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 - Until resolved, a [glibc bug](https://sourceware.org/bugzilla/show_bug.cgi?id=26034)
84 might cause bup to crash on startup for some (unusual) command line
85 argument values, when bup is configured to use Python 3.
87 - Any items in "Things that are stupid" below.
90 Notable changes introduced by a release
91 =======================================
93 - <a href="note/0.30.1-from-0.30.md">Changes in 0.30.1 as compared to 0.30</a>
94 - <a href="note/0.30-from-0.29.3.md">Changes in 0.30 as compared to 0.29.3</a>
95 - <a href="note/0.29.3-from-0.29.2.md">Changes in 0.29.3 as compared to 0.29.2</a>
96 - <a href="note/0.29.2-from-0.29.1.md">Changes in 0.29.2 as compared to 0.29.1</a>
97 - <a href="note/0.29.1-from-0.29.md">Changes in 0.29.1 as compared to 0.29</a>
98 - <a href="note/0.29-from-0.28.1.md">Changes in 0.29 as compared to 0.28.1</a>
99 - <a href="note/0.28.1-from-0.28.md">Changes in 0.28.1 as compared to 0.28</a>
100 - <a href="note/0.28-from-0.27.1.md">Changes in 0.28 as compared to 0.27.1</a>
101 - <a href="note/0.27.1-from-0.27.md">Changes in 0.27.1 as compared to 0.27</a>
107 | branch | Debian | FreeBSD | macOS |
108 |--------|------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------|
109 | master | [![Debian test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=master&task=debian)](https://cirrus-ci.com/github/bup/bup) | [![FreeBSD test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=master&task=freebsd)](https://cirrus-ci.com/github/bup/bup) | [![macOS test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=master&task=macos)](https://cirrus-ci.com/github/bup/bup) |
110 | 0.30.x | [![Debian test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.30.x&task=debian)](https://cirrus-ci.com/github/bup/bup) | [![FreeBSD test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.30.x&task=freebsd)](https://cirrus-ci.com/github/bup/bup) | [![macOS test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.30.x&task=macos)](https://cirrus-ci.com/github/bup/bup) |
111 | 0.29.x | [![Debian test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.29.x&task=debian)](https://cirrus-ci.com/github/bup/bup) | [![FreeBSD test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.29.x&task=freebsd)](https://cirrus-ci.com/github/bup/bup) | [![macOS test status](https://api.cirrus-ci.com/github/bup/bup.svg?branch=0.29.x&task=macos)](https://cirrus-ci.com/github/bup/bup) |
119 - Check out the bup source code using git:
122 git clone https://github.com/bup/bup
125 - This will leave you on the master branch, which is perfect if you
126 would like to help with development, but if you'd just like to use
127 bup, please check out the latest stable release like this:
133 You can see the latest stable release here:
134 https://github.com/bup/bup/releases.
136 - Install the required python libraries (including the development
139 On very recent Debian/Ubuntu versions, this may be sufficient (run
143 apt-get build-dep bup
146 Otherwise try this (substitute python2.6-dev if you have an older
150 apt-get install python2.7-dev python-fuse
151 apt-get install python-pyxattr
152 apt-get install pkg-config linux-libc-dev libacl1-dev
153 apt-get install acl attr
154 apt-get install libreadline-dev # optional (bup ftp)
155 apt-get install python-tornado # optional (bup web)
158 On CentOS (for CentOS 6, at least), this should be sufficient (run
162 yum groupinstall "Development Tools"
163 yum install python python-devel libacl-devel
164 yum install fuse-python pyxattr
165 yum install perl-Time-HiRes
166 yum install readline-devel # optional (bup ftp)
167 yum install python-tornado # optional (bup web)
170 In addition to the default CentOS repositories, you may need to add
171 RPMForge (for fuse-python) and EPEL (for pyxattr).
173 On Cygwin, install python, make, rsync, and gcc4.
175 If you would like to use the optional bup web server on systems
176 without a tornado package, you may want to try this:
182 - Build the python module and symlinks:
194 or if you're in a bit more of a hurry:
200 The tests should pass. If they don't pass for you, stop here and
201 send an email to bup-list@googlegroups.com. Though if there are
202 symbolic links along the current working directory path, the tests
203 may fail. Running something like this before "make test" should
204 sidestep the problem:
210 - You can install bup via "make install", and override the default
211 destination with DESTDIR and PREFIX.
