bup is a program that backs things up. It's short for "backup." Can you believe that nobody else has named an open source program "bup" after all this time? Me neither.
Despite its unassuming name, bup is pretty cool. To give you an idea of just how cool it is, I wrote you this poem:
Bup is teh awesome
What rhymes with awesome?
I guess maybe possum
But that's irrelevant.
Hmm. Did that help? Maybe prose is more useful after all.
bup has a few advantages over other backup software:
-
It uses a rolling checksum algorithm (similar to rsync) to split large files into chunks. The most useful result of this is you can backup huge virtual machine (VM) disk images, databases, and XML files incrementally, even though they're typically all in one huge file, and not use tons of disk space for multiple versions.
-
It uses the packfile format from git (the open source version control system), so you can access the stored data even if you don't like bup's user interface.
-
Unlike git, it writes packfiles directly (instead of having a separate garbage collection / repacking stage) so it's fast even with gratuitously huge amounts of data. bup's improved index formats also allow you to track far more filenames than git (millions) and keep track of far more objects (hundreds or thousands of gigabytes).
-
Data is "automagically" shared between incremental backups without having to know which backup is based on which other one - even if the backups are made from two different computers that don't even know about each other. You just tell bup to back stuff up, and it saves only the minimum amount of data needed.
-
You can back up directly to a remote bup server, without needing tons of temporary disk space on the computer being backed up. And if your backup is interrupted halfway through, the next run will pick up where you left off. And it's easy to set up a bup server: just install bup on any machine where you have ssh access.
-
Bup can use "par2" redundancy to recover corrupted backups even if your disk has undetected bad sectors.
-
Even when a backup is incremental, you don't have to worry about restoring the full backup, then each of the incrementals in turn; an incremental backup acts as if it's a full backup, it just takes less disk space.
-
You can mount your bup repository as a FUSE filesystem and access the content that way, and even export it over Samba.
-
It's written in python (with some C parts to make it faster) so it's easy for you to extend and maintain.
-
It's not remotely as well tested as something like tar, so it's more likely to eat your data. It's also missing some probably-critical features, though fewer than it used to be.
-
It requires python 3.7 or newer, a C compiler, and an installed git version >= 1.7.2. It also requires par2 if you want fsck to be able to generate the information needed to recover from some types of corruption.
-
It currently only works on Linux, FreeBSD, NetBSD, OS X >= 10.4, Solaris, or Windows (with Cygwin, and WSL). Patches to support other platforms are welcome.
-
Any items in "Things that are stupid" below.
- Changes in 0.33.7 as compared to 0.33.6
- Changes in 0.33.6 as compared to 0.33.5
- Changes in 0.33.5 as compared to 0.33.4
- Changes in 0.33.4 as compared to 0.33.3
- Changes in 0.33.3 as compared to 0.33.2
- Changes in 0.33.2 as compared to 0.33.1
- Changes in 0.33.1 as compared to 0.33
- Changes in 0.33 as compared to 0.32
- Changes in 0.32.1 as compared to 0.32
- Changes in 0.32 as compared to 0.31
- Changes in 0.31 as compared to 0.30.1
- Changes in 0.30.1 as compared to 0.30
- Changes in 0.30 as compared to 0.29.3
- Changes in 0.29.3 as compared to 0.29.2
- Changes in 0.29.2 as compared to 0.29.1
- Changes in 0.29.1 as compared to 0.29
- Changes in 0.29 as compared to 0.28.1
- Changes in 0.28.1 as compared to 0.28
- Changes in 0.28 as compared to 0.27.1
- Changes in 0.27.1 as compared to 0.27
| main |
|---|
 |
-
Check out the bup source code using git (for Cygwin, use a Cygwin installed git so that symlinks will work by defaul):
git clone https://github.com/bup/bup
-
This will leave you on the main branch, which is perfect if you would like to help with development, but if you'd just like to use bup, please check out the latest stable release like this:
git checkout 0.33.7
You can see the latest stable release here: https://github.com/bup/bup/tags
-
Ensure you have a Python 3.7+ development environment available and install the required python libraries (including the development libraries).
