and slightly more
432 information at the end. More than two bf(-v) flags should only be used if
433 you are debugging rsync.
435 Note that the names of the transferred files that are output are done using
436 a default bf(--log-format) of "%n%L", which tells you just the name of the
437 file and, if the item is a symlink, where it points. At the single bf(-v)
438 level of verbosity, this does not mention when a file gets its attributes
439 changed. If you ask for an itemized list of changed attributes (either
440 bf(--itemize-changes) or adding "%i" to the bf(--log-format) setting), the
441 output (on the client) increases to mention all items that are changed in
442 any way. See the bf(--log-format) option for more details.
444 dit(bf(-q, --quiet)) This option decreases the amount of information you
445 are given during the transfer, notably suppressing information messages
446 from the remote server. This flag is useful when invoking rsync from
449 dit(bf(-I, --ignore-times)) Normally rsync will skip any files that are
450 already the same size and have the same modification time-stamp.
451 This option turns off this "quick check" behavior.
453 dit(bf(--size-only)) Normally rsync will not transfer any files that are
454 already the same size and have the same modification time-stamp. With the
455 bf(--size-only) option, files will not be transferred if they have the same size,
456 regardless of timestamp. This is useful when starting to use rsync
457 after using another mirroring system which may not preserve timestamps
460 dit(bf(--modify-window)) When comparing two timestamps, rsync treats the
461 timestamps as being equal if they differ by no more than the modify-window
462 value. This is normally 0 (for an exact match), but you may find it useful
463 to set this to a larger value in some situations. In particular, when
464 transferring to or from an MS Windows FAT filesystem (which represents
465 times with a 2-second resolution), bf(--modify-window=1) is useful
466 (allowing times to differ by up to 1 second).
468 dit(bf(-c, --checksum)) This forces the sender to checksum all files using
469 a 128-bit MD4 checksum before transfer. The checksum is then
470 explicitly checked on the receiver and any files of the same name
471 which already exist and have the same checksum and size on the
472 receiver are not transferred. This option can be quite slow.
474 dit(bf(-a, --archive)) This is equivalent to bf(-rlptgoD). It is a quick
475 way of saying you want recursion and want to preserve almost
476 everything. The only exception to this is if bf(--files-from) was
477 specified, in which case bf(-r) is not implied.
479 Note that bf(-a) bf(does not preserve hardlinks), because
480 finding multiply-linked files is expensive. You must separately
483 dit(bf(-r, --recursive)) This tells rsync to copy directories
484 recursively. See also bf(--dirs) (bf(-d)).
486 dit(bf(-R, --relative)) Use relative paths. This means that the full path
487 names specified on the command line are sent to the server rather than
488 just the last parts of the filenames. This is particularly useful when
489 you want to send several different directories at the same time. For
490 example, if you used the command
492 quote(tt( rsync /foo/bar/foo.c remote:/tmp/))
494 then this would create a file called foo.c in /tmp/ on the remote
495 machine. If instead you used
497 quote(tt( rsync -R /foo/bar/foo.c remote:/tmp/))
499 then a file called /tmp/foo/bar/foo.c would be created on the remote
500 machine -- the full path name is preserved. To limit the amount of
501 path information that is sent, do something like this:
505 tt( rsync -R bar/foo.c remote:/tmp/)nl()
508 That would create /tmp/bar/foo.c on the remote machine.
510 dit(bf(--no-relative)) Turn off the bf(--relative) option. This is only
511 needed if you want to use bf(--files-from) without its implied bf(--relative)
514 dit(bf(--no-implied-dirs)) When combined with the bf(--relative) option, the
515 implied directories in each path are not explicitly duplicated as part
516 of the transfer. This makes the transfer more optimal and also allows
517 the two sides to have non-matching symlinks in the implied part of the
518 path. For instance, if you transfer the file "/path/foo/file" with bf(-R),
519 the default is for rsync to ensure that "/path" and "/path/foo" on the
520 destination exactly match the directories/symlinks of the source. Using
521 the bf(--no-implied-dirs) option would omit both of these implied dirs,
522 which means that if "/path" was a real directory on one machine and a
523 symlink of the other machine, rsync would not try to change this.
525 dit(bf(-b, --backup)) With this option, preexisting destination files are
526 renamed as each file is transferred or deleted. You can control where the
527 backup file goes and what (if any) suffix gets appended using the
528 bf(--backup-dir) and bf(--suffix) options.
529 Note that if you don't specify bf(--backup-dir), the bf(--omit-dir-times)
530 option will be enabled.
532 dit(bf(--backup-dir=DIR)) In combination with the bf(--backup) option, this
533 tells rsync to store all backups in the specified directory. This is
534 very useful for incremental backups. You can additionally
535 specify a backup suffix using the bf(--suffix) option
536 (otherwise the files backed up in the specified directory
537 will keep their original filenames).
539 dit(bf(--suffix=SUFFIX)) This option allows you to override the default
540 backup suffix used with the bf(--backup) (bf(-b)) option. The default suffix is a ~
541 if no -bf(-backup-dir) was specified, otherwise it is an empty string.
543 dit(bf(-u, --update)) This forces rsync to skip any files which exist on
544 the destination and have a modified time that is newer than the source
545 file. (If an existing destination file has a modify time equal to the
546 source file's, it will be updated if the sizes are different.)
548 In the current implementation of bf(--update), a difference of file format
549 between the sender and receiver is always
550 considered to be important enough for an update, no matter what date
551 is on the objects. In other words, if the source has a directory or a
552 symlink where the destination has a file, the transfer would occur
553 regardless of the timestamps. This might change in the future (feel
554 free to comment on this on the mailing list if you have an opinion).
556 dit(bf(--inplace)) This causes rsync not to create a new copy of the file
557 and then move it into place. Instead rsync will overwrite the existing
558 file, meaning that the rsync algorithm can't accomplish the full amount of
559 network reduction it might be able to otherwise (since it does not yet try
560 to sort data matches). One exception to this is if you combine the option
561 with bf(--backup), since rsync is smart enough to use the backup file as the
562 basis file for the transfer.
564 This option is useful for transfer of large files with block-based changes
565 or appended data, and also on systems that are disk bound, not network
568 The option implies bf(--partial) (since an interrupted transfer does not delete
569 the file), but conflicts with bf(--partial-dir) and bf(--delay-updates).
570 Prior to rsync 2.6.4 bf(--inplace) was also incompatible with bf(--compare-dest)
573 WARNING: The file's data will be in an inconsistent state during the
574 transfer (and possibly afterward if the transfer gets interrupted), so you
575 should not use this option to update files that are in use. Also note that
576 rsync will be unable to update a file in-place that is not writable by the
579 dit(bf(-d, --dirs)) Tell the sending side to include any directories that
580 are encountered. Unlike bf(--recursive), a directory's contents are not copied
581 unless the directory was specified on the command-line as either "." or a
582 name with a trailing slash (e.g. "foo/"). Without this option or the
583 bf(--recursive) option, rsync will skip all directories it encounters (and
584 output a message to that effect for each one).
586 dit(bf(-l, --links)) When symlinks are encountered, recreate the
587 symlink on the destination.
589 dit(bf(-L, --copy-links)) When symlinks are encountered, the file that
590 they point to (the referent) is copied, rather than the symlink. In older
591 versions of rsync, this option also had the side-effect of telling the
592 receiving side to follow symlinks, such as symlinks to directories. In a
593 modern rsync such as this one, you'll need to specify bf(--keep-dirlinks) (bf(-K))
594 to get this extra behavior. The only exception is when sending files to
595 an rsync that is too old to understand bf(-K) -- in that case, the bf(-L) option
596 will still have the side-effect of bf(-K) on that older receiving rsync.
598 dit(bf(--copy-unsafe-links)) This tells rsync to copy the referent of
599 symbolic links that point outside the copied tree. Absolute symlinks
600 are also treated like ordinary files, and so are any symlinks in the
601 source path itself when bf(--relative) is used.
603 dit(bf(--safe-links)) This tells rsync to ignore any symbolic links
604 which point outside the copied tree. All absolute symlinks are
605 also ignored. Using this option in conjunction with bf(--relative) may
606 give unexpected results.
608 dit(bf(-H, --hard-links)) This tells rsync to recreate hard links on
609 the remote system to be the same as the local system. Without this
610 option hard links are treated like regular files.
612 Note that rsync can only detect hard links if both parts of the link
613 are in the list of files being sent.
615 This option can be quite slow, so only use it if you need it.
617 dit(bf(-K, --keep-dirlinks)) On the receiving side, if a symlink is
618 pointing to a directory, it will be treated as matching a directory
621 dit(bf(-W, --whole-file)) With this option the incremental rsync algorithm
622 is not used and the whole file is sent as-is instead. The transfer may be
623 faster if this option is used when the bandwidth between the source and
624 destination machines is higher than the bandwidth to disk (especially when the
625 "disk" is actually a networked filesystem). This is the default when both
626 the source and destination are specified as local paths.
628 dit(bf(--no-whole-file)) Turn off bf(--whole-file), for use when it is the
631 dit(bf(-p, --perms)) This option causes rsync to set the destination
632 permissions to be the same as the source permissions.
634 Without this option, all existing files (including updated files) retain
635 their existing permissions, while each new file gets its permissions set
636 based on the source file's permissions, but masked by the receiving end's
638 (which is the same behavior as other file-copy utilities, such as cp).
640 dit(bf(-o, --owner)) This option causes rsync to set the owner of the
641 destination file to be the same as the source file. On most systems,
642 only the super-user can set file ownership. By default, the preservation
643 is done by name, but may fall back to using the ID number in some
644 circumstances. See the bf(--numeric-ids) option for a full discussion.
