1 mailto(rsync-bugs@samba.anu.edu.au)
2 manpage(rsync)(1)(13 May 1998)()()
3 manpagename(rsync)(faster, flexible replacement for rcp)
6 rsync [options] [user@]host:path path
8 rsync [options] path [user@]host:path
10 rsync [options] path path
12 rsync [options] [user@]host::module[/path] path
14 rsync [options] path [user@]host::module[/path]
18 rsync is a program that behaves in much the same way that rcp does,
19 but has many more options and uses the rsync remote-update protocol to
20 greatly speedup file transfers when the destination file already
23 The rsync remote-update protocol allows rsync to transfer just the
24 differences between two sets of files across the network link, using
25 an efficient checksum-search algorithm described in the technical
26 report that accompanies this package.
28 Some of the additional features of rsync are:
31 it() support for copying links, devices, owners, groups and permissions
32 it() exclude and exclude-from options similar to GNU tar
33 it() a CVS exclude mode for ignoring the same files that CVS would ignore
34 it() can use any transparent remote shell, including rsh or ssh
35 it() does not require root privileges
36 it() pipelining of file transfers to minimize latency costs
37 it() support for anonymous or authenticated rsync servers (ideal for
41 manpagesection(GENERAL)
43 There are five different ways of using rsync. They are:
46 it() for copying local files. This is invoked when neither
47 source nor destination path contains a : separator
49 it() for copying from the local machine to a remote machine using
50 a remote shell program as the transport (such as rsh or
51 ssh). This is invoked when the destination path contains a
54 it() for copying from a remote machine to the local machine
55 using a remote shell program. This is invoked when the local path
56 contains a : separator.
58 it() for copying from a remote rsync server to the local
59 machine. This is invoked when the source path contains a ::
62 it() for copying from the local machine to a remote rsync
63 server. This is invoked when the destination path contains a ::
67 Note that in all cases at least one of the source and destination
72 See the file README for installation instructions.
74 Once installed you can use rsync to any machine that you can use rsh
75 to. rsync uses rsh for its communications, unless both the source and
76 destination are local.
78 You can also specify a alternative to rsh, by either using the -e
79 command line option, or by setting the RSYNC_RSH environment variable.
81 One common substitute is to use ssh, which offers a high degree of
84 Note that rsync must be installed on both the source and destination
89 You use rsync in the same way you use rcp. You must specify a source
90 and a destination, one of which may be remote.
92 Perhaps the best way to explain the syntax is some examples:
94 quote(rsync *.c foo:src/)
96 this would transfer all files matching the pattern *.c from the
97 current directory to the directory src on the machine foo. If any of
98 the files already exist on the remote system then the rsync
99 remote-update protocol is used to update the file by sending only the
100 differences. See the tech report for details.
102 quote(rsync -avz foo:src/bar /data/tmp)
104 recursively transfer all files from the directory src/bar on the
105 machine foo into the /data/tmp/bar directory on the local machine. The
106 files are transferred in "archive" mode, which ensures that symbolic
107 links, devices, attributes, permissions, ownerships etc are preserved
108 in the transfer. Additionally compression will be used to reduce the
109 size of data portions of the transfer.
111 quote(rsync -avz foo:src/bar/ /data/tmp)
113 With a trailing slash on the source this behavior changes to transfer
114 all files from the directory src/bar on the machine foo into the
115 /data/tmp/. With a trailing / on a source name it means "copy the
116 contents of this directory". Without a trailing slash it means "copy
117 the directory". This difference becomes particularly important when
118 using the --delete option.
120 You can also use rsync in local-only mode, where both the source and
121 destination don't have a ':' in the name. In this case it behaves like
122 an improved copy command.
125 manpagesection(CONNECTING TO AN RSYNC SERVER)
127 It is also possible to use rsync without using rsh or ssh as the
128 transport. In this case you will connect to a remote rsync server
129 running on TCP port 873.
131 Using rsync in this was is the same as using it with rsh or ssh except
135 it() you use a double colon :: instead of a single colon to
136 separate the hostname from the path.
138 it() the remote server may print a message of the day when you
141 it() if you specify no path name on the remote server then the
142 list of accessible paths on the server will be shown.
