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 - This is a very early version. Therefore it will most probably not work
71 for you, but we don't know why. It is also missing some
72 probably-critical features.
74 - It requires python >= 2.5, a C compiler, and an installed git version >=
77 - It currently only works on Linux, MacOS X >= 10.4,
78 NetBSD, Solaris, or Windows (with Cygwin). Patches to support
79 other platforms are welcome.
89 - Check out the bup source code using git:
91 git clone git://github.com/bup/bup
93 - Install the needed python libraries (including the development
96 On Debian/Ubuntu this is usually sufficient (run as root):
98 apt-get install python2.6-dev python-fuse
99 apt-get install python-pyxattr python-pylibacl
100 apt-get install linux-libc-dev
102 Substitute python2.5-dev if you have an older system. Alternately,
103 on newer Debian/Ubuntu versions, you can try this:
105 apt-get build-dep bup
107 On CentOS (for CentOS 6, at least), this should be sufficient (run
110 yum groupinstall "Development Tools"
111 yum install python python-dev
112 yum install fuse-python pyxattr pylibacl
113 yum install perl-Time-HiRes
115 In addition to the default CentOS repositories, you may need to add
116 RPMForge (for fuse-python) and EPEL (for pyxattr and pylibacl).
118 On Cygwin, install python, make, rsync, and gcc4.
120 - Build the python module and symlinks:
128 (The tests should pass. If they don't pass for you, stop here and send
131 - You can install bup via "make install", and override the default
132 destination with DESTDIR and PREFIX.
134 Files are normally installed to "$DESTDIR/$PREFIX" where DESTDIR is
135 empty by default, and PREFIX is set to /usr. So if you wanted to
136 install bup to /opt/bup, you might do something like this:
138 make install DESTDIR=/opt/bup PREFIX=''
144 Binary packages of bup are known to be built for the following OSes:
147 http://packages.debian.org/search?searchon=names&keywords=bup
149 http://packages.ubuntu.com/search?searchon=names&keywords=bup
150 - pkgsrc (NetBSD, Dragonfly, and others)
151 http://pkgsrc.se/sysutils/bup
152 http://cvsweb.netbsd.org/bsdweb.cgi/pkgsrc/sysutils/bup/
158 - Try making a local backup as a tar file:
160 tar -cvf - /etc | bup split -n local-etc -vv
162 - Try restoring your backup tarball:
164 bup join local-etc | tar -tf -
166 - Look at how much disk space your backup took:
170 - Make another backup (which should be mostly identical to the last one;
171 notice that you don't have to *specify* that this backup is incremental,
172 it just saves space automatically):
174 tar -cvf - /etc | bup split -n local-etc -vv
176 - Look how little extra space your second backup used on top of the first:
180 - Restore your old backup again (the ~1 is git notation for "one older than
183 bup join local-etc~1 | tar -tf -
185 - Get a list of your previous backups:
187 GIT_DIR=~/.bup git log local-etc
189 - Make a backup on a remote server (which must already have the 'bup' command
190 somewhere in the server's PATH (see /etc/profile, etc/environment,
191 ~/.profile, or ~/.bashrc), and be accessible via ssh.
192 Make sure to replace SERVERNAME with the actual hostname of your server):
194 tar -cvf - /etc | bup split -r SERVERNAME: -n local-etc -vv
196 - Try restoring the remote backup tarball:
198 bup join -r SERVERNAME: local-etc | tar -tf -
200 - Try using the new (slightly experimental) 'bup index' and 'bup save'
201 style backups, which bypass 'tar' but have some missing features (see
202 "Things that are stupid" below):
205 bup save -n local-etc /etc
207 - Do it again and see how fast an incremental backup can be:
210 bup save -n local-etc /etc
212 (You can also use the "-r SERVERNAME:" option to 'bup save', just like
213 with 'bup split' and 'bup join'. The index itself is always local,
214 so you don't need -r there.)
216 That's all there is to it!
222 - FreeBSD's default 'make' command doesn't like bup's Makefile. In order to
223 compile the code, run tests and install bup, you need to install GNU Make
224 from the port named 'gmake' and use its executable instead in the commands
225 seen above. (i.e. 'gmake test' runs bup's test suite)
227 - Python's development headers are automatically installed with the 'python'
228 port so there's no need to install them separately.