646 dit(bf(-g, --group)) This option causes rsync to set the group of the
647 destination file to be the same as the source file. If the receiving
648 program is not running as the super-user, only groups that the
649 receiver is a member of will be preserved. By default, the preservation
650 is done by name, but may fall back to using the ID number in some
651 circumstances. See the bf(--numeric-ids) option for a full discussion.
653 dit(bf(-D, --devices)) This option causes rsync to transfer character and
654 block device information to the remote system to recreate these
655 devices. This option is only available to the super-user.
657 dit(bf(-t, --times)) This tells rsync to transfer modification times along
658 with the files and update them on the remote system. Note that if this
659 option is not used, the optimization that excludes files that have not been
660 modified cannot be effective; in other words, a missing bf(-t) or bf(-a) will
661 cause the next transfer to behave as if it used bf(-I), causing all files to be
662 updated (though the rsync algorithm will make the update fairly efficient
663 if the files haven't actually changed, you're much better off using bf(-t)).
665 dit(bf(-O, --omit-dir-times)) This tells rsync to omit directories when
666 it is preserving modification times (see bf(--times)). If NFS is sharing
667 the directories on the receiving side, it is a good idea to use bf(-O).
668 This option is inferred if you use bf(--backup) without bf(--backup-dir).
670 dit(bf(-n, --dry-run)) This tells rsync to not do any file transfers,
671 instead it will just report the actions it would have taken.
673 dit(bf(-S, --sparse)) Try to handle sparse files efficiently so they take
674 up less space on the destination.
676 NOTE: Don't use this option when the destination is a Solaris "tmpfs"
677 filesystem. It doesn't seem to handle seeks over null regions
678 correctly and ends up corrupting the files.
680 dit(bf(-x, --one-file-system)) This tells rsync not to cross filesystem
681 boundaries when recursing. This is useful for transferring the
682 contents of only one filesystem.
684 dit(bf(--existing)) This tells rsync not to create any new files --
685 only update files that already exist on the destination.
687 dit(bf(--ignore-existing))
688 This tells rsync not to update files that already exist on
691 dit(bf(--remove-sent-files)) This tells rsync to remove from the sending
692 side the files and/or symlinks that are newly created or whose content is
693 updated on the receiving side. Directories and devices are not removed,
694 nor are files/symlinks whose attributes are merely changed.
696 dit(bf(--delete)) This tells rsync to delete extraneous files from the
697 receiving side (ones that aren't on the sending side), but only for the
698 directories that are being synchronized. You must have asked rsync to
699 send the whole directory (e.g. "dir" or "dir/") without using a wildcard
700 for the directory's contents (e.g. "dir/*") since the wildcard is expanded
701 by the shell and rsync thus gets a request to transfer individual files, not
702 the files' parent directory. Files that are excluded from transfer are
703 also excluded from being deleted unless you use the bf(--delete-excluded)
704 option or mark the rules as only matching on the sending side (see the
705 include/exclude modifiers in the FILTER RULES section).
707 This option has no effect unless directory recursion is enabled.
709 This option can be dangerous if used incorrectly! It is a very good idea
710 to run first using the bf(--dry-run) option (bf(-n)) to see what files would be
711 deleted to make sure important files aren't listed.
713 If the sending side detects any I/O errors, then the deletion of any
714 files at the destination will be automatically disabled. This is to
715 prevent temporary filesystem failures (such as NFS errors) on the
716 sending side causing a massive deletion of files on the
717 destination. You can override this with the bf(--ignore-errors) option.
719 The bf(--delete) option may be combined with one of the --delete-WHEN options
720 without conflict, as well as bf(--delete-excluded). However, if none of the
721 --delete-WHEN options are specified, rsync will currently choose the
722 bf(--delete-before) algorithm. A future version may change this to choose the
723 bf(--delete-during) algorithm. See also bf(--delete-after).
725 dit(bf(--delete-before)) Request that the file-deletions on the receiving
726 side be done before the transfer starts. This is the default if bf(--delete)
727 or bf(--delete-excluded) is specified without one of the --delete-WHEN options.
728 See bf(--delete) (which is implied) for more details on file-deletion.
730 Deleting before the transfer is helpful if the filesystem is tight for space
731 and removing extraneous files would help to make the transfer possible.
732 However, it does introduce a delay before the start of the transfer,
733 and this delay might cause the transfer to timeout (if bf(--timeout) was
736 dit(bf(--delete-during, --del)) Request that the file-deletions on the
737 receiving side be done incrementally as the transfer happens. This is
738 a faster method than choosing the before- or after-transfer algorithm,
739 but it is only supported beginning with rsync version 2.6.4.
740 See bf(--delete) (which is implied) for more details on file-deletion.
742 dit(bf(--delete-after)) Request that the file-deletions on the receiving
743 side be done after the transfer has completed. This is useful if you
744 are sending new per-directory merge files as a part of the transfer and
745 you want their exclusions to take effect for the delete phase of the
747 See bf(--delete) (which is implied) for more details on file-deletion.
749 dit(bf(--delete-excluded)) In addition to deleting the files on the
750 receiving side that are not on the sending side, this tells rsync to also
751 delete any files on the receiving side that are excluded (see bf(--exclude)).
752 See the FILTER RULES section for a way to make individual exclusions behave
753 this way on the receiver, and for a way to protect files from
754 bf(--delete-excluded).
755 See bf(--delete) (which is implied) for more details on file-deletion.
757 dit(bf(--ignore-errors)) Tells bf(--delete) to go ahead and delete files
758 even when there are I/O errors.
760 dit(bf(--force)) This options tells rsync to delete directories even if
761 they are not empty when they are to be replaced by non-directories. This
762 is only relevant without bf(--delete) because deletions are now done depth-first.
763 Requires the bf(--recursive) option (which is implied by bf(-a)) to have any effect.
765 dit(bf(--max-delete=NUM)) This tells rsync not to delete more than NUM
766 files or directories (NUM must be non-zero).
767 This is useful when mirroring very large trees to prevent disasters.
769 dit(bf(--max-size=SIZE)) This tells rsync to avoid transferring any
770 file that is larger than the specified SIZE. The SIZE value can be
771 suffixed with a letter to indicate a size multiplier (K, M, or G) and
772 may be a fractional value (e.g. "bf(--max-size=1.5m)").
774 dit(bf(-B, --block-size=BLOCKSIZE)) This forces the block size used in
775 the rsync algorithm to a fixed value. It is normally selected based on
776 the size of each file being updated. See the technical report for details.
778 dit(bf(-e, --rsh=COMMAND)) This option allows you to choose an alternative
779 remote shell program to use for communication between the local and
780 remote copies of rsync. Typically, rsync is configured to use ssh by
781 default, but you may prefer to use rsh on a local network.
783 If this option is used with bf([user@]host::module/path), then the
784 remote shell em(COMMAND) will be used to run an rsync server on the
785 remote host, and all data will be transmitted through that remote
786 shell connection, rather than through a direct socket connection to a
787 running rsync server on the remote host. See the section "CONNECTING
788 TO AN RSYNC SERVER OVER A REMOTE SHELL PROGRAM" above.
790 Command-line arguments are permitted in COMMAND provided that COMMAND is
791 presented to rsync as a single argument. For example:
793 quote(tt( -e "ssh -p 2234"))
795 (Note that ssh users can alternately customize site-specific connect
796 options in their .ssh/config file.)
798 You can also choose the remote shell program using the RSYNC_RSH
799 environment variable, which accepts the same range of values as bf(-e).
801 See also the bf(--blocking-io) option which is affected by this option.
803 dit(bf(--rsync-path=PROGRAM)) Use this to specify what program is to be run
804 on the remote machine to start-up rsync. Often used when rsync is not in
805 the default remote-shell's path (e.g. --rsync-path=/usr/local/bin/rsync).
806 Note that PROGRAM is run with the help of a shell, so it can be any
807 program, script, or command sequence you'd care to run, so long as it does
808 not corrupt the standard-in & standard-out that rsync is using to
811 One tricky example is to set a different default directory on the remote
812 machine for use with the bf(--relative) option. For instance:
814 quote(tt( rsync -avR --rsync-path="cd /a/b && rsync" hst:c/d /e/))
816 dit(bf(-C, --cvs-exclude)) This is a useful shorthand for excluding a
817 broad range of files that you often don't want to transfer between
818 systems. It uses the same algorithm that CVS uses to determine if
819 a file should be ignored.
821 The exclude list is initialized to:
823 quote(quote(tt(RCS SCCS CVS CVS.adm RCSLOG cvslog.* tags TAGS .make.state
824 .nse_depinfo *~ #* .#* ,* _$* *$ *.old *.bak *.BAK *.orig *.rej
825 .del-* *.a *.olb *.o *.obj *.so *.exe *.Z *.elc *.ln core .svn/)))
827 then files listed in a $HOME/.cvsignore are added to the list and any
828 files listed in the CVSIGNORE environment variable (all cvsignore names
829 are delimited by whitespace).
831 Finally, any file is ignored if it is in the same directory as a
832 .cvsignore file and matches one of the patterns listed therein. Unlike
833 rsync's filter/exclude files, these patterns are split on whitespace.