145 Some paths on the remote server may require authentication. If so then
146 you will receive a password prompt when you connect. You can avoid the
147 password prompt by setting the environment variable RSYNC_PASSWORD to
148 the password you want to use. This may be useful when scripting rsync.
150 manpagesection(RUNNING AN RSYNC SERVER)
152 An rsync server is configured using a config file which by default is
153 called /etc/rsyncd.conf. Please see the rsyncd.conf(5) man page for more
156 manpagesection(EXAMPLES)
158 Here are some examples of how I use rsync.
160 To backup my wife's home directory, which consists of large MS word
161 files and mail folders I use a cron job that runs
163 quote(rsync -Cavz . arvidsjaur:backup)
165 each night over a PPP link to a duplicate directory on my machine
168 To synchronize my samba source trees I use the following Makefile
172 rsync -avuzb --exclude '*~' samba:samba/ .
175 rsync -Cavuzb . samba:samba/
179 this allows me to sync with a CVS directory at the other end of the
180 link. I then do cvs operations on the remote machine, which saves a
181 lot of time as the remote cvs protocol isn't very efficient.
183 I mirror a directory between my "old" and "new" ftp sites with the
186 quote(rsync -az -e ssh --delete ~ftp/pub/samba/ nimbus:"~ftp/pub/tridge/samba")
188 this is launched from cron every few hours.
190 manpagesection(OPTIONS SUMMARY)
192 Here is a short summary of the options avalable in rsync. Please refer
193 to the detailed description below for a complete description.
196 Usage: rsync [OPTION]... SRC [USER@]HOST:DEST
197 or rsync [OPTION]... [USER@]HOST:SRC DEST
198 or rsync [OPTION]... SRC DEST
199 or rsync [OPTION]... [USER@]HOST::SRC DEST
200 or rsync [OPTION]... SRC [USER@]HOST::DEST
203 -v, --verbose increase verbosity
204 -c, --checksum always checksum
205 -a, --archive archive mode
206 -r, --recursive recurse into directories
207 -R, --relative use relative path names
208 -b, --backup make backups (default ~ extension)
209 -u, --update update only (don't overwrite newer files)
210 -l, --links preserve soft links
211 -L, --copy-links treat soft links like regular files
212 --safe-links ignore links outside the destination tree
213 -H, --hard-links preserve hard links
214 -p, --perms preserve permissions
215 -o, --owner preserve owner (root only)
216 -g, --group preserve group
217 -D, --devices preserve devices (root only)
218 -t, --times preserve times
219 -S, --sparse handle sparse files efficiently
220 -n, --dry-run show what would have been transferred
221 -W, --whole-file copy whole files, no incremental checks
222 -x, --one-file-system don't cross filesystem boundaries
223 -B, --block-size=SIZE checksum blocking size
224 -e, --rsh=COMMAND specify rsh replacement
225 --rsync-path=PATH specify path to rsync on the remote machine
226 -C, --cvs-exclude auto ignore files in the same way CVS does
227 --delete delete files that don't exist on the sending side
228 --partial keep partially transferred files
229 --force force deletion of directories even if not empty
230 --numeric-ids don't map uid/gid values by user/group name
231 --timeout=TIME set IO timeout in seconds
232 -I, --ignore-times don't exclude files that match length and time
233 -T --temp-dir=DIR create temporary files in directory DIR
234 --compare-dest=DIR also compare destination files relative to DIR
235 -z, --compress compress file data
236 --exclude=PATTERN exclude file FILE
237 --exclude-from=PATTERN exclude files listed in FILE
238 --include=PATTERN don't exclude file FILE
239 --include-from=PATTERN don't exclude files listed in FILE
240 --suffix=SUFFIX override backup suffix
241 --version print version number
242 --daemon run as a rsync daemon
243 --config=FILE specify alternate rsyncd.conf file
244 --port=PORT specify alternate rsyncd port number
245 --stats give some file transfer stats
246 --progress show progress during transfer
247 -h, --help show this help screen
252 rsync uses the GNU long options package. Many of the command line
253 options have two variants, one short and one long. These are shown
254 below separated by commas. Some options only have a long variant.