213 Files are normally installed to "$DESTDIR/$PREFIX" where DESTDIR is
214 empty by default, and PREFIX is set to /usr/local. So if you wanted to
215 install bup to /opt/bup, you might do something like this:
218 make install DESTDIR=/opt/bup PREFIX=''
221 - The Python executable that bup will use is chosen by ./configure,
222 which will search for a reasonable version unless PYTHON is set in
223 the environment, in which case, bup will use that path. You can
224 see which Python executable was chosen by looking at the
225 configure output, or examining cmd/python-cmd.sh, and you can
226 change the selection by re-running ./configure.
231 Binary packages of bup are known to be built for the following OSes:
234 http://packages.debian.org/search?searchon=names&keywords=bup
236 http://packages.ubuntu.com/search?searchon=names&keywords=bup
237 - pkgsrc (NetBSD, Dragonfly, and others)
238 http://pkgsrc.se/sysutils/bup
239 http://cvsweb.netbsd.org/bsdweb.cgi/pkgsrc/sysutils/bup/
241 https://www.archlinux.org/packages/?sort=&q=bup
243 https://apps.fedoraproject.org/packages/bup
249 - Get help for any bup command:
260 - Initialize the default BUP_DIR (~/.bup -- you can choose another by
261 either specifying `bup -d DIR ...` or setting the `BUP_DIR`
262 environment variable for a command):
268 - Make a local backup (-v or -vv will increase the verbosity):
272 bup save -n local-etc /etc
275 - Restore a local backup to ./dest:
278 bup restore -C ./dest local-etc/latest/etc
282 - Look at how much disk space your backup took:
288 - Make another backup (which should be mostly identical to the last one;
289 notice that you don't have to *specify* that this backup is incremental,
290 it just saves space automatically):
294 bup save -n local-etc /etc
297 - Look how little extra space your second backup used (on top of the first):
303 - Get a list of your previous backups:
309 - Restore your first backup again:
312 bup restore -C ./dest-2 local-etc/2013-11-23-11195/etc
315 - Make a backup to a remote server which must already have the 'bup' command
316 somewhere in its PATH (see /etc/profile, etc/environment, ~/.profile, or
317 ~/.bashrc), and be accessible via ssh.
318 Make sure to replace SERVERNAME with the actual hostname of your server:
321 bup init -r SERVERNAME:path/to/remote-bup-dir
323 bup save -r SERVERNAME:path/to/remote-bup-dir -n local-etc /etc
326 - Make a remote backup to ~/.bup on SERVER:
330 bup save -r SERVER: -n local-etc /etc
333 - See what saves are available in ~/.bup on SERVER:
339 - Restore the remote backup to ./dest:
342 bup restore -r SERVER: -C ./dest local-etc/latest/etc
346 - Defend your backups from death rays (OK fine, more likely from the
347 occasional bad disk block). This writes parity information
348 (currently via par2) for all of the existing data so that bup may
349 be able to recover from some amount of repository corruption:
355 - Use split/join instead of index/save/restore. Try making a local
359 tar -cvf - /etc | bup split -n local-etc -vv
362 - Try restoring the tarball:
365 bup join local-etc | tar -tf -
368 - Look at how much disk space your backup took:
374 - Make another tar backup:
377 tar -cvf - /etc | bup split -n local-etc -vv
380 - Look at how little extra space your second backup used on top of
387 - Restore the first tar backup again (the ~1 is git notation for "one
388 older than the most recent"):
391 bup join local-etc~1 | tar -tf -
394 - Get a list of your previous split-based backups:
397 GIT_DIR=~/.bup git log local-etc
400 - Save a tar archive to a remote server (without tar -z to facilitate
404 tar -cvf - /etc | bup split -r SERVERNAME: -n local-etc -vv
407 - Restore the archive:
410 bup join -r SERVERNAME: local-etc | tar -tf -
413 That's all there is to it!
419 - FreeBSD's default 'make' command doesn't like bup's Makefile. In order to
420 compile the code, run tests and install bup, you need to install GNU Make
421 from the port named 'gmake' and use its executable instead in the commands
422 seen above. (i.e. 'gmake test' runs bup's test suite)
424 - Python's development headers are automatically installed with the 'python'
425 port so there's no need to install them separately.