834 See the bf(cvs(1)) manual for more information.
836 If you're combining bf(-C) with your own bf(--filter) rules, you should
837 note that these CVS excludes are appended at the end of your own rules,
838 regardless of where the bf(-C) was placed on the command-line. This makes them
839 a lower priority than any rules you specified explicitly. If you want to
840 control where these CVS excludes get inserted into your filter rules, you
841 should omit the bf(-C) as a command-line option and use a combination of
842 bf(--filter=:C) and bf(--filter=-C) (either on your command-line or by
843 putting the ":C" and "-C" rules into a filter file with your other rules).
844 The first option turns on the per-directory scanning for the .cvsignore
845 file. The second option does a one-time import of the CVS excludes
848 dit(bf(-f, --filter=RULE)) This option allows you to add rules to selectively
849 exclude certain files from the list of files to be transferred. This is
850 most useful in combination with a recursive transfer.
852 You may use as many bf(--filter) options on the command line as you like
853 to build up the list of files to exclude.
855 See the FILTER RULES section for detailed information on this option.
857 dit(bf(-F)) The bf(-F) option is a shorthand for adding two bf(--filter) rules to
858 your command. The first time it is used is a shorthand for this rule:
860 quote(tt( --filter=': /.rsync-filter'))
862 This tells rsync to look for per-directory .rsync-filter files that have
863 been sprinkled through the hierarchy and use their rules to filter the
864 files in the transfer. If bf(-F) is repeated, it is a shorthand for this
867 quote(tt( --filter='- .rsync-filter'))
869 This filters out the .rsync-filter files themselves from the transfer.
871 See the FILTER RULES section for detailed information on how these options
874 dit(bf(--exclude=PATTERN)) This option is a simplified form of the
875 bf(--filter) option that defaults to an exclude rule and does not allow
876 the full rule-parsing syntax of normal filter rules.
878 See the FILTER RULES section for detailed information on this option.
880 dit(bf(--exclude-from=FILE)) This option is similar to the bf(--exclude)
881 option, but instead it adds all exclude patterns listed in the file
882 FILE to the exclude list. Blank lines in FILE and lines starting with
883 ';' or '#' are ignored.
884 If em(FILE) is bf(-) the list will be read from standard input.
886 dit(bf(--include=PATTERN)) This option is a simplified form of the
887 bf(--filter) option that defaults to an include rule and does not allow
888 the full rule-parsing syntax of normal filter rules.
890 See the FILTER RULES section for detailed information on this option.
892 dit(bf(--include-from=FILE)) This specifies a list of include patterns
894 If em(FILE) is "-" the list will be read from standard input.
896 dit(bf(--files-from=FILE)) Using this option allows you to specify the
897 exact list of files to transfer (as read from the specified FILE or "-"
898 for standard input). It also tweaks the default behavior of rsync to make
899 transferring just the specified files and directories easier:
902 it() The bf(--relative) (bf(-R)) option is implied, which preserves the path
903 information that is specified for each item in the file (use
904 bf(--no-relative) if you want to turn that off).
905 it() The bf(--dirs) (bf(-d)) option is implied, which will create directories
906 specified in the list on the destination rather than noisily skipping
908 it() The bf(--archive) (bf(-a)) option's behavior does not imply bf(--recursive)
909 (bf(-r)), so specify it explicitly, if you want it.
912 The file names that are read from the FILE are all relative to the
913 source dir -- any leading slashes are removed and no ".." references are
914 allowed to go higher than the source dir. For example, take this
917 quote(tt( rsync -a --files-from=/tmp/foo /usr remote:/backup))
919 If /tmp/foo contains the string "bin" (or even "/bin"), the /usr/bin
920 directory will be created as /backup/bin on the remote host (but the
921 contents of the /usr/bin dir would not be sent unless you specified bf(-r)
922 or the names were explicitly listed in /tmp/foo). Also keep in mind
923 that the effect of the (enabled by default) bf(--relative) option is to
924 duplicate only the path info that is read from the file -- it does not
925 force the duplication of the source-spec path (/usr in this case).
927 In addition, the bf(--files-from) file can be read from the remote host
928 instead of the local host if you specify a "host:" in front of the file
929 (the host must match one end of the transfer). As a short-cut, you can
930 specify just a prefix of ":" to mean "use the remote end of the
931 transfer". For example:
933 quote(tt( rsync -a --files-from=:/path/file-list src:/ /tmp/copy))
935 This would copy all the files specified in the /path/file-list file that
936 was located on the remote "src" host.
938 dit(bf(-0, --from0)) This tells rsync that the rules/filenames it reads from a
939 file are terminated by a null ('\0') character, not a NL, CR, or CR+LF.
940 This affects bf(--exclude-from), bf(--include-from), bf(--files-from), and any
941 merged files specified in a bf(--filter) rule.
942 It does not affect bf(--cvs-exclude) (since all names read from a .cvsignore
943 file are split on whitespace).
945 dit(bf(-T, --temp-dir=DIR)) This option instructs rsync to use DIR as a
946 scratch directory when creating temporary copies of the files
947 transferred on the receiving side. The default behavior is to create
948 the temporary files in the receiving directory.
950 dit(bf(-y, --fuzzy)) This option tells rsync that it should look for a
951 basis file for any destination file that is missing. The current algorithm
952 looks in the same directory as the destination file for either a file that
953 has an identical size and modified-time, or a similarly-named file. If
954 found, rsync uses the fuzzy basis file to try to speed up the transfer.
956 Note that the use of the bf(--delete) option might get rid of any potential
957 fuzzy-match files, so either use bf(--delete-after) or specify some
958 filename exclusions if you need to prevent this.
960 dit(bf(--compare-dest=DIR)) This option instructs rsync to use em(DIR) on
961 the destination machine as an additional hierarchy to compare destination
962 files against doing transfers (if the files are missing in the destination
963 directory). If a file is found in em(DIR) that is identical to the
964 sender's file, the file will NOT be transferred to the destination
965 directory. This is useful for creating a sparse backup of just files that
966 have changed from an earlier backup.
968 Beginning in version 2.6.4, multiple bf(--compare-dest) directories may be
969 provided, which will cause rsync to search the list in the order specified
971 If a match is found that differs only in attributes, a local copy is made
972 and the attributes updated.
973 If a match is not found, a basis file from one of the em(DIR)s will be
974 selected to try to speed up the transfer.
976 If em(DIR) is a relative path, it is relative to the destination directory.
977 See also bf(--copy-dest) and bf(--link-dest).
979 dit(bf(--copy-dest=DIR)) This option behaves like bf(--compare-dest), but
980 rsync will also copy unchanged files found in em(DIR) to the destination
981 directory using a local copy.
982 This is useful for doing transfers to a new destination while leaving
983 existing files intact, and then doing a flash-cutover when all files have
984 been successfully transferred.
986 Multiple bf(--copy-dest) directories may be provided, which will cause
987 rsync to search the list in the order specified for an unchanged file.
988 If a match is not found, a basis file from one of the em(DIR)s will be
989 selected to try to speed up the transfer.
991 If em(DIR) is a relative path, it is relative to the destination directory.
992 See also bf(--compare-dest) and bf(--link-dest).
994 dit(bf(--link-dest=DIR)) This option behaves like bf(--copy-dest), but
995 unchanged files are hard linked from em(DIR) to the destination directory.
996 The files must be identical in all preserved attributes (e.g. permissions,
997 possibly ownership) in order for the files to be linked together.
1000 quote(tt( rsync -av --link-dest=$PWD/prior_dir host:src_dir/ new_dir/))
1002 Beginning in version 2.6.4, multiple bf(--link-dest) directories may be
1003 provided, which will cause rsync to search the list in the order specified
1005 If a match is found that differs only in attributes, a local copy is made
1006 and the attributes updated.
1007 If a match is not found, a basis file from one of the em(DIR)s will be
1008 selected to try to speed up the transfer.
1010 If em(DIR) is a relative path, it is relative to the destination directory.
1011 See also bf(--compare-dest) and bf(--copy-dest).
1013 Note that rsync versions prior to 2.6.1 had a bug that could prevent
1014 bf(--link-dest) from working properly for a non-root user when bf(-o) was specified
1015 (or implied by bf(-a)). You can work-around this bug by avoiding the bf(-o) option
1016 when sending to an old rsync.
1018 dit(bf(-z, --compress)) With this option, rsync compresses the file data
1019 as it is sent to the destination machine, which reduces the amount of data
1020 being transmitted -- something that is useful over a slow connection.
1022 Note this this option typically achieves better compression ratios that can
1023 be achieved by using a compressing remote shell or a compressing transport
1024 because it takes advantage of the implicit information in the matching data
1025 blocks that are not explicitly sent over the connection.
1027 dit(bf(--numeric-ids)) With this option rsync will transfer numeric group
1028 and user IDs rather than using user and group names and mapping them
1031 By default rsync will use the username and groupname to determine
1032 what ownership to give files. The special uid 0 and the special group
1033 0 are never mapped via user/group names even if the bf(--numeric-ids)
1034 option is not specified.
1036 If a user or group has no name on the source system or it has no match
1037 on the destination system, then the numeric ID
1038 from the source system is used instead. See also the comments on the
1039 "use chroot" setting in the rsyncd.conf manpage for information on how
1040 the chroot setting affects rsync's ability to look up the names of the
1041 users and groups and what you can do about it.
1043 dit(bf(--timeout=TIMEOUT)) This option allows you to set a maximum I/O
1044 timeout in seconds. If no data is transferred for the specified time
1045 then rsync will exit. The default is 0, which means no timeout.
1047 dit(bf(--address)) By default rsync will bind to the wildcard address when
1048 connecting to an rsync daemon. The bf(--address) option allows you to
1049 specify a specific IP address (or hostname) to bind to. See also this
1050 option in the bf(--daemon) mode section.