See the relevant platform specific information below, if available, and if not, refer to the Debian notes to get a general idea of what's required. You might also want to look at the
dev/prep-for-*scripts. Those include all the commands used to prepare to build bup for testing on the platform.For
bup fuseyou will need to install python-fuse rather than fusepy. For example, in Debian, install python3-fuse rather than python3-fusepy.If you would like to use the optional bup web server on systems without a tornado package, you may want to try this:
pip install tornado
-
Build (below, replace
makewith whatever GNU make is called on the current system if it isn'tmake, e.g. commonlygmake):make
At the moment the build treats compiler warnings as errors. If the build fails as a result, try this:
CFLAGS=-Wno-error ./configure make
-
Run the tests:
make long-check
or if you're in a bit more of a hurry:
make check
If you have the Python xdist module installed, then you can probably run the tests faster by adding the make -j option (see ./HACKING for additional information):
make -j check
The tests should pass (with some skipped tests that weren't applicable in your environment). If they don't pass for you, stop here and send an email to bup-list@googlegroups.com. Though if there are symbolic links along the current working directory path, the tests may fail. Running something like this before "make test" should sidestep the problem:
cd "$(pwd -P)"
-
You can install bup via "make install", and override the default destination with DESTDIR and PREFIX.
Files are normally installed to "$DESTDIR/$PREFIX" where DESTDIR is empty by default, and PREFIX is set to /usr/local. So if you wanted to install bup to /opt/bup, you might do something like this:
make install DESTDIR=/opt/bup PREFIX='' -
The Python version that bup will use is determined by the
python-configprogram chosen by./configure, which will search for a reasonable version unlessBUP_PYTHON_CONFIGis set in the environment. You can see which Python executable was chosen by looking at the configure output, or examiningconfig/config.var/bup-python-config, and you can change the selection by re-running./configure. -
If you want to specify your own
CPPFLAGS,CFLAGS, orLDFLAGS, you can set them for individualmakeinvocations, e.g.make CFLAGS=-O0 check, or persistently via./configurewithCFLAGS=-O0 ./configure. At the moment,make cleanclears the configuration, but we may change that at some point, perhaps by adding and requiring amake distcleanto clear the configuration.
Binary packages of bup are known to be built for the following OSes:
-
Get help for any bup command:
bup help bup help init bup help index bup help save bup help restore ...
-
Initialize the default bup repository (~/.bup -- you can choose another by either specifying
bup -d DIR ...or setting theBUP_DIRenvironment variable for a command):bup init
-
Make a local backup (-v or -vv will increase the verbosity):
bup index /etc bup save -n local-etc /etc
-
Restore a local backup to ./dest:
bup restore -C ./dest local-etc/latest/etc ls -l dest/etc
-
Look at how much disk space your backup took:
du -s ~/.bup -
Make another backup (which should be mostly identical to the last one; notice that you don't have to specify that this backup is incremental, it just saves space automatically):
bup index /etc bup save -n local-etc /etc
-
Look how little extra space your second backup used (on top of the first):
du -s ~/.bup -
Get a list of your previous backups:
bup ls local-etc
-
Restore your first backup again:
bup restore -C ./dest-2 local-etc/2013-11-23-11195/etc
-
Make a backup to a remote server which must already have the 'bup' command somewhere in its PATH (see /etc/profile, etc/environment, ~/.profile, or ~/.bashrc), and be accessible via ssh. Make sure to replace SERVERNAME with the actual hostname of your server:
bup init -r SERVERNAME:path/to/remote-bup-dir bup index /etc bup save -r SERVERNAME:path/to/remote-bup-dir -n local-etc /etc
-
Make a remote backup to ~/.bup on SERVER:
bup index /etc bup save -r SERVER: -n local-etc /etc
-
See what saves are available in ~/.bup on SERVER:
bup ls -r SERVER:
-
Restore the remote backup to ./dest:
bup restore -r SERVER: -C ./dest local-etc/latest/etc ls -l dest/etc
-
Defend your backups from death rays (OK fine, more likely from the occasional bad disk block). This writes parity information (currently via par2) for all of the existing data so that bup may be able to recover from some amount of repository corruption:
bup fsck -g
-
Use split/join instead of index/save/restore. Try making a local backup using tar:
tar -cvf - /etc | bup split -n local-etc -vv -
Try restoring the tarball:
bup join local-etc | tar -tf - -
Look at how much disk space your backup took:
du -s ~/.bup -
Make another tar backup:
tar -cvf - /etc | bup split -n local-etc -vv -
Look at how little extra space your second backup used on top of the first:
du -s ~/.bup -
Restore the first tar backup again (the ~1 is git notation for "one older than the most recent"):
bup join local-etc~1 | tar -tf - -
Get a list of your previous split-based backups:
GIT_DIR=~/.bup git log local-etc -
Save a tar archive to a remote server (without tar -z to facilitate deduplication):
tar -cvf - /etc | bup split -r SERVERNAME: -n local-etc -vv -
Restore the archive:
bup join -r SERVERNAME: local-etc | tar -tf -
That's all there is to it!