427 - To use the 'bup fuse' command, you need to install the fuse kernel module
428 from the 'fusefs-kmod' port in the 'sysutils' section and the libraries from
429 the port named 'py-fusefs' in the 'devel' section.
431 - The 'par2' command can be found in the port named 'par2cmdline'.
433 - In order to compile the documentation, you need pandoc which can be found in
434 the port named 'hs-pandoc' in the 'textproc' section.
437 Notes on NetBSD/pkgsrc
438 ----------------------
440 - See pkgsrc/sysutils/bup, which should be the most recent stable
441 release and includes man pages. It also has a reasonable set of
442 dependencies (git, par2, py-fuse-bindings).
444 - The "fuse-python" package referred to is hard to locate, and is a
445 separate tarball for the python language binding distributed by the
446 fuse project on sourceforge. It is available as
447 pkgsrc/filesystems/py-fuse-bindings and on NetBSD 5, "bup fuse"
450 - "bup fuse" presents every directory/file as inode 0. The directory
451 traversal code ("fts") in NetBSD's libc will interpret this as a
452 cycle and error out, so "ls -R" and "find" will not work.
454 - There is no support for ACLs. If/when some enterprising person
455 fixes this, adjust t/compare-trees.
461 - There is no support for ACLs. If/when some enterprising person
462 fixes this, adjust t/compare-trees.
464 - In t/test.sh, two tests have been disabled. These tests check to
465 see that repeated saves produce identical trees and that an
466 intervening index doesn't change the SHA1. Apparently Cygwin has
467 some unusual behaviors with respect to access times (that probably
468 warrant further investigation). Possibly related:
469 http://cygwin.com/ml/cygwin/2007-06/msg00436.html
475 - There is no support for ACLs. If/when some enterprising person
476 fixes this, adjust t/compare-trees.
485 bup stores its data in a git-formatted repository. Unfortunately, git
486 itself doesn't actually behave very well for bup's use case (huge numbers of
487 files, files with huge sizes, retaining file permissions/ownership are
488 important), so we mostly don't use git's *code* except for a few helper
489 programs. For example, bup has its own git packfile writer written in
492 Basically, 'bup split' reads the data on stdin (or from files specified on
493 the command line), breaks it into chunks using a rolling checksum (similar to
494 rsync), and saves those chunks into a new git packfile. There is at least one
495 git packfile per backup.
497 When deciding whether to write a particular chunk into the new packfile, bup
498 first checks all the other packfiles that exist to see if they already have that
499 chunk. If they do, the chunk is skipped.
501 git packs come in two parts: the pack itself (*.pack) and the index (*.idx).
502 The index is pretty small, and contains a list of all the objects in the
503 pack. Thus, when generating a remote backup, we don't have to have a copy
504 of the packfiles from the remote server: the local end just downloads a copy
505 of the server's *index* files, and compares objects against those when
506 generating the new pack, which it sends directly to the server.
508 The "-n" option to 'bup split' and 'bup save' is the name of the backup you
509 want to create, but it's actually implemented as a git branch. So you can
510 do cute things like checkout a particular branch using git, and receive a
511 bunch of chunk files corresponding to the file you split.
513 If you use '-b' or '-t' or '-c' instead of '-n', bup split will output a
514 list of blobs, a tree containing that list of blobs, or a commit containing
515 that tree, respectively, to stdout. You can use this to construct your own
516 scripts that do something with those values.
521 'bup index' walks through your filesystem and updates a file (whose name is,
522 by default, ~/.bup/bupindex) to contain the name, attributes, and an
523 optional git SHA1 (blob id) of each file and directory.
525 'bup save' basically just runs the equivalent of 'bup split' a whole bunch
526 of times, once per file in the index, and assembles a git tree
527 that contains all the resulting objects. Among other things, that makes
528 'git diff' much more useful (compared to splitting a tarball, which is
529 essentially a big binary blob). However, since bup splits large files into
530 smaller chunks, the resulting tree structure doesn't *exactly* correspond to
531 what git itself would have stored. Also, the tree format used by 'bup save'
532 will probably change in the future to support storing file ownership, more
533 complex file permissions, and so on.
535 If a file has previously been written by 'bup save', then its git blob/tree
536 id is stored in the index. This lets 'bup save' avoid reading that file to
537 produce future incremental backups, which means it can go *very* fast unless
538 a lot of files have changed.