257 dit(bf(-h, --help)) Print a short help page describing the options
260 dit(bf(--version)) print the rsync version number and exit
262 dit(bf(-v, --verbose)) This option increases the amount of information you
263 are given during the transfer. By default rsync works silently. A
264 single -v will give you information about what files are being
265 transferred and a brief summary at the end. Two -v flags will give you
266 information on what files are being skipped and slightly more
267 information at the end. More than two -v flags should only be used if
268 you are debugging rsync
270 dit(bf(-I, --ignore-times)) Normally rsync will skip any files that are
271 already the same length and have the same time-stamp. This option turns
274 dit(bf(-c, --checksum)) This forces the sender to checksum all files using
275 a 128-bit MD4 checksum before transfer. The checksum is then
276 explicitly checked on the receiver and any files of the same name
277 which already exist and have the same checksum and size on the
278 receiver are skipped. This option can be quite slow.
280 dit(bf(-a, --archive)) This is equivalent to -rlptDg. It is a quick way
281 of saying I want recursion and want to preserve everything.
283 Note: if the user launching rsync is root then the -o option (preserve
284 uid) is also implied.
286 dit(bf(-r, --recursive)) This tells rsync to copy directories recursively
288 dit(bf(-R, --relative)) Use relative paths. This means that the full path
289 names specified on the command line are sent to the server rather than
290 just the last parts of the filenames. This is particularly useful when
291 you want to sent several different directories at the same time. For
292 example if you used the command
294 verb(rsync foo/bar/foo.c remote:/tmp/)
296 then this would create a file called foo.c in /tmp/ on the remote
297 machine. If instead you used
299 verb(rsync -R foo/bar/foo.c remote:/tmp/)
301 then a file called /tmp/foo/bar/foo.c would be created on the remote
302 machine. The full path name is preserved.
304 dit(bf(-b, --backup)) With this option preexisting destination files are
305 renamed with a ~ extension as each file is transferred. You can
306 control the backup suffix using the --suffix option.
308 dit(bf(-u, --update)) This forces rsync to skip any files for which the
309 destination file already exists and has a date later than the source
312 dit(bf(-l, --links)) This tells rsync to recreate symbolic links on the
313 remote system to be the same as the local system. Without this
314 option all symbolic links are skipped.
316 dit(bf(-L, --copy-links)) This tells rsync to treat symbolic links just
319 dit(bf(--safe-links)) This tells rsync to ignore any symbolic links
320 which point outside the destination tree. All absolute symlinks are
321 also ignored. Using this option in conjunction with --relative may
322 give unexpecetd results.
324 dit(bf(-H, --hard-links)) This tells rsync to recreate hard links on
325 the remote system to be the same as the local system. Without this
326 option hard links are treated like regular files.
328 Note that rsync can only detect hard links if both parts of the link
329 are in the list of files being sent.
331 This option can be quite slow, so only use it if you need it.
333 dit(bf(-W, --whole-file)) With this option the incremental rsync algorithm
334 is not used and the whole file is sent as-is instead. This may be
335 useful when using rsync with a local machine.
337 dit(bf(--partial)) By default rsync will delete any partially
338 transferred file if the transfer is interrupted. In some circumstances
339 it is more desirable to keep partially transferred files. Using the
340 --partial option tells rsync to keep the partial file which should
341 make a subsequent transfer of the rest of the file much faster.
343 dit(bf(-p, --perms)) This option causes rsync to update the remote
344 permissions to be the same as the local permissions.
346 dit(bf(-o, --owner)) This option causes rsync to update the remote owner
347 of the file to be the same as the local owner. This is only available
350 dit(bf(-g, --group)) This option causes rsync to update the remote group
351 of the file to be the same as the local group.
353 dit(bf(-D, --devices)) This option causes rsync to transfer character and
354 block device information to the remote system to recreate these
355 devices. This option is only available to the super-user.
357 dit(bf(-t, --times)) This tells rsync to transfer modification times along
358 with the files and update them on the remote system
360 dit(bf(-n, --dry-run)) This tells rsync to not do any file transfers,
361 instead it will just report the actions it would have taken.
363 dit(bf(-S, --sparse)) Try to handle sparse files efficiently so they take
364 up less space on the destination.
