1 bup: It backs things up
2 =======================
4 bup is a program that backs things up. It's short for "backup." Can you
5 believe that nobody else has named an open source program "bup" after all
8 Despite its unassuming name, bup is pretty cool. To give you an idea of
9 just how cool it is, I wrote you this poem:
12 What rhymes with awesome?
14 But that's irrelevant.
16 Hmm. Did that help? Maybe prose is more useful after all.
19 Reasons bup is awesome
20 ----------------------
22 bup has a few advantages over other backup software:
24 - It uses a rolling checksum algorithm (similar to rsync) to split large
25 files into chunks. The most useful result of this is you can backup huge
26 virtual machine (VM) disk images, databases, and XML files incrementally,
27 even though they're typically all in one huge file, and not use tons of
28 disk space for multiple versions.
30 - It uses the packfile format from git (the open source version control
31 system), so you can access the stored data even if you don't like bup's
34 - Unlike git, it writes packfiles *directly* (instead of having a separate
35 garbage collection / repacking stage) so it's fast even with gratuitously
36 huge amounts of data. bup's improved index formats also allow you to
37 track far more filenames than git (millions) and keep track of far more
38 objects (hundreds or thousands of gigabytes).
40 - Data is "automagically" shared between incremental backups without having
41 to know which backup is based on which other one - even if the backups
42 are made from two different computers that don't even know about each
43 other. You just tell bup to back stuff up, and it saves only the minimum
44 amount of data needed.
46 - You can back up directly to a remote bup server, without needing tons of
47 temporary disk space on the computer being backed up. And if your backup
48 is interrupted halfway through, the next run will pick up where you left
49 off. And it's easy to set up a bup server: just install bup on any
50 machine where you have ssh access.
52 - Bup can use "par2" redundancy to recover corrupted backups even if your
53 disk has undetected bad sectors.
55 - Even when a backup is incremental, you don't have to worry about
56 restoring the full backup, then each of the incrementals in turn; an
57 incremental backup *acts* as if it's a full backup, it just takes less
60 - You can mount your bup repository as a FUSE filesystem and access the
61 content that way, and even export it over Samba.
63 - It's written in python (with some C parts to make it faster) so it's easy
64 for you to extend and maintain.
67 Reasons you might want to avoid bup
68 -----------------------------------
70 - It's not remotely as well tested as something like tar, so it's
71 more likely to eat your data. It's also missing some
72 probably-critical features, though fewer than it used to be.
74 - It requires python >= 2.6, a C compiler, and an installed git
75 version >= 1.5.6. It also requires par2 if you want fsck to be
76 able to generate the information needed to recover from some types
79 - It currently only works on Linux, FreeBSD, NetBSD, OS X >= 10.4,
80 Solaris, or Windows (with Cygwin, and maybe with WSL). Patches to
81 support other platforms are welcome.
83 - Any items in "Things that are stupid" below.
86 Notable changes introduced by a release
87 =======================================
89 - <a href="note/0.29.2-from-0.29.1.md">Changes in 0.29.2 as compared to 0.29.1</a>
90 - <a href="note/0.29.1-from-0.29.md">Changes in 0.29.1 as compared to 0.29</a>
91 - <a href="note/0.29-from-0.28.1.md">Changes in 0.29 as compared to 0.28.1</a>
92 - <a href="note/0.28.1-from-0.28.md">Changes in 0.28.1 as compared to 0.28</a>
93 - <a href="note/0.28-from-0.27.1.md">Changes in 0.28 as compared to 0.27.1</a>
94 - <a href="note/0.27.1-from-0.27.md">Changes in 0.27.1 as compared to 0.27</a>
97 Test status: [![Debian test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=tmp/test-cirrus&task=debian)](https://cirrus-ci.com/github/bup/bup)
98 [![FreeBSD test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=tmp/test-cirrus&task=freebsd)](https://cirrus-ci.com/github/bup/bup)
99 [![macOS test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=tmp/test-cirrus&task=macos)](https://cirrus-ci.com/github/bup/bup)
100 ==============================================================
110 - Check out the bup source code using git:
112 git clone https://github.com/bup/bup
114 - This will leave you on the master branch, which is perfect if you
115 would like to help with development, but if you'd just like to use
116 bup, please check out the latest stable release like this:
120 You can see the latest stable release here:
121 https://github.com/bup/bup/releases.