230 - To use the 'bup fuse' command, you need to install the fuse kernel module
231 from the 'fusefs-kmod' port in the 'sysutils' section and the libraries from
232 the port named 'py-fusefs' in the 'devel' section.
234 - The 'par2' command can be found in the port named 'par2cmdline'.
236 - In order to compile the documentation, you need pandoc which can be found in
237 the port named 'hs-pandoc' in the 'textproc' section.
240 Notes on NetBSD/pkgsrc
241 ----------------------
243 - See pkgsrc/sysutils/bup, which should be the most recent stable
244 release and includes man pages. It also has a reasonable set of
245 dependencies (git, par2, py-fuse-bindings).
247 - The "fuse-python" package referred to is hard to locate, and is a
248 separate tarball for the python language binding distributed by the
249 fuse project on sourceforge. It is available as
250 pkgsrc/filesystems/py-fuse-bindings and on NetBSD 5, "bup fuse"
253 - "bup fuse" presents every directory/file as inode 0. The directory
254 traversal code ("fts") in NetBSD's libc will interpret this as a
255 cycle and error out, so "ls -R" and "find" will not work.
257 - It is not clear if extended attribute and POSIX acl support does
264 - There is no support for ACLs. If/when some entrprising person
265 fixes this, adjust t/compare-trees.
273 bup stores its data in a git-formatted repository. Unfortunately, git
274 itself doesn't actually behave very well for bup's use case (huge numbers of
275 files, files with huge sizes, retaining file permissions/ownership are
276 important), so we mostly don't use git's *code* except for a few helper
277 programs. For example, bup has its own git packfile writer written in
280 Basically, 'bup split' reads the data on stdin (or from files specified on
281 the command line), breaks it into chunks using a rolling checksum (similar to
282 rsync), and saves those chunks into a new git packfile. There is one git
285 When deciding whether to write a particular chunk into the new packfile, bup
286 first checks all the other packfiles that exist to see if they already have that
287 chunk. If they do, the chunk is skipped.
289 git packs come in two parts: the pack itself (*.pack) and the index (*.idx).
290 The index is pretty small, and contains a list of all the objects in the
291 pack. Thus, when generating a remote backup, we don't have to have a copy
292 of the packfiles from the remote server: the local end just downloads a copy
293 of the server's *index* files, and compares objects against those when
294 generating the new pack, which it sends directly to the server.
296 The "-n" option to 'bup split' and 'bup save' is the name of the backup you
297 want to create, but it's actually implemented as a git branch. So you can
298 do cute things like checkout a particular branch using git, and receive a
299 bunch of chunk files corresponding to the file you split.
301 If you use '-b' or '-t' or '-c' instead of '-n', bup split will output a
302 list of blobs, a tree containing that list of blobs, or a commit containing
303 that tree, respectively, to stdout. You can use this to construct your own
304 scripts that do something with those values.
308 'bup index' walks through your filesystem and updates a file (whose name is,
309 by default, ~/.bup/bupindex) to contain the name, attributes, and an
310 optional git SHA1 (blob id) of each file and directory.
312 'bup save' basically just runs the equivalent of 'bup split' a whole bunch
313 of times, once per file in the index, and assembles a git tree
314 that contains all the resulting objects. Among other things, that makes
315 'git diff' much more useful (compared to splitting a tarball, which is
316 essentially a big binary blob). However, since bup splits large files into
317 smaller chunks, the resulting tree structure doesn't *exactly* correspond to
318 what git itself would have stored. Also, the tree format used by 'bup save'
319 will probably change in the future to support storing file ownership, more
320 complex file permissions, and so on.
322 If a file has previously been written by 'bup save', then its git blob/tree
323 id is stored in the index. This lets 'bup save' avoid reading that file to
324 produce future incremental backups, which means it can go *very* fast unless
325 a lot of files have changed.
328 Things that are stupid for now but which we'll fix later
329 --------------------------------------------------------
331 Help with any of these problems, or others, is very welcome. Join the
332 mailing list (see below) if you'd like to help.