366 NOTE: Don't use this option when the destination is a Solaris "tmpfs"
367 filesystem. It doesn't seem to handle seeks over null regions
368 correctly and ends up corrupting the files.
370 dit(bf(-x, --one-file-system)) This tells rsync not to cross filesystem
371 boundaries when recursing. This is useful for transferring the
372 contents of only one filesystem.
374 dit(bf(--delete)) This tells rsync to delete any files on the receiving
375 side that aren't on the sending side. This option can be dangerous if
378 It is a very good idea to run first using the dry run option (-n) to
379 see what files would be deleted to make sure important files aren't
382 rsync 1.6.4 changed the behavior of --delete to make it less
383 dangerous. rsync now only scans directories on the receiving side
384 that are explicitly transferred from the sending side. Only files in
385 these directories are deleted.
387 Still, it is probably easy to get burnt with this option. The moral
388 of the story is to use the -n option until you get used to the
389 behavior of --delete.
391 NOTE: It also may delete files on the destination if the sending side
392 can't open them or stat them. This is a bug that hopefully will be
393 fixed in a future release.
395 dit(bf(--force)) This options tells rsync to delete directories even if
396 they are not empty. This applies to both the --delete option and to
397 cases where rsync tries to copy a normal file but the destination
398 contains a directory of the same name. Normally rsync will refuse to
399 do a recursive directory deletion in such cases, by using --force
400 the recursive deletion will be done.
402 Use this option with caution!
404 dit(bf(-B , --block_size BLOCKSIZE)) This controls the block size used in
405 the rsync algorithm. See the technical report for details.
407 dit(bf(-e, --rsh COMMAND)) This option allows you to choose an alternative
408 remote shell program to use for communication between the local and
409 remote copies of rsync. By default rsync will use rsh, but you may
410 like to instead use ssh because of its high security.
412 You can also choose the remote shell program using the RSYNC_RSH
413 environment variable.
415 dit(bf(--rsync-path PATH)) Use this to specify the path to the copy of
416 rsync on the remote machine. Useful when its not in your path.
418 dit(bf(--exclude pattern)) This option allows you to selectively exclude
419 certain files from the list of files to be transferred. This is most
420 useful in combination with a recursive transfer.
422 You may use as many --exclude options on the command line as you like
423 to build up the list of files to exclude.
425 See the section of exclude patterns for information on the syntax of
428 dit(bf(--exclude-from FILE)) This option is similar to the --exclude
429 option, but instead it adds all filenames listed in the file FILE to
432 dit(bf(--include pattern)) This option tells rsync to not exclude the
433 specified pattern of filenames. This is useful as it allows you to
434 build up quite complex exclude/include rules.
436 See the section of exclude patterns for information on the syntax of
439 dit(bf(--include-from FILE)) This specifies a list of include patterns
442 dit(bf(-C, --cvs-exclude)) This is a useful shorthand for excluding a
443 broad range of files that you often don't want to transfer between
444 systems. It uses the same algorithm that CVS uses to determine if
445 a file should be ignored.
447 The exclude list is initialized to:
449 quote(RCS SCCS CVS CVS.adm RCSLOG cvslog.* tags TAGS .make.state
450 .nse_depinfo *~ #* .#* ,* *.old *.bak *.BAK *.orig *.rej .del-*
451 *.a *.o *.obj *.so *.Z *.elc *.ln core)
453 then files listed in a $HOME/.cvsignore are added to the list and any
454 files listed in the CVSIGNORE environment variable (space delimited).
456 Finally in each directory any files listed in the .cvsignore file in
457 that directory are added to the list.
459 dit(bf(--suffix SUFFIX)) This option allows you to override the default
460 backup suffix used with the -b option. The default is a ~.
462 dit(bf(--csum-length LENGTH)) By default the primary checksum used in
463 rsync is a very strong 16 byte MD4 checksum. In most cases you will
464 find that a truncated version of this checksum is quite efficient, and
465 this will decrease the size of the checksum data sent over the link,
466 making things faster.
468 You can choose the number of bytes in the truncated checksum using the
469 --csum-length option. Any value less than or equal to 16 is valid.
