Commit | Line | Data |
---|---|---|
e484585e PBG |
1 | Device-mapper snapshot support |
2 | ============================== | |
3 | ||
4 | Device-mapper allows you, without massive data copying: | |
5 | ||
6 | *) To create snapshots of any block device i.e. mountable, saved states of | |
7 | the block device which are also writable without interfering with the | |
8 | original content; | |
9 | *) To create device "forks", i.e. multiple different versions of the | |
10 | same data stream. | |
d698aa45 MP |
11 | *) To merge a snapshot of a block device back into the snapshot's origin |
12 | device. | |
e484585e | 13 | |
d698aa45 MP |
14 | In the first two cases, dm copies only the chunks of data that get |
15 | changed and uses a separate copy-on-write (COW) block device for | |
16 | storage. | |
e484585e | 17 | |
d698aa45 MP |
18 | For snapshot merge the contents of the COW storage are merged back into |
19 | the origin device. | |
e484585e PBG |
20 | |
21 | ||
d698aa45 MP |
22 | There are three dm targets available: |
23 | snapshot, snapshot-origin, and snapshot-merge. | |
e484585e PBG |
24 | |
25 | *) snapshot-origin <origin> | |
26 | ||
27 | which will normally have one or more snapshots based on it. | |
e484585e PBG |
28 | Reads will be mapped directly to the backing device. For each write, the |
29 | original data will be saved in the <COW device> of each snapshot to keep | |
30 | its visible content unchanged, at least until the <COW device> fills up. | |
31 | ||
32 | ||
33 | *) snapshot <origin> <COW device> <persistent?> <chunksize> | |
34 | ||
411f1140 | 35 | A snapshot of the <origin> block device is created. Changed chunks of |
e484585e PBG |
36 | <chunksize> sectors will be stored on the <COW device>. Writes will |
37 | only go to the <COW device>. Reads will come from the <COW device> or | |
38 | from <origin> for unchanged data. <COW device> will often be | |
39 | smaller than the origin and if it fills up the snapshot will become | |
40 | useless and be disabled, returning errors. So it is important to monitor | |
41 | the amount of free space and expand the <COW device> before it fills up. | |
42 | ||
43 | <persistent?> is P (Persistent) or N (Not persistent - will not survive | |
b0d3cc01 MS |
44 | after reboot). O (Overflow) can be added as a persistent store option |
45 | to allow userspace to advertise its support for seeing "Overflow" in the | |
46 | snapshot status. So supported store types are "P", "PO" and "N". | |
47 | ||
48 | The difference between persistent and transient is with transient | |
49 | snapshots less metadata must be saved on disk - they can be kept in | |
50 | memory by the kernel. | |
e484585e PBG |
51 | |
52 | ||
d698aa45 MP |
53 | * snapshot-merge <origin> <COW device> <persistent> <chunksize> |
54 | ||
55 | takes the same table arguments as the snapshot target except it only | |
56 | works with persistent snapshots. This target assumes the role of the | |
57 | "snapshot-origin" target and must not be loaded if the "snapshot-origin" | |
58 | is still present for <origin>. | |
59 | ||
60 | Creates a merging snapshot that takes control of the changed chunks | |
61 | stored in the <COW device> of an existing snapshot, through a handover | |
62 | procedure, and merges these chunks back into the <origin>. Once merging | |
63 | has started (in the background) the <origin> may be opened and the merge | |
64 | will continue while I/O is flowing to it. Changes to the <origin> are | |
65 | deferred until the merging snapshot's corresponding chunk(s) have been | |
66 | merged. Once merging has started the snapshot device, associated with | |
67 | the "snapshot" target, will return -EIO when accessed. | |
68 | ||
69 | ||
70 | How snapshot is used by LVM2 | |
71 | ============================ | |
e484585e PBG |
72 | When you create the first LVM2 snapshot of a volume, four dm devices are used: |
73 | ||
74 | 1) a device containing the original mapping table of the source volume; | |
75 | 2) a device used as the <COW device>; | |
76 | 3) a "snapshot" device, combining #1 and #2, which is the visible snapshot | |
77 | volume; | |
78 | 4) the "original" volume (which uses the device number used by the original | |