334 - 'bup save' and 'bup restore' have immature metadata support.
336 On the plus side, they actually do have support now, but it's new,
337 and not remotely as well tested as tar/rsync/whatever's. If you'd
338 like to help test, please do (see t/compare-trees for one
341 In addition, at the moment, if any strip or graft-style options
342 are specified to 'bup save', then no metadata will be written for
343 the root directory. That's obviously less than ideal.
345 - bup is overly optimistic about mmap. Right now bup just assumes
346 that it can mmap as large a block as it likes, and that mmap will
347 never fail. Yeah, right... If nothing else, this has failed on
348 32-bit architectures (and 31-bit is even worse -- looking at you,
351 To fix this, we might just implement a FakeMmap[1] class that uses
352 normal file IO and handles all of the mmap methods[2] that bup
353 actually calls. Then we'd swap in one of those whenever mmap
356 This would also require implementing some of the methods needed to
357 support "[]" array access, probably at a minimum __getitem__,
358 __setitem__, and __setslice__ [3].
360 [1] http://comments.gmane.org/gmane.comp.sysutils.backup.bup/613
361 [2] http://docs.python.org/2/library/mmap.html
362 [3] http://docs.python.org/2/reference/datamodel.html#emulating-container-types
364 - 'bup index' is slower than it should be.
366 It's still rather fast: it can iterate through all the filenames on my
367 600,000 file filesystem in a few seconds. But it still needs to rewrite
368 the entire index file just to add a single filename, which is pretty
369 nasty; it should just leave the new files in a second "extra index" file
372 - bup could use inotify for *really* efficient incremental backups.
374 You could even have your system doing "continuous" backups: whenever a
375 file changes, we immediately send an image of it to the server. We could
376 give the continuous-backup process a really low CPU and I/O priority so
377 you wouldn't even know it was running.
379 - bup currently has no features that prune away *old* backups.
381 Because of the way the packfile system works, backups become "entangled"
382 in weird ways and it's not actually possible to delete one pack
383 (corresponding approximately to one backup) without risking screwing up
386 git itself has lots of ways of optimizing this sort of thing, but its
387 methods aren't really applicable here; bup packfiles are just too huge.
388 We'll have to do it in a totally different way. There are lots of
389 options. For now: make sure you've got lots of disk space :)
391 - bup has never been tested on anything but Linux, MacOS, and Windows+Cygwin.
393 There's nothing that makes it *inherently* non-portable, though, so
394 that's mostly a matter of someone putting in some effort. (For a
395 "native" Windows port, the most annoying thing is the absence of ssh in
396 a default Windows installation.)
398 - bup needs better documentation.
400 According to a recent article about git in Linux Weekly News
401 (https://lwn.net/Articles/380983/), "it's a bit short on examples and
402 a user guide would be nice." Documentation is the sort of thing that
403 will never be great unless someone from outside contributes it (since
404 the developers can never remember which parts are hard to understand).
406 - bup is "relatively speedy" and has "pretty good" compression.
408 ...according to the same LWN article. Clearly neither of those is good
409 enough. We should have awe-inspiring speed and crazy-good compression.
410 Must work on that. Writing more parts in C might help with the speed.
414 Actually, that's not stupid, but you might consider it a limitation.
415 There are a bunch of Linux GUI backup programs; someday I expect someone
416 will adapt one of them to use bup.
422 bup has an extensive set of man pages. Try using 'bup help' to get
423 started, or use 'bup help SUBCOMMAND' for any bup subcommand (like split,
424 join, index, save, etc.) to get details on that command.
426 For further technical details, please see ./DESIGN.
432 bup is a work in progress and there are many ways it can still be improved.
433 If you'd like to contribute patches, ideas, or bug reports, please join the
436 You can find the mailing list archives here:
438 http://groups.google.com/group/bup-list
440 and you can subscribe by sending a message to:
442 bup-list+subscribe@googlegroups.com
444 Please see <a href="HACKING">./HACKING</a> for
445 additional information, i.e. how to submit patches (hint - no pull
446 requests), how we handle branches, etc.