1052 dit(bf(--port=PORT)) This specifies an alternate TCP port number to use
1053 rather than the default of 873. This is only needed if you are using the
1054 double-colon (::) syntax to connect with an rsync daemon (since the URL
1055 syntax has a way to specify the port as a part of the URL). See also this
1056 option in the bf(--daemon) mode section.
1058 dit(bf(--blocking-io)) This tells rsync to use blocking I/O when launching
1059 a remote shell transport. If the remote shell is either rsh or remsh,
1060 rsync defaults to using
1061 blocking I/O, otherwise it defaults to using non-blocking I/O. (Note that
1062 ssh prefers non-blocking I/O.)
1064 dit(bf(--no-blocking-io)) Turn off bf(--blocking-io), for use when it is the
1067 dit(bf(-i, --itemize-changes)) Requests a simple itemized list of the
1068 changes that are being made to each file, including attribute changes.
1069 This is exactly the same as specifying bf(--log-format='%i %n%L').
1071 The "%i" escape has a cryptic output that is 9 letters long. The general
1072 format is like the string bf(UXcstpoga)), where bf(U) is replaced by the
1073 kind of update being done, bf(X) is replaced by the file-type, and the
1074 other letters represent attributes that may be output if they are being
1077 The update types that replace the bf(U) are as follows:
1080 it() A bf(<) means that a file is being transferred to the remote host
1082 it() A bf(>) means that a file is being transferred to the local host
1084 it() A bf(c) means that a local change/creation is occurring for the item
1085 (such as the creation of a directory or the changing of a symlink, etc.).
1086 it() A bf(h) means that the item is a hard-link to another item (requires
1088 it() A bf(.) means that the item is not being updated (though it might
1089 have attributes that are being modified).
1092 The file-types that replace the bf(X) are: bf(f) for a file, a bf(d) for a
1093 directory, an bf(L) for a symlink, and a bf(D) for a device.
1095 The other letters in the string above are the actual letters that
1096 will be output if the associated attribute for the item is being updated or
1097 a "." for no change. Three exceptions to this are: (1) a newly created
1098 item replaces each letter with a "+", (2) an identical item replaces the
1099 dots with spaces, and (3) an unknown attribute replaces each letter with
1100 a "?" (this happens when talking to an older rsync).
1102 The attribute that is associated with each letter is as follows:
1105 it() A bf(c) means the checksum of the file is different and will be
1106 updated by the file transfer (requires bf(--checksum)).
1107 it() A bf(s) means the size of the file is different and will be updated
1108 by the file transfer.
1109 it() A bf(t) means the modification time is different and is being updated
1110 to the server's value (requires bf(--times)). An alternate value of bf(T)
1111 means that the time will be set to the transfer time, which happens
1112 anytime a symlink is transferred, or when a file or device is transferred
1113 without bf(--times).
1114 it() A bf(p) means the permissions are different and are being updated to
1115 the server's value (requires bf(--perms)).
1116 it() An bf(o) means the owner is different and is being updated to the
1117 server's value (requires bf(--owner) and root privileges).
1118 it() A bf(g) means the group is different and is being updated to the
1119 server's value (requires bf(--group) and the authority to set the group).
1120 it() The bf(a) is reserved for a future enhanced version that supports
1121 extended file attributes, such as ACLs.
1124 One other output is possible: when deleting files, the "%i" will output
1125 the string "*deleting" for each item that is being removed (assuming that
1126 you are talking to a recent enough rsync that it logs deletions instead of
1127 outputting them as a verbose message).
1129 dit(bf(--log-format=FORMAT)) This allows you to specify exactly what the
1130 rsync client outputs to the user on a per-file basis. The format is a text
1131 string containing embedded single-character escape sequences prefixed with
1132 a percent (%) character. For a list of the possible escape characters, see
1133 the "log format" setting in the rsyncd.conf manpage. (Note that this
1134 option does not affect what a daemon logs to its logfile.)
1136 Specifying this option will mention each file, dir, etc. that gets updated
1137 in a significant way (a transferred file, a recreated symlink/device, or a
1138 touched directory) unless the itemized-changes escape (%i) is included in
1139 the string, in which case the logging of names increases to mention any
1140 item that is updated in any way (as long as the receiving side is version
1141 2.6.4). See the bf(--itemized-changes) option for a description of the
1144 The bf(--verbose) option implies a format of "%n%L", but you can use
1145 bf(--log-format) without bv(--verbose) if you like, or you can override
1146 the format of its per-file output using this option.
1148 Rsync will output the log-format string prior to a file's transfer unless
1149 one of the transfer-statistic escapes is requested, in which case the
1150 logging is done at the end of the file's transfer. When this late logging
1151 is in effect and bf(--progress) is also specified, rsync will also output
1152 the name of the file being transferred prior to its progress information
1153 (followed, of course, by the log-format output).
1155 dit(bf(--stats)) This tells rsync to print a verbose set of statistics
1156 on the file transfer, allowing you to tell how effective the rsync
1157 algorithm is for your data.
1159 dit(bf(--partial)) By default, rsync will delete any partially
1160 transferred file if the transfer is interrupted. In some circumstances
1161 it is more desirable to keep partially transferred files. Using the
1162 bf(--partial) option tells rsync to keep the partial file which should
1163 make a subsequent transfer of the rest of the file much faster.
1165 dit(bf(--partial-dir=DIR)) A better way to keep partial files than the
1166 bf(--partial) option is to specify a em(DIR) that will be used to hold the
1167 partial data (instead of writing it out to the destination file).
1168 On the next transfer, rsync will use a file found in this
1169 dir as data to speed up the resumption of the transfer and then deletes it
1170 after it has served its purpose.
1171 Note that if bf(--whole-file) is specified (or implied), any partial-dir
1172 file that is found for a file that is being updated will simply be removed
1174 rsync is sending files without using the incremental rsync algorithm).
1176 Rsync will create the em(DIR) if it is missing (just the last dir -- not
1177 the whole path). This makes it easy to use a relative path (such as
1178 "bf(--partial-dir=.rsync-partial)") to have rsync create the
1179 partial-directory in the destination file's directory when needed, and then
1180 remove it again when the partial file is deleted.
1182 If the partial-dir value is not an absolute path, rsync will also add a directory
1183 bf(--exclude) of this value at the end of all your existing excludes. This
1184 will prevent partial-dir files from being transferred and also prevent the
1185 untimely deletion of partial-dir items on the receiving side. An example:
1186 the above bf(--partial-dir) option would add an "bf(--exclude=.rsync-partial/)"
1187 rule at the end of any other filter rules. Note that if you are
1188 supplying your own filter rules, you may need to manually insert a
1189 rule for this directory exclusion somewhere higher up in the list so that
1190 it has a high enough priority to be effective (e.g., if your rules specify
1191 a trailing bf(--exclude='*') rule, the auto-added rule would never be
1194 IMPORTANT: the bf(--partial-dir) should not be writable by other users or it
1195 is a security risk. E.g. AVOID "/tmp".
1197 You can also set the partial-dir value the RSYNC_PARTIAL_DIR environment
1198 variable. Setting this in the environment does not force bf(--partial) to be
1199 enabled, but rather it effects where partial files go when bf(--partial) is
1200 specified. For instance, instead of using bf(--partial-dir=.rsync-tmp)
1201 along with bf(--progress), you could set RSYNC_PARTIAL_DIR=.rsync-tmp in your
1202 environment and then just use the bf(-P) option to turn on the use of the
1203 .rsync-tmp dir for partial transfers. The only time that the bf(--partial)
1204 option does not look for this environment value is (1) when bf(--inplace) was
1205 specified (since bf(--inplace) conflicts with bf(--partial-dir)), or (2) when
1206 bf(--delay-updates) was specified (see below).
1208 For the purposes of the server-config's "refuse options" setting,
1209 bf(--partial-dir) does em(not) imply bf(--partial). This is so that a
1210 refusal of the bf(--partial) option can be used to disallow the overwriting
1211 of destination files with a partial transfer, while still allowing the
1212 safer idiom provided by bf(--partial-dir).
1214 dit(bf(--delay-updates)) This option puts the temporary file from each
1215 updated file into a holding directory until the end of the
1216 transfer, at which time all the files are renamed into place in rapid
1217 succession. This attempts to make the updating of the files a little more
1218 atomic. By default the files are placed into a directory named ".~tmp~" in
1219 each file's destination directory, but you can override this by specifying
1220 the bf(--partial-dir) option. (Note that RSYNC_PARTIAL_DIR has no effect
1221 on this value, nor is bf(--partial-dir) considered to be implied for the
1222 purposes of the server-config's "refuse options" setting.)
1223 Conflicts with bf(--inplace).
1225 This option uses more memory on the receiving side (one bit per file
1226 transferred) and also requires enough free disk space on the receiving
1227 side to hold an additional copy of all the updated files. Note also that
1228 you should not use an absolute path to bf(--partial-dir) unless there is no
1229 chance of any of the files in the transfer having the same name (since all
1230 the updated files will be put into a single directory if the path is
1233 See also the "atomic-rsync" perl script in the "support" subdir for an
1234 update algorithm that is even more atomic (it uses bf(--link-dest) and a
1235 parallel hierarchy of files).
1237 dit(bf(--progress)) This option tells rsync to print information
1238 showing the progress of the transfer. This gives a bored user
1240 Implies bf(--verbose) if it wasn't already specified.
1242 When the file is transferring, the data looks like this:
1244 verb( 782448 63% 110.64kB/s 0:00:04)
1246 This tells you the current file size, the percentage of the transfer that
1247 is complete, the current calculated file-completion rate (including both
1248 data over the wire and data being matched locally), and the estimated time
1249 remaining in this transfer.
