]> Dogcows Code - chaz/tar/blob - doc/snapshot.texi
Update
[chaz/tar] / doc / snapshot.texi
1 @c This is part of the paxutils manual.
2 @c Copyright (C) 2005, 2007 Free Software Foundation, Inc.
3 @c Written by Sergey Poznyakoff
4 @c This file is distributed under GFDL 1.1 or any later version
5 @c published by the Free Software Foundation.
6
7 A @dfn{snapshot file} (or @dfn{directory file}) is created during
8 incremental backups (@pxref{Incremental Dumps}). It
9 contains the status of the file system at the time of the dump and is
10 used to determine which files were modified since the last backup.
11
12 @GNUTAR{} version @value{VERSION} supports three snapshot file
13 formats. The first format, called @dfn{format 0}, is the one used by
14 @GNUTAR{} versions up to 1.15.1. The second format, called @dfn{format
15 1} is an extended version of this format, that contains more metadata
16 and allows for further extensions. It was used by version
17 1.15.1. Starting from version 1.16 and up to @value{VERSION}, the
18 @dfn{format 2} is used.
19
20 @GNUTAR{} is able to read all three formats, but will create
21 snapshots only in format 2.
22
23 This appendix describes all three formats in detail.
24
25 @enumerate 0
26 @cindex format 0, snapshot file
27 @cindex snapshot file, format 0
28 @item
29 @samp{Format 0} snapshot file begins with a line containing a
30 decimal number that represents a @acronym{UNIX} timestamp of the
31 beginning of the last archivation. This line is followed by directory
32 metadata descriptions, one per line. Each description has the
33 following format:
34
35 @smallexample
36 @var{nfs}@var{dev} @var{inode} @var{name}
37 @end smallexample
38
39 @noindent
40 where:
41
42 @table @var
43 @item nfs
44 A single plus character (@samp{+}), if this directory is located on
45 an @acronym{NFS}-mounted partition, or a single space otherwise;
46
47 @item dev
48 Device number of the directory;
49
50 @item inode
51 I-node number of the directory;
52
53 @item name
54 Name of the directory. Any special characters (white-space,
55 backslashes, etc.) are quoted.
56 @end table
57
58 @cindex format 1, snapshot file
59 @cindex snapshot file, format 1
60 @item
61 @samp{Format 1} snapshot file begins with a line specifying the
62 format of the file. This line has the following structure:
63
64 @smallexample
65 @samp{GNU tar-}@var{tar-version}@samp{-}@var{incr-format-version}
66 @end smallexample
67
68 @noindent
69 where @var{tar-version} is the version number of @GNUTAR{}
70 implementation that created this snapshot, and
71 @var{incr-format-version} is the version number of the snapshot format
72 (in this case @samp{1}).
73
74 Next line contains two decimal numbers, representing the
75 time of the last backup. First number is the number of seconds, the
76 second one is the number of nanoseconds, since the beginning of the
77 epoch.
78
79 Lines that follow contain directory metadata, one line per
80 directory. Each line is formatted as follows:
81
82 @smallexample
83 [@var{nfs}]@var{mtime-sec} @var{mtime-nsec} @var{dev} @var{inode} @var{name}
84 @end smallexample
85
86 @noindent
87 where @var{mtime-sec} and @var{mtime-nsec} represent last
88 modification time of this directory with nanosecond precision;
89 @var{nfs}, @var{dev}, @var{inode} and @var{name} have the same meaning
90 as with @samp{format 0}.
91
92 @cindex format 2, snapshot file
93 @cindex snapshot file, format 2
94 @item
95 @FIXME{}
96 A snapshot file begins with a format identifier, as described for
97 version 1, e.g.:
98
99 @smallexample
100 GNU tar-@value{VERSION}-2
101 @end smallexample
102
103 This line is followed by newline. Rest of file consists of
104 records, separated by null (@acronym{ASCII} 0)
105 characters. Thus, in contrast to the previous formats, format 2
106 snapshot is a binary file.
107
108 First two records are decimal numbers, representing the
109 time of the last backup. First number is the number of seconds, the
110 second one is the number of nanoseconds, since the beginning of the
111 epoch. These are followed by arbitrary number of directory records.
112
113 Each @dfn{directory record} contains a set of metadata describing a
114 particular directory. Parts of a directory record are delimited with
115 @acronym{ASCII} 0 characters. The following table describes each
116 part. The @dfn{Number} type in this table stands for a decimal number
117 in @acronym{ASCII} notation.
118
119 @multitable @columnfractions 0.2 0.2 0.6
120 @headitem Field @tab Type @tab Description
121 @item nfs @tab Character @tab @samp{1} if the directory is located on
122 an @acronym{NFS}-mounted partition, or @samp{0} otherwise;
123 @item mtime-sec @tab Number @tab Modification time, seconds;
124 @item mtime-nano @tab Number @tab Modification time, nanoseconds;
125 @item dev-no @tab Number @tab Device number;
126 @item i-no @tab Number @tab I-node number;
127 @item name @tab String @tab Directory name; In contrast to the
128 previous versions it is not quoted.
129 @item contents @tab Dumpdir @tab Contents of the directory;
130 @xref{Dumpdir}, for a description of its format.
131 @item
132 @end multitable
133
134 Dumpdirs stored in snapshot files contain only records of types
135 @samp{Y}, @samp{N} and @samp{D}.
136
137 @end enumerate
138
139 @c End of snapshot.texi
140
141
This page took 0.041497 seconds and 4 git commands to generate.