1 % bup-split(1) Bup %BUP_VERSION%
2 % Avery Pennarun <apenwarr@gmail.com>
7 bup-split - save individual files to bup backup sets
11 bup split [-r *host*:*path*] <-b|-t|-c|-n *name*> [-v] [-q]
12 [--bench] [--max-pack-size=*bytes*]
13 [--max-pack-objects=*n*] [--fanout=*count] [filenames...]
17 `bup split` concatenates the contents of the given files
18 (or if no filenames are given, reads from stdin), splits
19 the content into chunks of around 8k using a rolling
20 checksum algorithm, and saves the chunks into a bup
21 repository. Chunks which have previously been stored are
22 not stored again (ie. they are "deduplicated").
24 Because of the way the rolling checksum works, chunks
25 tend to be very stable across changes to a given file,
26 including adding, deleting, and changing bytes.
28 For example, if you use `bup split` to back up an XML dump
29 of a database, and the XML file changes slightly from one
30 run to the next, nearly all the data will still be
31 deduplicated and the size of each backup after the first
32 will typically be quite small.
34 Another technique is to pipe the output of the `tar`(1) or
35 `cpio`(1) programs to `bup split`. When individual files
36 in the tarball change slightly or are added or removed, bup
37 still processes the remainder of the tarball efficiently.
38 (Note that `bup save` is usually a more efficient way to
39 accomplish this, however.)
41 To get the data back, use `bup-join`(1).
45 -r, --remote=*host*:*path*
46 : save the backup set to the given remote server. If
47 *path* is omitted, uses the default path on the remote
48 server (you still need to include the ':')
51 : output a series of git blob ids that correspond to the
52 chunks in the dataset.
55 : output the git tree id of the resulting dataset.
58 : output the git commit id of the resulting dataset.
61 : after creating the dataset, create a git branch
62 named *name* so that it can be accessed using
63 that name. If *name* already exists, the new dataset
64 will be considered a descendant of the old *name*.
65 (Thus, you can continually create new datasets with
66 the same name, and later view the history of that
67 dataset to see how it has changed over time.)
70 : disable progress messages.
73 : increase verbosity (can be used more than once).
76 : read the data and split it into blocks based on the "bupsplit"
77 rolling checksum algorithm, but don't do anything with
78 the blocks. This is mostly useful for benchmarking.
81 : like --noop, but also write the data to stdout. This
82 can be useful for benchmarking the speed of read+bupsplit+write
83 for large amounts of data.
86 : print benchmark timings to stderr.
88 --max-pack-size=*bytes*
89 : never create git packfiles larger than the given number
90 of bytes. Default is 1 billion bytes. Usually there
91 is no reason to change this.
93 --max-pack-objects=*numobjs*
94 : never create git packfiles with more than the given
95 number of objects. Default is 200 thousand objects.
96 Usually there is no reason to change this.
99 : when splitting very large files, never put more than
100 this number of git blobs in a single git tree. Instead,
101 generate a new tree and link to that. Default is
102 4096 objects per tree.
104 --bwlimit=*bytes/sec*
105 : don't transmit more than *bytes/sec* bytes per second
106 to the server. This is good for making your backups
107 not suck up all your network bandwidth. Use a suffix
108 like k, M, or G to specify multiples of 1024,
109 1024*1024, 1024*1024*1024 respectively.
114 $ tar -cf - /etc | bup split -r myserver: -n mybackup-tar
115 tar: Removing leading /' from member names
116 Indexing objects: 100% (196/196), done.
118 $ bup join -r myserver: mybackup-tar | tar -tf - | wc -l
124 `bup-join`(1), `bup-index`(1), `bup-save`(1), `bup-on`(1)
128 Part of the `bup`(1) suite.