1251 After a file is complete, the data looks like this:
1253 verb( 1238099 100% 146.38kB/s 0:00:08 (5, 57.1% of 396))
1255 This tells you the final file size, that it's 100% complete, the final
1256 transfer rate for the file, the amount of elapsed time it took to transfer
1257 the file, and the addition of a total-transfer summary in parentheses.
1258 These additional numbers tell you how many files have been updated, and
1259 what percent of the total number of files has been scanned.
1261 dit(bf(-P)) The bf(-P) option is equivalent to bf(--partial) bf(--progress). Its
1262 purpose is to make it much easier to specify these two options for a long
1263 transfer that may be interrupted.
1265 dit(bf(--password-file)) This option allows you to provide a password
1266 in a file for accessing a remote rsync server. Note that this option
1267 is only useful when accessing an rsync server using the built in
1268 transport, not when using a remote shell as the transport. The file
1269 must not be world readable. It should contain just the password as a
1272 dit(bf(--list-only)) This option will cause the source files to be listed
1273 instead of transferred. This option is inferred if there is no destination
1274 specified, so you don't usually need to use it explicitly. However, it can
1275 come in handy for a power user that wants to avoid the "bf(-r --exclude='/*/*')"
1276 options that rsync might use as a compatibility kluge when generating a
1277 non-recursive listing.
1279 dit(bf(--bwlimit=KBPS)) This option allows you to specify a maximum
1280 transfer rate in kilobytes per second. This option is most effective when
1281 using rsync with large files (several megabytes and up). Due to the nature
1282 of rsync transfers, blocks of data are sent, then if rsync determines the
1283 transfer was too fast, it will wait before sending the next data block. The
1284 result is an average transfer rate equaling the specified limit. A value
1285 of zero specifies no limit.
1287 dit(bf(--write-batch=FILE)) Record a file that can later be applied to
1288 another identical destination with bf(--read-batch). See the "BATCH MODE"
1289 section for details, and also the bf(--only-write-batch) option.
1291 dit(bf(--only-write-batch=FILE)) Works like bf(--write-batch), except that
1292 no updates are made on the destination system when creating the batch.
1293 This lets you transport the changes to the destination system via some
1294 other means and then apply the changes via bf(--read-batch).
1296 Note that you can feel free to write the batch directly to some portable
1297 media: if this media fills to capacity before the end of the transfer, you
1298 can just apply that partial transfer to the destination and repeat the
1299 whole process to get the rest of the changes (as long as you don't mind a
1300 partially updated destination system while the multi-update cycle is
1303 Also note that you only save bandwidth when pushing changes to a remote
1304 system because this allows the batched data to be diverted from the sender
1305 into the batch file without having to flow over the wire to the receiver
1306 (when pulling, the sender is remote, and thus can't write the batch).
1308 dit(bf(--read-batch=FILE)) Apply all of the changes stored in FILE, a
1309 file previously generated by bf(--write-batch).
1310 If em(FILE) is "-" the batch data will be read from standard input.
1311 See the "BATCH MODE" section for details.
1313 dit(bf(--protocol=NUM)) Force an older protocol version to be used. This
1314 is useful for creating a batch file that is compatible with an older
1315 version of rsync. For instance, if rsync 2.6.4 is being used with the
1316 bf(--write-batch) option, but rsync 2.6.3 is what will be used to run the
1317 bf(--read-batch) option, you should use "--protocol=28" (when creating the
1318 batch file) to force the older protocol version to be used in the batch
1319 file (assuming you can't upgrade the rsync on the reading system to 2.6.4).
1321 dit(bf(-4, --ipv4) or bf(-6, --ipv6)) Tells rsync to prefer IPv4/IPv6
1322 when creating sockets. This only affects sockets that rsync has direct
1323 control over, such as the outgoing socket when directly contacting an
1324 rsync daemon. See also these options in the bf(--daemon) mode section.
1326 dit(bf(--checksum-seed=NUM)) Set the MD4 checksum seed to the integer
1327 NUM. This 4 byte checksum seed is included in each block and file
1328 MD4 checksum calculation. By default the checksum seed is generated
1329 by the server and defaults to the current time(). This option
1330 is used to set a specific checksum seed, which is useful for
1331 applications that want repeatable block and file checksums, or
1332 in the case where the user wants a more random checksum seed.
1333 Note that setting NUM to 0 causes rsync to use the default of time()
1337 manpagesection(DAEMON OPTIONS)
1339 The options allowed when starting an rsync daemon are as follows:
1342 dit(bf(--daemon)) This tells rsync that it is to run as a daemon. The
1343 daemon you start running may be accessed using an rsync client using
1344 the bf(host::module) or bf(rsync://host/module/) syntax.
1346 If standard input is a socket then rsync will assume that it is being
1347 run via inetd, otherwise it will detach from the current terminal and
1348 become a background daemon. The daemon will read the config file
1349 (rsyncd.conf) on each connect made by a client and respond to
1350 requests accordingly. See the rsyncd.conf(5) man page for more
1353 dit(bf(--address)) By default rsync will bind to the wildcard address when
1354 run as a daemon with the bf(--daemon) option. The bf(--address) option
1355 allows you to specify a specific IP address (or hostname) to bind to. This
1356 makes virtual hosting possible in conjunction with the bf(--config) option.
1357 See also the "address" global option in the rsyncd.conf manpage.
1359 dit(bf(--bwlimit=KBPS)) This option allows you to specify a maximum
1360 transfer rate in kilobytes per second for the data the daemon sends.
1361 The client can still specify a smaller bf(--bwlimit) value, but their
1362 requested value will be rounded down if they try to exceed it. See the
1363 client version of this option (above) for some extra details.
1365 dit(bf(--config=FILE)) This specifies an alternate config file than
1366 the default. This is only relevant when bf(--daemon) is specified.
1367 The default is /etc/rsyncd.conf unless the daemon is running over
1368 a remote shell program and the remote user is not root; in that case
1369 the default is rsyncd.conf in the current directory (typically $HOME).
1371 dit(bf(--no-detach)) When running as a daemon, this option instructs
1372 rsync to not detach itself and become a background process. This
1373 option is required when running as a service on Cygwin, and may also
1374 be useful when rsync is supervised by a program such as
1375 bf(daemontools) or AIX's bf(System Resource Controller).
1376 bf(--no-detach) is also recommended when rsync is run under a
1377 debugger. This option has no effect if rsync is run from inetd or
1380 dit(bf(--port=PORT)) This specifies an alternate TCP port number for the
1381 daemon to listen on rather than the default of 873. See also the "port"
1382 global option in the rsyncd.conf manpage.
1384 dit(bf(-v, --verbose)) This option increases the amount of information the
1385 daemon logs during its startup phase. After the client connects, the
1386 daemon's verbosity level will be controlled by the options that the client
1387 used and the "max verbosity" setting in the module's config section.
1389 dit(bf(-4, --ipv4) or bf(-6, --ipv6)) Tells rsync to prefer IPv4/IPv6
1390 when creating the incoming sockets that the rsync daemon will use to
1391 listen for connections. One of these options may be required in older
1392 versions of Linux to work around an IPv6 bug in the kernel (if you see
1393 an "address already in use" error when nothing else is using the port,
1394 try specifying bf(--ipv6) or bf(--ipv4) when starting the daemon).
1396 dit(bf(-h, --help)) When specified after bf(--daemon), print a short help
1397 page describing the options available for starting an rsync daemon.
1400 manpagesection(FILTER RULES)
1402 The filter rules allow for flexible selection of which files to transfer
1403 (include) and which files to skip (exclude). The rules either directly
1404 specify include/exclude patterns or they specify a way to acquire more
1405 include/exclude patterns (e.g. to read them from a file).
1407 As the list of files/directories to transfer is built, rsync checks each
1408 name to be transferred against the list of include/exclude patterns in
1409 turn, and the first matching pattern is acted on: if it is an exclude
1410 pattern, then that file is skipped; if it is an include pattern then that
1411 filename is not skipped; if no matching pattern is found, then the
1412 filename is not skipped.
1414 Rsync builds an ordered list of filter rules as specified on the
1415 command-line. Filter rules have the following syntax:
1418 tt(RULE [PATTERN_OR_FILENAME])nl()
1419 tt(RULE,MODIFIERS [PATTERN_OR_FILENAME])nl()
1422 You have your choice of using either short or long RULE names, as described
1423 below. If you use a short-named rule, the ',' separating the RULE from the
1424 MODIFIERS is optional. The PATTERN or FILENAME that follows (when present)
1425 must come after either a single space or an underscore (_).
1426 Here are the available rule prefixes:
1429 bf(exclude, -) specifies an exclude pattern. nl()
1430 bf(include, +) specifies an include pattern. nl()
1431 bf(merge, .) specifies a merge-file to read for more rules. nl()
1432 bf(dir-merge, :) specifies a per-directory merge-file. nl()
1433 bf(hide, H) specifies a pattern for hiding files from the transfer. nl()
1434 bf(show, S) files that match the pattern are not hidden. nl()
1435 bf(protect, P) specifies a pattern for protecting files from deletion. nl()
1436 bf(risk, R) files that match the pattern are not protected. nl()
1437 bf(clear, !) clears the current include/exclude list (takes no arg) nl()
1440 When rules are being read from a file, empty lines are ignored, as are
1441 comment lines that start with a "#".