79 | source volume), whose table is replaced by a "snapshot-origin" mapping | |
80 | from device #1. | |
81 | ||
82 | A fixed naming scheme is used, so with the following commands: | |
83 | ||
84 | lvcreate -L 1G -n base volumeGroup | |
85 | lvcreate -L 100M --snapshot -n snap volumeGroup/base | |
86 | ||
87 | we'll have this situation (with volumes in above order): | |
88 | ||
89 | # dmsetup table|grep volumeGroup | |
90 | ||
91 | volumeGroup-base-real: 0 2097152 linear 8:19 384 | |
92 | volumeGroup-snap-cow: 0 204800 linear 8:19 2097536 | |
93 | volumeGroup-snap: 0 2097152 snapshot 254:11 254:12 P 16 | |
94 | volumeGroup-base: 0 2097152 snapshot-origin 254:11 | |
95 | ||
96 | # ls -lL /dev/mapper/volumeGroup-* | |
97 | brw------- 1 root root 254, 11 29 ago 18:15 /dev/mapper/volumeGroup-base-real | |
98 | brw------- 1 root root 254, 12 29 ago 18:15 /dev/mapper/volumeGroup-snap-cow | |
99 | brw------- 1 root root 254, 13 29 ago 18:15 /dev/mapper/volumeGroup-snap | |
100 | brw------- 1 root root 254, 10 29 ago 18:14 /dev/mapper/volumeGroup-base | |
101 | ||
d698aa45 MP |
102 | |
103 | How snapshot-merge is used by LVM2 | |
104 | ================================== | |
105 | A merging snapshot assumes the role of the "snapshot-origin" while | |
106 | merging. As such the "snapshot-origin" is replaced with | |
107 | "snapshot-merge". The "-real" device is not changed and the "-cow" | |
108 | device is renamed to <origin name>-cow to aid LVM2's cleanup of the | |
109 | merging snapshot after it completes. The "snapshot" that hands over its | |
110 | COW device to the "snapshot-merge" is deactivated (unless using lvchange | |
111 | --refresh); but if it is left active it will simply return I/O errors. | |
112 | ||
113 | A snapshot will merge into its origin with the following command: | |
114 | ||
115 | lvconvert --merge volumeGroup/snap | |
116 | ||
117 | we'll now have this situation: | |
118 | ||
119 | # dmsetup table|grep volumeGroup | |
120 | ||
121 | volumeGroup-base-real: 0 2097152 linear 8:19 384 | |
122 | volumeGroup-base-cow: 0 204800 linear 8:19 2097536 | |
123 | volumeGroup-base: 0 2097152 snapshot-merge 254:11 254:12 P 16 | |
124 | ||
125 | # ls -lL /dev/mapper/volumeGroup-* | |
126 | brw------- 1 root root 254, 11 29 ago 18:15 /dev/mapper/volumeGroup-base-real | |
127 | brw------- 1 root root 254, 12 29 ago 18:16 /dev/mapper/volumeGroup-base-cow | |
128 | brw------- 1 root root 254, 10 29 ago 18:16 /dev/mapper/volumeGroup-base | |
c53a381e MS |
129 | |
130 | ||
131 | How to determine when a merging is complete | |
132 | =========================================== | |
133 | The snapshot-merge and snapshot status lines end with: | |
134 | <sectors_allocated>/<total_sectors> <metadata_sectors> | |
135 | ||
136 | Both <sectors_allocated> and <total_sectors> include both data and metadata. | |
137 | During merging, the number of sectors allocated gets smaller and | |
138 | smaller. Merging has finished when the number of sectors holding data | |
139 | is zero, in other words <sectors_allocated> == <metadata_sectors>. | |
140 | ||
141 | Here is a practical example (using a hybrid of lvm and dmsetup commands): | |
142 | ||
143 | # lvs | |
144 | LV VG Attr LSize Origin Snap% Move Log Copy% Convert | |
145 | base volumeGroup owi-a- 4.00g | |
146 | snap volumeGroup swi-a- 1.00g base 18.97 | |
147 | ||
148 | # dmsetup status volumeGroup-snap | |
149 | 0 8388608 snapshot 397896/2097152 1560 | |
150 | ^^^^ metadata sectors | |
151 | ||
152 | # lvconvert --merge -b volumeGroup/snap | |
153 | Merging of volume snap started. | |
154 | ||
155 | # lvs volumeGroup/snap | |
156 | LV VG Attr LSize Origin Snap% Move Log Copy% Convert | |
157 | base volumeGroup Owi-a- 4.00g 17.23 | |
158 | ||
159 | # dmsetup status volumeGroup-base | |
160 | 0 8388608 snapshot-merge 281688/2097152 1104 | |
161 | ||
162 | # dmsetup status volumeGroup-base | |
163 | 0 8388608 snapshot-merge 180480/2097152 712 | |
164 | ||
165 | # dmsetup status volumeGroup-base | |
166 | 0 8388608 snapshot-merge 16/2097152 16 | |
167 | ||
168 | Merging has finished. | |
169 | ||
170 | # lvs | |
171 | LV VG Attr LSize Origin Snap% Move Log Copy% Convert | |
172 | base volumeGroup owi-a- 4.00g |