1BTRFS-QGROUP(8) BTRFS BTRFS-QGROUP(8)
2
6 btrfs-qgroup - control the quota group of a btrfs filesystem
7
9 btrfs qgroup <subcommand> <args>
10
12 btrfs qgroup is used to control quota group (qgroup) of a btrfs
13 filesystem.
14 NOTE:
16 To use qgroup you need to enable quota first using btrfs quota en‐
17 able command.
18 WARNING:
20 Qgroup is not stable yet and will impact performance in current
21 mainline kernel (v4.14).
22
24 Quota groups or qgroup in btrfs make a tree hierarchy, the leaf qgroups
25 are attached to subvolumes. The size limits are set per qgroup and ap‐
26 ply when any limit is reached in tree that contains a given subvolume.
27 The limits are separated between shared and exclusive and reflect the
29 extent ownership. For example a fresh snapshot shares almost all the
30 blocks with the original subvolume, new writes to either subvolume will
31 raise towards the exclusive limit.
32 The qgroup identifiers conform to level/id where level 0 is reserved to
34 the qgroups associated with subvolumes. Such qgroups are created auto‐
35 matically.
36 The qgroup hierarchy is built by commands create and assign.
38 NOTE:
40 If the qgroup of a subvolume is destroyed, quota about the subvolume
41 will not be functional until qgroup 0/<subvolume id> is created
42 again.
43
45 assign [options] <src> <dst> <path>
46 Assign qgroup src as the child qgroup of dst in the btrfs
47 filesystem identified by path.
48 Options
50 --rescan
52 (default since: 4.19) Automatically schedule quota rescan
53 if the new qgroup assignment would lead to quota incon‐
54 sistency. See QUOTA RESCAN for more information.
55 --no-rescan
57 Explicitly ask not to do a rescan, even if the assignment
58 will make the quotas inconsistent. This may be useful for
59 repeated calls where the rescan would add unnecessary
60 overhead.
61 create <qgroupid> <path>
63 Create a subvolume quota group.
64 For the 0/<subvolume id> qgroup, a qgroup can be created even
66 before the subvolume is created.
67 destroy <qgroupid> <path>
69 Destroy a qgroup.
70 If a qgroup is not isolated, meaning it is a parent or child
72 qgroup, then it can only be destroyed after the relationship is
73 removed.
74 clear-stale <path>
76 Clear all stale qgroups whose subvolume does not exist anymore,
77 this is the level 0 qgroup like 0/subvolid. Higher level qgroups
78 are not deleted even if they don't have any child qgroups.
79 limit [options] <size>|none [<qgroupid>] <path>
81 Limit the size of a qgroup to size or no limit in the btrfs
82 filesystem identified by path.
83 If qgroupid is not given, qgroup of the subvolume identified by
85 path is used if possible.
86 Options
88 -c limit amount of data after compression. This is the de‐
90 fault, it is currently not possible to turn off this op‐
91 tion.
92 -e limit space exclusively assigned to this qgroup.
94 remove <src> <dst> <path>
96 Remove the relationship between child qgroup src and parent
97 qgroup dst in the btrfs filesystem identified by path.
98 Options
100 --rescan
102 (default since: 4.19) Automatically schedule quota rescan
103 if the removed qgroup relation would lead to quota incon‐
104 sistency. See QUOTA RESCAN for more information.
105 --no-rescan
107 Explicitly ask not to do a rescan, even if the removal
108 will make the quotas inconsistent. This may be useful for
109 repeated calls where the rescan would add unnecessary
110 overhead.
111 show [options] <path>
113 Show all qgroups in the btrfs filesystem identified by <path>.
114 Options
116 -p print parent qgroup id.
118 -c print child qgroup id.
120 -r print limit of referenced size of qgroup.
122 -e print limit of exclusive size of qgroup.
124 -F list all qgroups which impact the given path(include an‐
126 cestral qgroups)
127 -f list all qgroups which impact the given path(exclude an‐
129 cestral qgroups)
130 --raw raw numbers in bytes, without the B suffix.
132 --human-readable
134 print human friendly numbers, base 1024, this is the de‐
135 fault
136 --iec select the 1024 base for the following options, according
138 to the IEC standard.
139 --si select the 1000 base for the following options, according
141 to the SI standard.
142 --kbytes
144 show sizes in KiB, or kB with --si.
145 --mbytes
147 show sizes in MiB, or MB with --si.
148 --gbytes
150 show sizes in GiB, or GB with --si.
151 --tbytes
153 show sizes in TiB, or TB with --si.
154 --sort=[+/-]<attr>[,[+/-]<attr>]...
156 list qgroups in order of <attr>.
157 <attr> can be one or more of
159 qgroupid,rfer,excl,max_rfer,max_excl.
160 Prefix + means ascending order and - means descending or‐
162 der of attr. If no prefix is given, use ascending order
163 by default.
164 If multiple attr values are given, use comma to separate.
166 --sync To retrieve information after updating the state of
168 qgroups, force sync of the filesystem identified by path
169 before getting information.
170
172 The rescan reads all extent sharing metadata and updates the respective
173 qgroups accordingly.
174 The information consists of bytes owned exclusively (excl) or
176 shared/referred to (rfer). There's no explicit information about which
177 extents are shared or owned exclusively. This means when qgroup rela‐
178 tionship changes, extent owners change and qgroup numbers are no longer
179 consistent unless we do a full rescan.
180 However there are cases where we can avoid a full rescan, if a subvol‐
182 ume whose rfer number equals its excl number, which means all bytes are
183 exclusively owned, then assigning/removing this subvolume only needs to
184 add/subtract rfer number from its parent qgroup. This can speed up the
185 rescan.
186
188 Make a parent group that has two quota group children
189 Given the following filesystem mounted at /mnt/my-vault
190 Label: none uuid: 60d2ab3b-941a-4f22-8d1a-315f329797b2
192 Total devices 1 FS bytes used 128.00KiB
193 devid 1 size 5.00GiB used 536.00MiB path /dev/vdb
194 Enable quota and create subvolumes. Check subvolume ids.
196 $ cd /mnt/my-vault
198 $ btrfs quota enable .
199 $ btrfs subvolume create a
200 $ btrfs subvolume create b
201 $ btrfs subvolume list .
202 ID 261 gen 61 top level 5 path a
204 ID 262 gen 62 top level 5 path b
205 Create qgroup and set limit to 10MiB.
207 $ btrfs qgroup create 1/100 .
209 $ btrfs qgroup limit 10M 1/100 .
210 $ btrfs qgroup assign 0/261 1/100 .
211 $ btrfs qgroup assign 0/262 1/100 .
212 And check qgroups.
214 $ btrfs qgroup show .
216 qgroupid rfer excl
218 -------- ---- ----
219 0/5 16.00KiB 16.00KiB
220 0/261 16.00KiB 16.00KiB
221 0/262 16.00KiB 16.00KiB
222 1/100 32.00KiB 32.00KiB
223
225 btrfs qgroup returns a zero exit status if it succeeds. Non zero is re‐
226 turned in case of failure.
227
229 btrfs is part of btrfs-progs. Please refer to the documentation at
230 https://btrfs.readthedocs.io or wiki http://btrfs.wiki.kernel.org for
231 further information.
232
234 btrfs-quota(8), btrfs-subvolume(8), mkfs.btrfs(8),
2356.1.3 Jan 25, 2023 BTRFS-QGROUP(8)