541 Things that are stupid for now but which we'll fix later
542 ========================================================
544 Help with any of these problems, or others, is very welcome. Join the
545 mailing list (see below) if you'd like to help.
547 - 'bup save' and 'bup restore' have immature metadata support.
549 On the plus side, they actually do have support now, but it's new,
550 and not remotely as well tested as tar/rsync/whatever's. However,
551 you have to start somewhere, and as of 0.25, we think it's ready
552 for more general use. Please let us know if you have any trouble.
554 Also, if any strip or graft-style options are specified to 'bup
555 save', then no metadata will be written for the root directory.
556 That's obviously less than ideal.
558 - bup is overly optimistic about mmap. Right now bup just assumes
559 that it can mmap as large a block as it likes, and that mmap will
560 never fail. Yeah, right... If nothing else, this has failed on
561 32-bit architectures (and 31-bit is even worse -- looking at you,
564 To fix this, we might just implement a FakeMmap[1] class that uses
565 normal file IO and handles all of the mmap methods[2] that bup
566 actually calls. Then we'd swap in one of those whenever mmap
569 This would also require implementing some of the methods needed to
570 support "[]" array access, probably at a minimum __getitem__,
571 __setitem__, and __setslice__ [3].
573 [1] http://comments.gmane.org/gmane.comp.sysutils.backup.bup/613
574 [2] http://docs.python.org/2/library/mmap.html
575 [3] http://docs.python.org/2/reference/datamodel.html#emulating-container-types
577 - 'bup index' is slower than it should be.
579 It's still rather fast: it can iterate through all the filenames on my
580 600,000 file filesystem in a few seconds. But it still needs to rewrite
581 the entire index file just to add a single filename, which is pretty
582 nasty; it should just leave the new files in a second "extra index" file
585 - bup could use inotify for *really* efficient incremental backups.
587 You could even have your system doing "continuous" backups: whenever a
588 file changes, we immediately send an image of it to the server. We could
589 give the continuous-backup process a really low CPU and I/O priority so
590 you wouldn't even know it was running.
592 - bup only has experimental support for pruning old backups.
594 While you should now be able to drop old saves and branches with
595 `bup rm`, and reclaim the space occupied by data that's no longer
596 needed by other backups with `bup gc`, these commands are
597 experimental, and should be handled with great care. See the
598 man pages for more information.
600 Unless you want to help test the new commands, one possible
601 workaround is to just start a new BUP_DIR occasionally,
602 i.e. bup-2013, bup-2014...
604 - bup has never been tested on anything but Linux, FreeBSD, NetBSD,
605 OS X, and Windows+Cygwin.
607 There's nothing that makes it *inherently* non-portable, though, so
608 that's mostly a matter of someone putting in some effort. (For a
609 "native" Windows port, the most annoying thing is the absence of ssh in
610 a default Windows installation.)
612 - bup needs better documentation.
614 According to an article about bup in Linux Weekly News
615 (https://lwn.net/Articles/380983/), "it's a bit short on examples and
616 a user guide would be nice." Documentation is the sort of thing that
617 will never be great unless someone from outside contributes it (since
618 the developers can never remember which parts are hard to understand).
620 - bup is "relatively speedy" and has "pretty good" compression.
622 ...according to the same LWN article. Clearly neither of those is good
623 enough. We should have awe-inspiring speed and crazy-good compression.
624 Must work on that. Writing more parts in C might help with the speed.
628 Actually, that's not stupid, but you might consider it a
629 limitation. See the ["Related Projects"](https://bup.github.io/)
630 list for some possible options.
635 bup has an extensive set of man pages. Try using 'bup help' to get
636 started, or use 'bup help SUBCOMMAND' for any bup subcommand (like split,
637 join, index, save, etc.) to get details on that command.
639 For further technical details, please see ./DESIGN.
645 bup is a work in progress and there are many ways it can still be improved.
646 If you'd like to contribute patches, ideas, or bug reports, please join the
649 You can find the mailing list archives here:
651 http://groups.google.com/group/bup-list
653 and you can subscribe by sending a message to:
655 bup-list+subscribe@googlegroups.com
657 Please see <a href="HACKING">./HACKING</a> for
658 additional information, i.e. how to submit patches (hint - no pull
659 requests), how we handle branches, etc.