471 Note that if you use this option then you run the risk of ending up
472 with an incorrect target file. The risk with a value of 16 is
473 microscopic and can be safely ignored (the universe will probably end
474 before it fails) but with smaller values the risk is higher.
476 Current versions of rsync actually use an adaptive algorithm for the
477 checksum length by default, using a 16 byte file checksum to determine
478 if a 2nd pass is required with a longer block checksum. Only use this
479 option if you have read the source code and know what you are doing.
481 dit(bf(-T, --temp-dir DIR)) This option instructs rsync to use DIR as a
482 scratch directory when creating temporary copies of the files
483 transferred on the receiving side. The default behavior is to create
484 the temporary files in the receiving directory.
486 dit(bf(--compare-dest DIR)) This option instructs rsync to use DIR as an
487 additional directory to compare destination files against when doing
488 transfers. This is useful for doing transfers to a new destination while
489 leaving existing files intact, and then doing a flash-cutover when all
490 files have been successfully transfered (for example by moving directories
491 around and removing the old directory). This option increases the
492 usefulness of --partial because partially transferred files will remain in
493 the new temporary destination until they have a chance to be completed.
494 If DIR is a relative path, it is relative to the destination directory.
496 dit(bf(-z, --compress)) With this option, rsync compresses any data from
497 the source file(s) which it sends to the destination machine. This
498 option is useful on slow links. The compression method used is the
499 same method that gzip uses.
501 Note this this option typically achieves better compression ratios
502 that can be achieved by using a compressing remote shell, or a
503 compressing transport, as it takes advantage of the implicit
504 information sent for matching data blocks.
506 dit(bf(--numeric-ids)) With this option rsync will transfer numeric group
507 and user ids rather than using user and group names and mapping them
510 By default rsync will use the user name and group name to determine
511 what ownership to give files. The special uid 0 and the special group
512 0 and never mapped via user/group names even if the --numeric-ids
513 option is not specified.
515 If a user or group name does not exist on the destination system then
516 the numeric id from the source system is used instead.
518 dit(bf(--timeout=TIMEOUT)) This option allows you to set a maximum IO
519 timeout in seconds. If no data is transferred for the specified time
520 then rsync will exit. The default is 0, which means no timeout.
522 dit(bf(--daemon)) This tells rsync that it is to run as a rsync
523 daemon. If standard input is a socket then rsync will assume that it
524 is being run via inetd, otherwise it will detach from the current
525 terminal and become a background daemon. The daemon will read the
526 config file (/etc/rsyncd.conf) on each connect made by a client and
527 respond to requests accordingly. See the rsyncd.conf(5) man page for more
530 dit(bf(--config FILE)) This specifies an alternate config file than
531 the default /etc/rsyncd.conf. This is only relevant when --daemon is
534 dit(bf(--port PORT)) This specifies an alternate TCP port number to use
535 rather than the default port 873.
537 dit(bf(--stats)) This tells rsync to print a verbose set of statistics
538 on the file transfer, allowing you to tell how effective the rsync
539 algorithm is for your data. This option only works in conjunction with
540 the -v (verbose) option.
542 dit(bf(--progress)) This option tells rsync to print information
543 showing the progress of the transfer. This gives a bored user
548 manpagesection(EXCLUDE PATTERNS)
550 The exclude and include patterns specified to rsync allow for flexible
551 selection of what files to transfer and what files to skip.
553 rsync build a ordered list of include/exclude options as specified on
554 the command line. When a filename is encountered rsync then checks the
555 name against each exclude/include pattern in turn. The first matching
556 pattern is acted on. If it is an exclude pattern than that file is
557 skipped. If it is an include pattern then that filename is not
558 skipped. If no matching include/exclude pattern is found then the
559 filename is not skipped.
561 The patterns themselves can take several forms. The rules are:
564 it() if the pattern starts with a / then it is matched against the
565 start of the filename, otherwise it is matched against the end of
566 the filename. Thus /foo would match a file called foo
567 at the base of the tree whereas foo would match any file
568 called foo anywhere in the tree.
570 it() if the pattern ends with a / then it will only match a
571 directory, not a file, link or device.