123 - Install the required python libraries (including the development
126 On very recent Debian/Ubuntu versions, this may be sufficient (run
129 apt-get build-dep bup
131 Otherwise try this (substitute python2.6-dev if you have an older
134 apt-get install python2.7-dev python-fuse
135 apt-get install python-pyxattr python-pylibacl
136 apt-get install linux-libc-dev
137 apt-get install acl attr
138 apt-get install python-tornado # optional
140 On CentOS (for CentOS 6, at least), this should be sufficient (run
143 yum groupinstall "Development Tools"
144 yum install python python-devel
145 yum install fuse-python pyxattr pylibacl
146 yum install perl-Time-HiRes
148 In addition to the default CentOS repositories, you may need to add
149 RPMForge (for fuse-python) and EPEL (for pyxattr and pylibacl).
151 On Cygwin, install python, make, rsync, and gcc4.
153 If you would like to use the optional bup web server on systems
154 without a tornado package, you may want to try this:
158 - Build the python module and symlinks:
166 or if you're in a bit more of a hurry:
170 The tests should pass. If they don't pass for you, stop here and
171 send an email to bup-list@googlegroups.com. Though if there are
172 symbolic links along the current working directory path, the tests
173 may fail. Running something like this before "make test" should
174 sidestep the problem:
178 - You can install bup via "make install", and override the default
179 destination with DESTDIR and PREFIX.
181 Files are normally installed to "$DESTDIR/$PREFIX" where DESTDIR is
182 empty by default, and PREFIX is set to /usr/local. So if you wanted to
183 install bup to /opt/bup, you might do something like this:
185 make install DESTDIR=/opt/bup PREFIX=''
187 - The Python executable that bup will use is chosen by ./configure,
188 which will search for a reasonable version unless PYTHON is set in
189 the environment, in which case, bup will use that path. You can
190 see which Python executable was chosen by looking at the
191 configure output, or examining cmd/python-cmd.sh, and you can
192 change the selection by re-running ./configure.
197 Binary packages of bup are known to be built for the following OSes:
200 http://packages.debian.org/search?searchon=names&keywords=bup
202 http://packages.ubuntu.com/search?searchon=names&keywords=bup
203 - pkgsrc (NetBSD, Dragonfly, and others)
204 http://pkgsrc.se/sysutils/bup
205 http://cvsweb.netbsd.org/bsdweb.cgi/pkgsrc/sysutils/bup/
207 https://www.archlinux.org/packages/?sort=&q=bup
209 https://apps.fedoraproject.org/packages/bup
215 - Get help for any bup command:
224 - Initialize the default BUP_DIR (~/.bup -- you can choose another by
225 either specifying `bup -d DIR ...` or setting the `BUP_DIR`
226 environment variable for a command):
230 - Make a local backup (-v or -vv will increase the verbosity):
233 bup save -n local-etc /etc
235 - Restore a local backup to ./dest:
237 bup restore -C ./dest local-etc/latest/etc
240 - Look at how much disk space your backup took:
244 - Make another backup (which should be mostly identical to the last one;
245 notice that you don't have to *specify* that this backup is incremental,
246 it just saves space automatically):
249 bup save -n local-etc /etc
251 - Look how little extra space your second backup used (on top of the first):
255 - Get a list of your previous backups:
259 - Restore your first backup again:
261 bup restore -C ./dest-2 local-etc/2013-11-23-11195/etc
263 - Make a backup to a remote server which must already have the 'bup' command
264 somewhere in its PATH (see /etc/profile, etc/environment, ~/.profile, or
265 ~/.bashrc), and be accessible via ssh.
266 Make sure to replace SERVERNAME with the actual hostname of your server:
268 bup init -r SERVERNAME:path/to/remote-bup-dir
270 bup save -r SERVERNAME:path/to/remote-bup-dir -n local-etc /etc
272 - Make a remote backup to ~/.bup on SERVER:
275 bup save -r SERVER: -n local-etc /etc
277 - See what saves are available in ~/.bup on SERVER:
281 - Restore the remote backup to ./dest:
283 bup restore -r SERVER: -C ./dest local-etc/latest/etc
286 - Defend your backups from death rays (OK fine, more likely from the
287 occasional bad disk block). This writes parity information
288 (currently via par2) for all of the existing data so that bup may
289 be able to recover from some amount of repository corruption:
293 - Use split/join instead of index/save/restore. Try making a local
296 tar -cvf - /etc | bup split -n local-etc -vv
298 - Try restoring the tarball:
300 bup join local-etc | tar -tf -
302 - Look at how much disk space your backup took:
306 - Make another tar backup:
308 tar -cvf - /etc | bup split -n local-etc -vv
310 - Look at how little extra space your second backup used on top of
315 - Restore the first tar backup again (the ~1 is git notation for "one
316 older than the most recent"):
318 bup join local-etc~1 | tar -tf -
320 - Get a list of your previous split-based backups:
322 GIT_DIR=~/.bup git log local-etc
324 - Save a tar archive to a remote server (without tar -z to facilitate
327 tar -cvf - /etc | bup split -r SERVERNAME: -n local-etc -vv
329 - Restore the archive:
331 bup join -r SERVERNAME: local-etc | tar -tf -
333 That's all there is to it!
339 - FreeBSD's default 'make' command doesn't like bup's Makefile. In order to
340 compile the code, run tests and install bup, you need to install GNU Make
341 from the port named 'gmake' and use its executable instead in the commands
342 seen above. (i.e. 'gmake test' runs bup's test suite)
344 - Python's development headers are automatically installed with the 'python'
345 port so there's no need to install them separately.