1443 Note that the bf(--include)/bf(--exclude) command-line options do not allow the
1444 full range of rule parsing as described above -- they only allow the
1445 specification of include/exclude patterns plus a "!" token to clear the
1446 list (and the normal comment parsing when rules are read from a file).
1448 does not begin with "- " (dash, space) or "+ " (plus, space), then the
1449 rule will be interpreted as if "+ " (for an include option) or "- " (for
1450 an exclude option) were prefixed to the string. A bf(--filter) option, on
1451 the other hand, must always contain either a short or long rule name at the
1454 Note also that the bf(--filter), bf(--include), and bf(--exclude) options take one
1455 rule/pattern each. To add multiple ones, you can repeat the options on
1456 the command-line, use the merge-file syntax of the bf(--filter) option, or
1457 the bf(--include-from)/bf(--exclude-from) options.
1459 manpagesection(INCLUDE/EXCLUDE PATTERN RULES)
1461 You can include and exclude files by specifying patterns using the "+",
1462 "-", etc. filter rules (as introduced in the FILTER RULES section above).
1463 The include/exclude rules each specify a pattern that is matched against
1464 the names of the files that are going to be transferred. These patterns
1465 can take several forms:
1468 it() if the pattern starts with a / then it is anchored to a
1469 particular spot in the hierarchy of files, otherwise it is matched
1470 against the end of the pathname. This is similar to a leading ^ in
1471 regular expressions.
1472 Thus "/foo" would match a file called "foo" at either the "root of the
1473 transfer" (for a global rule) or in the merge-file's directory (for a
1474 per-directory rule).
1475 An unqualified "foo" would match any file or directory named "foo"
1476 anywhere in the tree because the algorithm is applied recursively from
1478 top down; it behaves as if each path component gets a turn at being the
1479 end of the file name. Even the unanchored "sub/foo" would match at
1480 any point in the hierarchy where a "foo" was found within a directory
1481 named "sub". See the section on ANCHORING INCLUDE/EXCLUDE PATTERNS for
1482 a full discussion of how to specify a pattern that matches at the root
1484 it() if the pattern ends with a / then it will only match a
1485 directory, not a file, link, or device.
1486 it() if the pattern contains a wildcard character from the set
1487 *?[ then expression matching is applied using the shell filename
1488 matching rules. Otherwise a simple string match is used.
1489 it() the double asterisk pattern "**" will match slashes while a
1490 single asterisk pattern "*" will stop at slashes.
1491 it() if the pattern contains a / (not counting a trailing /) or a "**"
1492 then it is matched against the full pathname, including any leading
1493 directories. If the pattern doesn't contain a / or a "**", then it is
1494 matched only against the final component of the filename.
1495 (Remember that the algorithm is applied recursively so "full filename"
1496 can actually be any portion of a path from the starting directory on
1500 Note that, when using the bf(--recursive) (bf(-r)) option (which is implied by
1501 bf(-a)), every subcomponent of every path is visited from the top down, so
1502 include/exclude patterns get applied recursively to each subcomponent's
1503 full name (e.g. to include "/foo/bar/baz" the subcomponents "/foo" and
1504 "/foo/bar" must not be excluded).
1505 The exclude patterns actually short-circuit the directory traversal stage
1506 when rsync finds the files to send. If a pattern excludes a particular
1507 parent directory, it can render a deeper include pattern ineffectual
1508 because rsync did not descend through that excluded section of the
1509 hierarchy. This is particularly important when using a trailing '*' rule.
1510 For instance, this won't work:
1513 tt(+ /some/path/this-file-will-not-be-found)nl()
1514 tt(+ /file-is-included)nl()
1518 This fails because the parent directory "some" is excluded by the '*'
1519 rule, so rsync never visits any of the files in the "some" or "some/path"
1520 directories. One solution is to ask for all directories in the hierarchy
1521 to be included by using a single rule: "+ */" (put it somewhere before the
1522 "- *" rule). Another solution is to add specific include rules for all
1523 the parent dirs that need to be visited. For instance, this set of rules
1528 tt(+ /some/path/)nl()
1529 tt(+ /some/path/this-file-is-found)nl()
1530 tt(+ /file-also-included)nl()
1534 Here are some examples of exclude/include matching:
1537 it() "- *.o" would exclude all filenames matching *.o
1538 it() "- /foo" would exclude a file called foo in the transfer-root directory
1539 it() "- foo/" would exclude any directory called foo
1540 it() "- /foo/*/bar" would exclude any file called bar two
1541 levels below a directory called foo in the transfer-root directory
1542 it() "- /foo/**/bar" would exclude any file called bar two
1543 or more levels below a directory called foo in the transfer-root directory
1544 it() The combination of "+ */", "+ *.c", and "- *" would include all
1545 directories and C source files but nothing else.
1546 it() The combination of "+ foo/", "+ foo/bar.c", and "- *" would include
1547 only the foo directory and foo/bar.c (the foo directory must be
1548 explicitly included or it would be excluded by the "*")
1551 manpagesection(MERGE-FILE FILTER RULES)
1553 You can merge whole files into your filter rules by specifying either a
1554 merge (.) or a dir-merge (:) filter rule (as introduced in the FILTER RULES
1557 There are two kinds of merged files -- single-instance ('.') and
1558 per-directory (':'). A single-instance merge file is read one time, and
1559 its rules are incorporated into the filter list in the place of the "."
1560 rule. For per-directory merge files, rsync will scan every directory that
1561 it traverses for the named file, merging its contents when the file exists
1562 into the current list of inherited rules. These per-directory rule files
1563 must be created on the sending side because it is the sending side that is
1564 being scanned for the available files to transfer. These rule files may
1565 also need to be transferred to the receiving side if you want them to
1566 affect what files don't get deleted (see PER-DIRECTORY RULES AND DELETE
1572 tt(merge /etc/rsync/default.rules)nl()
1573 tt(. /etc/rsync/default.rules)nl()
1574 tt(dir-merge .per-dir-filter)nl()
1575 tt(dir-merge,n- .non-inherited-per-dir-excludes)nl()
1576 tt(:n- .non-inherited-per-dir-excludes)nl()
1579 The following modifiers are accepted after a merge or dir-merge rule:
1582 it() A bf(-) specifies that the file should consist of only exclude
1583 patterns, with no other rule-parsing except for in-file comments.
1584 it() A bf(+) specifies that the file should consist of only include
1585 patterns, with no other rule-parsing except for in-file comments.
1586 it() A bf(C) is a way to specify that the file should be read in a
1587 CVS-compatible manner. This turns on 'n', 'w', and '-', but also
1588 allows the list-clearing token (!) to be specified. If no filename is
1589 provided, ".cvsignore" is assumed.
1590 it() A bf(e) will exclude the merge-file name from the transfer; e.g.
1591 "dir-merge,e .rules" is like "dir-merge .rules" and "- .rules".
1592 it() An bf(n) specifies that the rules are not inherited by subdirectories.
1593 it() A bf(w) specifies that the rules are word-split on whitespace instead
1594 of the normal line-splitting. This also turns off comments. Note: the
1595 space that separates the prefix from the rule is treated specially, so
1596 "- foo + bar" is parsed as two rules (assuming that prefix-parsing wasn't
1598 it() You may also specify any of the modifiers for the "+" or "-" rules
1599 (below) in order to have the rules that are read-in from the file
1600 default to having that modifier set. For instance, "merge,-/ .excl" would
1601 treat the contents of .excl as absolute-path excludes,
1602 while "dir-merge,s .filt" and ":sC" would each make all their
1603 per-directory rules apply only on the server side.
1606 The following modifiers are accepted after a "+" or "-":
1609 it() A "/" specifies that the include/exclude should be treated as an
1610 absolute path, relative to the root of the filesystem. For example,
1611 "-/ /etc/passwd" would exclude the passwd file any time the transfer
1612 was sending files from the "/etc" directory.
1613 it() A "!" specifies that the include/exclude should take effect if
1614 the pattern fails to match. For instance, "-! */" would exclude all
1616 it() A bf(C) is used to indicate that all the global CVS-exclude rules
1617 should be inserted as excludes in place of the "-C". No arg should
1619 it() An bf(s) is used to indicate that the rule applies to the sending
1620 side. When a rule affects the sending side, it prevents files from
1621 being transferred. The default is for a rule to affect both sides
1622 unless bf(--delete-excluded) was specified, in which case default rules
1623 become sender-side only. See also the hide (H) and show (S) rules,
1624 which are an alternate way to specify server-side includes/excludes.
1625 it() An bf(r) is used to indicate that the rule applies to the receiving
1626 side. When a rule affects the receiving side, it prevents files from
1627 being deleted. See the bf(s) modifier for more info. See also the
1628 protect (P) and risk (R) rules, which are an alternate way to
1629 specify receiver-side includes/excludes.
1632 Per-directory rules are inherited in all subdirectories of the directory
1633 where the merge-file was found unless the 'n' modifier was used. Each
1634 subdirectory's rules are prefixed to the inherited per-directory rules
1635 from its parents, which gives the newest rules a higher priority than the
1636 inherited rules. The entire set of dir-merge rules are grouped together in
1637 the spot where the merge-file was specified, so it is possible to override
1638 dir-merge rules via a rule that got specified earlier in the list of global
1639 rules. When the list-clearing rule ("!") is read from a per-directory
1640 file, it only clears the inherited rules for the current merge file.