If your distribution is recent enough, or includes new enough apt
sources, this may be sufficient (run as root):
apt-get build-dep bupOtherwise try this:
apt-get install python3-dev python3-fuse
apt-get install python3-pyxattr python3-pytest
apt-get install python3-distutils
apt-get install pkg-config linux-libc-dev libacl1-dev
apt-get install gcc make acl attr rsync
apt-get isntall python3-pytest-xdist # optional (parallel tests)
apt-get install par2 # optional (error correction)
apt-get install libreadline-dev # optional (bup ftp)
apt-get install python3-tornado # optional (bup web)-
In order to compile the code, run tests and install bup, you need to install and run GNU Make from the
gmakeport. -
Python's development headers are automatically installed with the 'python' port so there's no need to install them separately.
-
To use the 'bup fuse' command, you need to install the fuse kernel module from the 'fusefs-kmod' port in the 'sysutils' section and the libraries from the port named 'py-fusefs' in the 'devel' section.
-
The 'par2' command can be found in the port named 'par2cmdline'.
-
In order to compile the documentation, you need pandoc which can be found in the port named 'hs-pandoc' in the 'textproc' section.
-
See pkgsrc/sysutils/bup, which should be the most recent stable release and includes man pages. It also has a reasonable set of dependencies (git, par2, py-fuse-bindings).
-
The "fuse-python" package referred to is hard to locate, and is a separate tarball for the python language binding distributed by the fuse project on sourceforge. It is available as pkgsrc/filesystems/py-fuse-bindings and on NetBSD 5, "bup fuse" works with it.
-
"bup fuse" presents every directory/file as inode 0. The directory traversal code ("fts") in NetBSD's libc will interpret this as a cycle and error out, so "ls -R" and "find" will not work.
-
There is no support for ACLs. If/when some enterprising person fixes this, adjust dev/compare-trees.
-
To prepare to build build bup, install the gcc-core, git, python3-devel, make, and rsync packages. Optionally, install libreadline-devel (for
bup ftp) and par2. To generate the help pages, install pandoc (outside of Cygwin).Use a Cygwin installed git so that symlinks will work by default. If you try some other git version, ensure that the symlinks (e.g.
./bup) are not converted to text files. -
There is no support for ACLs. If/when some enterprising person fixes this, adjust dev/compare-trees.
-
In test/ext/test-misc, two tests have been disabled. These tests check to see that repeated saves produce identical trees and that an intervening index doesn't change the SHA1. Apparently Cygwin has some unusual behaviors with respect to access times (that probably warrant further investigation). Possibly related: http://cygwin.com/ml/cygwin/2007-06/msg00436.html
- There is no support for ACLs. If/when some enterprising person fixes this, adjust dev/compare-trees.
bup stores its data in a git-formatted repository. Unfortunately, git itself doesn't actually behave very well for bup's use case (huge numbers of files, files with huge sizes, retaining file permissions/ownership are important), so we mostly don't use git's code except for a few helper programs. For example, bup has its own git packfile writer written in python.
Basically, 'bup split' reads the data on stdin (or from files specified on the command line), breaks it into chunks using a rolling checksum (similar to rsync), and saves those chunks into a new git packfile. There is at least one git packfile per backup.
When deciding whether to write a particular chunk into the new packfile, bup first checks all the other packfiles that exist to see if they already have that chunk. If they do, the chunk is skipped.
git packs come in two parts: the pack itself (.pack) and the index (.idx). The index is pretty small, and contains a list of all the objects in the pack. Thus, when generating a remote backup, we don't have to have a copy of the packfiles from the remote server: the local end just downloads a copy of the server's index files, and compares objects against those when generating the new pack, which it sends directly to the server.
The "-n" option to 'bup split' and 'bup save' is the name of the backup you want to create, but it's actually implemented as a git branch. So you can do cute things like checkout a particular branch using git, and receive a bunch of chunk files corresponding to the file you split.
If you use '-b' or '-t' or '-c' instead of '-n', bup split will output a list of blobs, a tree containing that list of blobs, or a commit containing that tree, respectively, to stdout. You can use this to construct your own scripts that do something with those values.