347 - To use the 'bup fuse' command, you need to install the fuse kernel module
348 from the 'fusefs-kmod' port in the 'sysutils' section and the libraries from
349 the port named 'py-fusefs' in the 'devel' section.
351 - The 'par2' command can be found in the port named 'par2cmdline'.
353 - In order to compile the documentation, you need pandoc which can be found in
354 the port named 'hs-pandoc' in the 'textproc' section.
357 Notes on NetBSD/pkgsrc
358 ----------------------
360 - See pkgsrc/sysutils/bup, which should be the most recent stable
361 release and includes man pages. It also has a reasonable set of
362 dependencies (git, par2, py-fuse-bindings).
364 - The "fuse-python" package referred to is hard to locate, and is a
365 separate tarball for the python language binding distributed by the
366 fuse project on sourceforge. It is available as
367 pkgsrc/filesystems/py-fuse-bindings and on NetBSD 5, "bup fuse"
370 - "bup fuse" presents every directory/file as inode 0. The directory
371 traversal code ("fts") in NetBSD's libc will interpret this as a
372 cycle and error out, so "ls -R" and "find" will not work.
374 - There is no support for ACLs. If/when some entrprising person
375 fixes this, adjust t/compare-trees.
381 - There is no support for ACLs. If/when some enterprising person
382 fixes this, adjust t/compare-trees.
384 - In t/test.sh, two tests have been disabled. These tests check to
385 see that repeated saves produce identical trees and that an
386 intervening index doesn't change the SHA1. Apparently Cygwin has
387 some unusual behaviors with respect to access times (that probably
388 warrant further investigation). Possibly related:
389 http://cygwin.com/ml/cygwin/2007-06/msg00436.html
395 - There is no support for ACLs. If/when some enterprising person
396 fixes this, adjust t/compare-trees.
405 bup stores its data in a git-formatted repository. Unfortunately, git
406 itself doesn't actually behave very well for bup's use case (huge numbers of
407 files, files with huge sizes, retaining file permissions/ownership are
408 important), so we mostly don't use git's *code* except for a few helper
409 programs. For example, bup has its own git packfile writer written in
412 Basically, 'bup split' reads the data on stdin (or from files specified on
413 the command line), breaks it into chunks using a rolling checksum (similar to
414 rsync), and saves those chunks into a new git packfile. There is at least one
415 git packfile per backup.
417 When deciding whether to write a particular chunk into the new packfile, bup
418 first checks all the other packfiles that exist to see if they already have that
419 chunk. If they do, the chunk is skipped.
421 git packs come in two parts: the pack itself (*.pack) and the index (*.idx).
422 The index is pretty small, and contains a list of all the objects in the
423 pack. Thus, when generating a remote backup, we don't have to have a copy
424 of the packfiles from the remote server: the local end just downloads a copy
425 of the server's *index* files, and compares objects against those when
426 generating the new pack, which it sends directly to the server.
428 The "-n" option to 'bup split' and 'bup save' is the name of the backup you
429 want to create, but it's actually implemented as a git branch. So you can
430 do cute things like checkout a particular branch using git, and receive a
431 bunch of chunk files corresponding to the file you split.
433 If you use '-b' or '-t' or '-c' instead of '-n', bup split will output a
434 list of blobs, a tree containing that list of blobs, or a commit containing
435 that tree, respectively, to stdout. You can use this to construct your own
436 scripts that do something with those values.
441 'bup index' walks through your filesystem and updates a file (whose name is,
442 by default, ~/.bup/bupindex) to contain the name, attributes, and an
443 optional git SHA1 (blob id) of each file and directory.
445 'bup save' basically just runs the equivalent of 'bup split' a whole bunch
446 of times, once per file in the index, and assembles a git tree
447 that contains all the resulting objects. Among other things, that makes
448 'git diff' much more useful (compared to splitting a tarball, which is
449 essentially a big binary blob). However, since bup splits large files into
450 smaller chunks, the resulting tree structure doesn't *exactly* correspond to
451 what git itself would have stored. Also, the tree format used by 'bup save'
452 will probably change in the future to support storing file ownership, more
453 complex file permissions, and so on.