1642 Another way to prevent a single rule from a dir-merge file from being inherited is to
1643 anchor it with a leading slash. Anchored rules in a per-directory
1644 merge-file are relative to the merge-file's directory, so a pattern "/foo"
1645 would only match the file "foo" in the directory where the dir-merge filter
1648 Here's an example filter file which you'd specify via bf(--filter=". file":)
1651 tt(merge /home/user/.global-filter)nl()
1653 tt(dir-merge .rules)nl()
1658 This will merge the contents of the /home/user/.global-filter file at the
1659 start of the list and also turns the ".rules" filename into a per-directory
1660 filter file. All rules read-in prior to the start of the directory scan
1661 follow the global anchoring rules (i.e. a leading slash matches at the root
1664 If a per-directory merge-file is specified with a path that is a parent
1665 directory of the first transfer directory, rsync will scan all the parent
1666 dirs from that starting point to the transfer directory for the indicated
1667 per-directory file. For instance, here is a common filter (see bf(-F)):
1669 quote(tt(--filter=': /.rsync-filter'))
1671 That rule tells rsync to scan for the file .rsync-filter in all
1672 directories from the root down through the parent directory of the
1673 transfer prior to the start of the normal directory scan of the file in
1674 the directories that are sent as a part of the transfer. (Note: for an
1675 rsync daemon, the root is always the same as the module's "path".)
1677 Some examples of this pre-scanning for per-directory files:
1680 tt(rsync -avF /src/path/ /dest/dir)nl()
1681 tt(rsync -av --filter=': ../../.rsync-filter' /src/path/ /dest/dir)nl()
1682 tt(rsync -av --filter=': .rsync-filter' /src/path/ /dest/dir)nl()
1685 The first two commands above will look for ".rsync-filter" in "/" and
1686 "/src" before the normal scan begins looking for the file in "/src/path"
1687 and its subdirectories. The last command avoids the parent-dir scan
1688 and only looks for the ".rsync-filter" files in each directory that is
1689 a part of the transfer.
1691 If you want to include the contents of a ".cvsignore" in your patterns,
1692 you should use the rule ":C", which creates a dir-merge of the .cvsignore
1693 file, but parsed in a CVS-compatible manner. You can
1694 use this to affect where the bf(--cvs-exclude) (bf(-C)) option's inclusion of the
1695 per-directory .cvsignore file gets placed into your rules by putting the
1696 ":C" wherever you like in your filter rules. Without this, rsync would
1697 add the dir-merge rule for the .cvsignore file at the end of all your other
1698 rules (giving it a lower priority than your command-line rules). For
1702 tt(cat <<EOT | rsync -avC --filter='. -' a/ b)nl()
1707 tt(rsync -avC --include=foo.o -f :C --exclude='*.old' a/ b)nl()
1710 Both of the above rsync commands are identical. Each one will merge all
1711 the per-directory .cvsignore rules in the middle of the list rather than
1712 at the end. This allows their dir-specific rules to supersede the rules
1713 that follow the :C instead of being subservient to all your rules. To
1714 affect the other CVS exclude rules (i.e. the default list of exclusions,
1715 the contents of $HOME/.cvsignore, and the value of $CVSIGNORE) you should
1716 omit the bf(-C) command-line option and instead insert a "-C" rule into
1717 your filter rules; e.g. "--filter=-C".
1719 manpagesection(LIST-CLEARING FILTER RULE)
1721 You can clear the current include/exclude list by using the "!" filter
1722 rule (as introduced in the FILTER RULES section above). The "current"
1723 list is either the global list of rules (if the rule is encountered while
1724 parsing the filter options) or a set of per-directory rules (which are
1725 inherited in their own sub-list, so a subdirectory can use this to clear
1726 out the parent's rules).
1728 manpagesection(ANCHORING INCLUDE/EXCLUDE PATTERNS)
1730 As mentioned earlier, global include/exclude patterns are anchored at the
1731 "root of the transfer" (as opposed to per-directory patterns, which are
1732 anchored at the merge-file's directory). If you think of the transfer as
1733 a subtree of names that are being sent from sender to receiver, the
1734 transfer-root is where the tree starts to be duplicated in the destination
1735 directory. This root governs where patterns that start with a / match.
1737 Because the matching is relative to the transfer-root, changing the
1738 trailing slash on a source path or changing your use of the bf(--relative)
1739 option affects the path you need to use in your matching (in addition to
1740 changing how much of the file tree is duplicated on the destination
1741 host). The following examples demonstrate this.
1743 Let's say that we want to match two source files, one with an absolute
1744 path of "/home/me/foo/bar", and one with a path of "/home/you/bar/baz".
1745 Here is how the various command choices differ for a 2-source transfer:
1748 Example cmd: rsync -a /home/me /home/you /dest nl()
1749 +/- pattern: /me/foo/bar nl()
1750 +/- pattern: /you/bar/baz nl()
1751 Target file: /dest/me/foo/bar nl()
1752 Target file: /dest/you/bar/baz nl()
1756 Example cmd: rsync -a /home/me/ /home/you/ /dest nl()
1757 +/- pattern: /foo/bar (note missing "me") nl()
1758 +/- pattern: /bar/baz (note missing "you") nl()
1759 Target file: /dest/foo/bar nl()
1760 Target file: /dest/bar/baz nl()
1764 Example cmd: rsync -a --relative /home/me/ /home/you /dest nl()
1765 +/- pattern: /home/me/foo/bar (note full path) nl()
1766 +/- pattern: /home/you/bar/baz (ditto) nl()
1767 Target file: /dest/home/me/foo/bar nl()
1768 Target file: /dest/home/you/bar/baz nl()
1772 Example cmd: cd /home; rsync -a --relative me/foo you/ /dest nl()
1773 +/- pattern: /me/foo/bar (starts at specified path) nl()
1774 +/- pattern: /you/bar/baz (ditto) nl()
1775 Target file: /dest/me/foo/bar nl()
1776 Target file: /dest/you/bar/baz nl()
1779 The easiest way to see what name you should filter is to just
1780 look at the output when using bf(--verbose) and put a / in front of the name
1781 (use the bf(--dry-run) option if you're not yet ready to copy any files).
1783 manpagesection(PER-DIRECTORY RULES AND DELETE)
1785 Without a delete option, per-directory rules are only relevant on the
1786 sending side, so you can feel free to exclude the merge files themselves
1787 without affecting the transfer. To make this easy, the 'e' modifier adds
1788 this exclude for you, as seen in these two equivalent commands:
1791 tt(rsync -av --filter=': .excl' --exclude=.excl host:src/dir /dest)nl()
1792 tt(rsync -av --filter=':e .excl' host:src/dir /dest)nl()
1795 However, if you want to do a delete on the receiving side AND you want some
1796 files to be excluded from being deleted, you'll need to be sure that the
1797 receiving side knows what files to exclude. The easiest way is to include
1798 the per-directory merge files in the transfer and use bf(--delete-after),
1799 because this ensures that the receiving side gets all the same exclude
1800 rules as the sending side before it tries to delete anything:
1802 quote(tt(rsync -avF --delete-after host:src/dir /dest))
1804 However, if the merge files are not a part of the transfer, you'll need to
1805 either specify some global exclude rules (i.e. specified on the command
1806 line), or you'll need to maintain your own per-directory merge files on
1807 the receiving side. An example of the first is this (assume that the
1808 remote .rules files exclude themselves):
1810 verb(rsync -av --filter=': .rules' --filter='. /my/extra.rules'
1811 --delete host:src/dir /dest)
1813 In the above example the extra.rules file can affect both sides of the
1814 transfer, but (on the sending side) the rules are subservient to the rules
1815 merged from the .rules files because they were specified after the
1816 per-directory merge rule.
1818 In one final example, the remote side is excluding the .rsync-filter
1819 files from the transfer, but we want to use our own .rsync-filter files
1820 to control what gets deleted on the receiving side. To do this we must
1821 specifically exclude the per-directory merge files (so that they don't get
1822 deleted) and then put rules into the local files to control what else
1823 should not get deleted. Like one of these commands:
1825 verb( rsync -av --filter=':e /.rsync-filter' --delete \
1827 rsync -avFF --delete host:src/dir /dest)
1829 manpagesection(BATCH MODE)
1831 Batch mode can be used to apply the same set of updates to many
1832 identical systems. Suppose one has a tree which is replicated on a
1833 number of hosts. Now suppose some changes have been made to this
1834 source tree and those changes need to be propagated to the other
1835 hosts. In order to do this using batch mode, rsync is run with the
1836 write-batch option to apply the changes made to the source tree to one
1837 of the destination trees. The write-batch option causes the rsync
1838 client to store in a "batch file" all the information needed to repeat
1839 this operation against other, identical destination trees.
1841 To apply the recorded changes to another destination tree, run rsync
1842 with the read-batch option, specifying the name of the same batch
1843 file, and the destination tree. Rsync updates the destination tree
1844 using the information stored in the batch file.
1846 For convenience, one additional file is creating when the write-batch
1847 option is used. This file's name is created by appending
1848 ".sh" to the batch filename. The .sh file contains
1849 a command-line suitable for updating a destination tree using that
1850 batch file. It can be executed using a Bourne(-like) shell, optionally
1851 passing in an alternate destination tree pathname which is then used
1852 instead of the original path. This is useful when the destination tree
1853 path differs from the original destination tree path.
1855 Generating the batch file once saves having to perform the file
1856 status, checksum, and data block generation more than once when
1857 updating multiple destination trees. Multicast transport protocols can
1858 be used to transfer the batch update files in parallel to many hosts
1859 at once, instead of sending the same data to every host individually.