573 it() if the pattern contains a wildcard character from the set
574 *?[ then regular expression matching is applied using the
575 normal shell filename matching rules. Otherwise a simple string
578 it() if the pattern contains a / (not counting a trailing /) then it
579 is matched against the full filename, including any leading
580 directory. If the pattern doesn't contain a / then it is matched
581 only against the final component of the filename.
583 it() if the pattern starts with "+ " (a plus followed by a space)
584 then it is always considered a include pattern, even if specified as
585 part of an exclude option. The "+ " part is discarded before matching.
587 it() if the pattern starts with "- " (a minus followed by a space)
588 then it is always considered a exclude pattern, even if specified as
589 part of an include option. The "- " part is discarded before matching.
591 it() if the pattern is a single exclamation mark ! then the current
592 exclude list is reset, removing all previous exclude patterns.
595 The +/- rules are most useful in exclude lists, allowing you to have a
596 single exclude list that contains both include and exclude options.
598 Here are some examples:
601 it() --exclude "*.o" would exclude all filenames matching *.o
602 it() --exclude "/foo" would exclude a file in the base directory called foo
603 it() --exclude "foo/" would exclude any directory called foo
604 it() --include "*/" --include "*.c" --exclude "*" would include all
605 directories and C source files.
608 manpagesection(DIAGNOSTICS)
610 rsync occasinally produces error messages that may seem a little
611 cryptic. The one that seems to cause the most confusion is "protocol
612 version mismatch - is your shell clean?".
614 This message is usually caused by your startup scripts or remote shell
615 facility producing unwanted garbage on the stream that rsync is using
616 for its transport. The way ot diagnose this problem is to run your
617 remote shell like this:
620 rsh remotehost /bin/true > out.dat
623 then look at out.dat. If everything is working correctly then out.dat
624 should be a zero length file. You you are getting the above error from
625 rsync then you will probably find that out.dat contains some text or
626 data. Look at the contents and try to work out what is producing
627 it. The most common cause is incorrectly configued shell startup
628 scripts (such as .cshrc or .profile) that contain output statements
629 for non-interactive logins.
631 manpagesection(ENVIRONMENT VARIABLES)
635 dit(bf(CVSIGNORE)) The CVSIGNORE environment variable supplements any
636 ignore patterns in .cvsignore files. See the --cvs-exclude option for
639 dit(bf(RSYNC_RSH)) The RSYNC_RSH environment variable allows you to
640 override the default shell used as the transport for rsync. This can
641 be used instead of the -e option.
643 dit(bf(RSYNC_PASSWORD)) Setting RSYNC_PASSWORD to the required
644 password allows you to run authenticated rsync connections to a rsync
645 daemon without user intervention. Note that this does not supply a
646 password to a shell transport such as ssh.
648 dit(bf(USER) or bf(LOGNAME)) The USER or LOGNAME environment variables
649 are used to determine the default username sent to a rsync server.
651 dit(bf(HOME)) The HOME environment variable is used to find the users
652 default .cvsignore file.
668 times are transferred as unix time_t values
670 file permissions, devices etc are transferred as native numerical
673 see also the comments on the --delete option
675 Please report bugs! The rsync bug tracking system is online at
676 url(http://samba.anu.edu.au/rsync/)(http://samba.anu.edu.au/rsync/)
678 manpagesection(VERSION)
679 This man page is current for version 2.0 of rsync
681 manpagesection(CREDITS)
683 rsync is distributed under the GNU public license. See the file
686 The primary ftp site for rsync is
687 url(ftp://samba.anu.edu.au/pub/rsync)(ftp://samba.anu.edu.au/pub/rsync).
689 A WEB site is available at
690 url(http://samba.anu.edu.au/rsync/)(http://samba.anu.edu.au/rsync/)
692 We would be delighted to hear from you if you like this program.
694 This program uses the zlib compression library written by Jean-loup
695 Gailly and Mark Adler.
697 manpagesection(THANKS)
699 Thanks to Richard Brent, Brendan Mackay, Bill Waite, Stephen Rothwell
700 and David Bell for helpful suggestions and testing of rsync. I've
701 probably missed some people, my apologies if I have.
706 rsync was written by Andrew Tridgell and Paul Mackerras. They may be
707 contacted via email at tridge@samba.anu.edu.au and
708 Paul.Mackerras@cs.anu.edu.au