455 If a file has previously been written by 'bup save', then its git blob/tree
456 id is stored in the index. This lets 'bup save' avoid reading that file to
457 produce future incremental backups, which means it can go *very* fast unless
458 a lot of files have changed.
461 Things that are stupid for now but which we'll fix later
462 ========================================================
464 Help with any of these problems, or others, is very welcome. Join the
465 mailing list (see below) if you'd like to help.
467 - 'bup save' and 'bup restore' have immature metadata support.
469 On the plus side, they actually do have support now, but it's new,
470 and not remotely as well tested as tar/rsync/whatever's. However,
471 you have to start somewhere, and as of 0.25, we think it's ready
472 for more general use. Please let us know if you have any trouble.
474 Also, if any strip or graft-style options are specified to 'bup
475 save', then no metadata will be written for the root directory.
476 That's obviously less than ideal.
478 - bup is overly optimistic about mmap. Right now bup just assumes
479 that it can mmap as large a block as it likes, and that mmap will
480 never fail. Yeah, right... If nothing else, this has failed on
481 32-bit architectures (and 31-bit is even worse -- looking at you,
484 To fix this, we might just implement a FakeMmap[1] class that uses
485 normal file IO and handles all of the mmap methods[2] that bup
486 actually calls. Then we'd swap in one of those whenever mmap
489 This would also require implementing some of the methods needed to
490 support "[]" array access, probably at a minimum __getitem__,
491 __setitem__, and __setslice__ [3].
493 [1] http://comments.gmane.org/gmane.comp.sysutils.backup.bup/613
494 [2] http://docs.python.org/2/library/mmap.html
495 [3] http://docs.python.org/2/reference/datamodel.html#emulating-container-types
497 - 'bup index' is slower than it should be.
499 It's still rather fast: it can iterate through all the filenames on my
500 600,000 file filesystem in a few seconds. But it still needs to rewrite
501 the entire index file just to add a single filename, which is pretty
502 nasty; it should just leave the new files in a second "extra index" file
505 - bup could use inotify for *really* efficient incremental backups.
507 You could even have your system doing "continuous" backups: whenever a
508 file changes, we immediately send an image of it to the server. We could
509 give the continuous-backup process a really low CPU and I/O priority so
510 you wouldn't even know it was running.
512 - bup only has experimental support for pruning old backups.
514 While you should now be able to drop old saves and branches with
515 `bup rm`, and reclaim the space occupied by data that's no longer
516 needed by other backups with `bup gc`, these commands are
517 experimental, and should be handled with great care. See the
518 man pages for more information.
520 Unless you want to help test the new commands, one possible
521 workaround is to just start a new BUP_DIR occasionally,
522 i.e. bup-2013, bup-2014...
524 - bup has never been tested on anything but Linux, FreeBSD, NetBSD,
525 OS X, and Windows+Cygwin.
527 There's nothing that makes it *inherently* non-portable, though, so
528 that's mostly a matter of someone putting in some effort. (For a
529 "native" Windows port, the most annoying thing is the absence of ssh in
530 a default Windows installation.)
532 - bup needs better documentation.
534 According to an article about bup in Linux Weekly News
535 (https://lwn.net/Articles/380983/), "it's a bit short on examples and
536 a user guide would be nice." Documentation is the sort of thing that
537 will never be great unless someone from outside contributes it (since
538 the developers can never remember which parts are hard to understand).
540 - bup is "relatively speedy" and has "pretty good" compression.
542 ...according to the same LWN article. Clearly neither of those is good
543 enough. We should have awe-inspiring speed and crazy-good compression.
544 Must work on that. Writing more parts in C might help with the speed.
548 Actually, that's not stupid, but you might consider it a
549 limitation. See the ["Related Projects"](https://bup.github.io/)
550 list for some possible options.
555 bup has an extensive set of man pages. Try using 'bup help' to get
556 started, or use 'bup help SUBCOMMAND' for any bup subcommand (like split,
557 join, index, save, etc.) to get details on that command.
559 For further technical details, please see ./DESIGN.
565 bup is a work in progress and there are many ways it can still be improved.
566 If you'd like to contribute patches, ideas, or bug reports, please join the
569 You can find the mailing list archives here:
571 http://groups.google.com/group/bup-list
573 and you can subscribe by sending a message to:
575 bup-list+subscribe@googlegroups.com
577 Please see <a href="HACKING">./HACKING</a> for
578 additional information, i.e. how to submit patches (hint - no pull
579 requests), how we handle branches, etc.