1864 tt($ rsync --write-batch=foo -a host:/source/dir/ /adest/dir/)nl()
1865 tt($ scp foo* remote:)nl()
1866 tt($ ssh remote ./foo.sh /bdest/dir/)nl()
1870 tt($ rsync --write-batch=foo -a /source/dir/ /adest/dir/)nl()
1871 tt($ ssh remote rsync --read-batch=- -a /bdest/dir/ <foo)nl()
1874 In these examples, rsync is used to update /adest/dir/ from /source/dir/
1875 and the information to repeat this operation is stored in "foo" and
1876 "foo.sh". The host "remote" is then updated with the batched data going
1877 into the directory /bdest/dir. The differences between the two examples
1878 reveals some of the flexibility you have in how you deal with batches:
1881 it() The first example shows that the initial copy doesn't have to be
1882 local -- you can push or pull data to/from a remote host using either the
1883 remote-shell syntax or rsync daemon syntax, as desired.
1884 it() The first example uses the created "foo.sh" file to get the right
1885 rsync options when running the read-batch command on the remote host.
1886 it() The second example reads the batch data via standard input so that
1887 the batch file doesn't need to be copied to the remote machine first.
1888 This example avoids the foo.sh script because it needed to use a modified
1889 bf(--read-batch) option, but you could edit the script file if you wished to
1890 make use of it (just be sure that no other option is trying to use
1891 standard input, such as the "bf(--exclude-from=-)" option).
1896 The read-batch option expects the destination tree that it is updating
1897 to be identical to the destination tree that was used to create the
1898 batch update fileset. When a difference between the destination trees
1899 is encountered the update might be discarded with a warning (if the file
1900 appears to be up-to-date already) or the file-update may be attempted
1901 and then, if the file fails to verify, the update discarded with an
1902 error. This means that it should be safe to re-run a read-batch operation
1903 if the command got interrupted. If you wish to force the batched-update to
1904 always be attempted regardless of the file's size and date, use the bf(-I)
1905 option (when reading the batch).
1906 If an error occurs, the destination tree will probably be in a
1907 partially updated state. In that case, rsync can
1908 be used in its regular (non-batch) mode of operation to fix up the
1911 The rsync version used on all destinations must be at least as new as the
1912 one used to generate the batch file. Rsync will die with an error if the
1913 protocol version in the batch file is too new for the batch-reading rsync
1914 to handle. See also the bf(--protocol) option for a way to have the
1915 creating rsync generate a batch file that an older rsync can understand.
1916 (Note that batch files changed format in version 2.6.3, so mixing versions
1917 older than that with newer versions will not work.)
1919 When reading a batch file, rsync will force the value of certain options
1920 to match the data in the batch file if you didn't set them to the same
1921 as the batch-writing command. Other options can (and should) be changed.
1922 For instance bf(--write-batch) changes to bf(--read-batch),
1923 bf(--files-from) is dropped, and the
1924 bf(--filter)/bf(--include)/bf(--exclude) options are not needed unless
1925 one of the bf(--delete) options is specified.
1927 The code that creates the BATCH.sh file transforms any filter/include/exclude
1928 options into a single list that is appended as a "here" document to the
1929 shell script file. An advanced user can use this to modify the exclude
1930 list if a change in what gets deleted by bf(--delete) is desired. A normal
1931 user can ignore this detail and just use the shell script as an easy way
1932 to run the appropriate bf(--read-batch) command for the batched data.
1934 The original batch mode in rsync was based on "rsync+", but the latest
1935 version uses a new implementation.
1937 manpagesection(SYMBOLIC LINKS)
1939 Three basic behaviors are possible when rsync encounters a symbolic
1940 link in the source directory.
1942 By default, symbolic links are not transferred at all. A message
1943 "skipping non-regular" file is emitted for any symlinks that exist.
1945 If bf(--links) is specified, then symlinks are recreated with the same
1946 target on the destination. Note that bf(--archive) implies
1949 If bf(--copy-links) is specified, then symlinks are "collapsed" by
1950 copying their referent, rather than the symlink.
1952 rsync also distinguishes "safe" and "unsafe" symbolic links. An
1953 example where this might be used is a web site mirror that wishes
1954 ensure the rsync module they copy does not include symbolic links to
1955 bf(/etc/passwd) in the public section of the site. Using
1956 bf(--copy-unsafe-links) will cause any links to be copied as the file
1957 they point to on the destination. Using bf(--safe-links) will cause
1958 unsafe links to be omitted altogether.
1960 Symbolic links are considered unsafe if they are absolute symlinks
1961 (start with bf(/)), empty, or if they contain enough bf("..")
1962 components to ascend from the directory being copied.
1964 manpagediagnostics()
1966 rsync occasionally produces error messages that may seem a little
1967 cryptic. The one that seems to cause the most confusion is "protocol
1968 version mismatch -- is your shell clean?".
1970 This message is usually caused by your startup scripts or remote shell
1971 facility producing unwanted garbage on the stream that rsync is using
1972 for its transport. The way to diagnose this problem is to run your
1973 remote shell like this:
1975 quote(tt(ssh remotehost /bin/true > out.dat))
1977 then look at out.dat. If everything is working correctly then out.dat
1978 should be a zero length file. If you are getting the above error from
1979 rsync then you will probably find that out.dat contains some text or
1980 data. Look at the contents and try to work out what is producing
1981 it. The most common cause is incorrectly configured shell startup
1982 scripts (such as .cshrc or .profile) that contain output statements
1983 for non-interactive logins.
1985 If you are having trouble debugging filter patterns, then
1986 try specifying the bf(-vv) option. At this level of verbosity rsync will
1987 show why each individual file is included or excluded.
1989 manpagesection(EXIT VALUES)
1993 dit(bf(1)) Syntax or usage error
1994 dit(bf(2)) Protocol incompatibility
1995 dit(bf(3)) Errors selecting input/output files, dirs
1996 dit(bf(4)) Requested action not supported: an attempt
1997 was made to manipulate 64-bit files on a platform that cannot support
1998 them; or an option was specified that is supported by the client and
2000 dit(bf(5)) Error starting client-server protocol
2001 dit(bf(6)) Daemon unable to append to log-file
2002 dit(bf(10)) Error in socket I/O
2003 dit(bf(11)) Error in file I/O
2004 dit(bf(12)) Error in rsync protocol data stream
2005 dit(bf(13)) Errors with program diagnostics
2006 dit(bf(14)) Error in IPC code
2007 dit(bf(20)) Received SIGUSR1 or SIGINT
2008 dit(bf(21)) Some error returned by waitpid()
2009 dit(bf(22)) Error allocating core memory buffers
2010 dit(bf(23)) Partial transfer due to error
2011 dit(bf(24)) Partial transfer due to vanished source files
2012 dit(bf(25)) The --max-delete limit stopped deletions
2013 dit(bf(30)) Timeout in data send/receive
2016 manpagesection(ENVIRONMENT VARIABLES)
2019 dit(bf(CVSIGNORE)) The CVSIGNORE environment variable supplements any
2020 ignore patterns in .cvsignore files. See the bf(--cvs-exclude) option for
2022 dit(bf(RSYNC_RSH)) The RSYNC_RSH environment variable allows you to
2023 override the default shell used as the transport for rsync. Command line
2024 options are permitted after the command name, just as in the bf(-e) option.
2025 dit(bf(RSYNC_PROXY)) The RSYNC_PROXY environment variable allows you to
2026 redirect your rsync client to use a web proxy when connecting to a
2027 rsync daemon. You should set RSYNC_PROXY to a hostname:port pair.
2028 dit(bf(RSYNC_PASSWORD)) Setting RSYNC_PASSWORD to the required
2029 password allows you to run authenticated rsync connections to an rsync
2030 daemon without user intervention. Note that this does not supply a
2031 password to a shell transport such as ssh.
2032 dit(bf(USER) or bf(LOGNAME)) The USER or LOGNAME environment variables
2033 are used to determine the default username sent to an rsync server.
2034 If neither is set, the username defaults to "nobody".
2035 dit(bf(HOME)) The HOME environment variable is used to find the user's
2036 default .cvsignore file.
2041 /etc/rsyncd.conf or rsyncd.conf
2049 times are transferred as unix time_t values
2051 When transferring to FAT filesystems rsync may re-sync
2053 See the comments on the bf(--modify-window) option.
2055 file permissions, devices, etc. are transferred as native numerical
2058 see also the comments on the bf(--delete) option
2060 Please report bugs! See the website at
2061 url(http://rsync.samba.org/)(http://rsync.samba.org/)
2063 manpagesection(CREDITS)
2065 rsync is distributed under the GNU public license. See the file
2066 COPYING for details.
2068 A WEB site is available at
2069 url(http://rsync.samba.org/)(http://rsync.samba.org/). The site
2070 includes an FAQ-O-Matic which may cover questions unanswered by this
2073 The primary ftp site for rsync is
2074 url(ftp://rsync.samba.org/pub/rsync)(ftp://rsync.samba.org/pub/rsync).
2076 We would be delighted to hear from you if you like this program.
2078 This program uses the excellent zlib compression library written by
2079 Jean-loup Gailly and Mark Adler.
2081 manpagesection(THANKS)
2083 Thanks to Richard Brent, Brendan Mackay, Bill Waite, Stephen Rothwell
2084 and David Bell for helpful suggestions, patches and testing of rsync.
2085 I've probably missed some people, my apologies if I have.
2087 Especial thanks also to: David Dykstra, Jos Backus, Sebastian Krahmer,
2088 Martin Pool, Wayne Davison, J.W. Schultz.
2092 rsync was originally written by Andrew Tridgell and Paul Mackerras.
2093 Many people have later contributed to it.
2095 Mailing lists for support and development are available at
2096 url(http://lists.samba.org)(lists.samba.org)