'bup index' walks through your filesystem and updates a file (whose name is, by default, ~/.bup/bupindex) to contain the name, attributes, and an optional git SHA1 (blob id) of each file and directory.
'bup save' basically just runs the equivalent of 'bup split' a whole bunch of times, once per file in the index, and assembles a git tree that contains all the resulting objects. Among other things, that makes 'git diff' much more useful (compared to splitting a tarball, which is essentially a big binary blob). However, since bup splits large files into smaller chunks, the resulting tree structure doesn't exactly correspond to what git itself would have stored. Also, the tree format used by 'bup save' will probably change in the future to support storing file ownership, more complex file permissions, and so on.
If a file has previously been written by 'bup save', then its git blob/tree id is stored in the index. This lets 'bup save' avoid reading that file to produce future incremental backups, which means it can go very fast unless a lot of files have changed.
Help with any of these problems, or others, is very welcome. Join the mailing list (see below) if you'd like to help.
-
'bup save' and 'bup restore' have immature metadata support.
On the plus side, they actually do have support now, but it's new, and not remotely as well tested as tar/rsync/whatever's. However, you have to start somewhere, and as of 0.25, we think it's ready for more general use. Please let us know if you have any trouble.
Also, if any strip or graft-style options are specified to 'bup save', then no metadata will be written for the root directory. That's obviously less than ideal.
-
bup is overly optimistic about mmap. Right now bup just assumes that it can mmap as large a block as it likes, and that mmap will never fail. Yeah, right... If nothing else, this has failed on 32-bit architectures (and 31-bit is even worse -- looking at you, s390).
To fix this, we might just implement a FakeMmap[1] class that uses normal file IO and handles all of the mmap methods[2] that bup actually calls. Then we'd swap in one of those whenever mmap fails.
This would also require implementing some of the methods needed to support "[]" array access, probably at a minimum getitem, setitem, and setslice [3].
[1] http://comments.gmane.org/gmane.comp.sysutils.backup.bup/613 [2] http://docs.python.org/3/library/mmap.html [3] http://docs.python.org/3/reference/datamodel.html#emulating-container-types
-
'bup index' is slower than it should be.
It's still rather fast: it can iterate through all the filenames on my 600,000 file filesystem in a few seconds. But it still needs to rewrite the entire index file just to add a single filename, which is pretty nasty; it should just leave the new files in a second "extra index" file or something.
-
bup could use inotify for really efficient incremental backups.
You could even have your system doing "continuous" backups: whenever a file changes, we immediately send an image of it to the server. We could give the continuous-backup process a really low CPU and I/O priority so you wouldn't even know it was running.
-
bup has never been tested on anything but Linux, FreeBSD, NetBSD, OS X, and Windows+Cygwin.
There's nothing that makes it inherently non-portable, though, so that's mostly a matter of someone putting in some effort. (For a "native" Windows port, the most annoying thing is the absence of ssh in a default Windows installation.)
-
bup needs better documentation.
According to an article about bup in Linux Weekly News (https://lwn.net/Articles/380983/), "it's a bit short on examples and a user guide would be nice." Documentation is the sort of thing that will never be great unless someone from outside contributes it (since the developers can never remember which parts are hard to understand).
-
bup is "relatively speedy" and has "pretty good" compression.
...according to the same LWN article. Clearly neither of those is good enough. We should have awe-inspiring speed and crazy-good compression. Must work on that. Writing more parts in C might help with the speed.
-
bup has no GUI.
Actually, that's not stupid, but you might consider it a limitation. See the "Related Projects" list for some possible options.
bup has an extensive set of man pages. Try using 'bup help' to get started, or use 'bup help SUBCOMMAND' for any bup subcommand (like split, join, index, save, etc.) to get details on that command.
For further technical details, please see ./DESIGN.
bup is a work in progress and there are many ways it can still be improved. If you'd like to contribute patches, ideas, or bug reports, please join the bup mailing list:
You can find the mailing list archives here:
http://groups.google.com/group/bup-list
and you can subscribe by sending a message to:
bup-list+subscribe@googlegroups.com
You can also reach us via the #bup IRC channel at ircs://irc.libera.chat:6697/bup on the libera.chat network or via this web interface.
Please see ./HACKING for additional information, i.e. how to submit patches (hint - no pull requests), how we handle branches, etc.
Have fun,
Avery