1BTRFS-CHECK(8) BTRFS BTRFS-CHECK(8)
2
6 btrfs-check - check or repair a btrfs filesystem
7
9 btrfs check [options] <device>
10
12 The filesystem checker is used to verify structural integrity of a
13 filesystem and attempt to repair it if requested. It is recommended to
14 unmount the filesystem prior to running the check, but it is possible
15 to start checking a mounted filesystem (see --force).
16 By default, btrfs check will not modify the device but you can reaffirm
18 that by the option --readonly.
19 btrfsck is an alias of btrfs check command and is now deprecated.
21 WARNING:
23 Do not use --repair unless you are advised to do so by a developer
24 or an experienced user, and then only after having accepted that no
25 fsck successfully repair all types of filesystem corruption. E.g.
26 some other software or hardware bugs can fatally damage a volume.
27 The structural integrity check verifies if internal filesystem objects
29 or data structures satisfy the constraints, point to the right objects
30 or are correctly connected together.
31 There are several cross checks that can detect wrong reference counts
33 of shared extents, backreferences, missing extents of inodes, directory
34 and inode connectivity etc.
35 The amount of memory required can be high, depending on the size of the
37 filesystem, similarly the run time. Check the modes that can also af‐
38 fect that.
39
41 -b|--backup
42 use the first valid set of backup roots stored in the superblock
43 This can be combined with --super if some of the superblocks are
45 damaged.
46 --check-data-csum
48 verify checksums of data blocks
49 This expects that the filesystem is otherwise OK, and is basi‐
51 cally an offline scrub that does not repair data from spare
52 copies.
53 --chunk-root <bytenr>
55 use the given offset bytenr for the chunk tree root
56 -E|--subvol-extents <subvolid>
58 show extent state for the given subvolume
59 -p|--progress
61 indicate progress at various checking phases
62 -Q|--qgroup-report
64 verify qgroup accounting and compare against filesystem account‐
65 ing
66 -r|--tree-root <bytenr>
68 use the given offset 'bytenr' for the tree root
69 --readonly
71 (default) run in read-only mode, this option exists to calm po‐
72 tential panic when users are going to run the checker
73 -s|--super <N>
75 use Nth superblock copy, valid values are 0, 1 or 2 if the re‐
76 spective superblock offset is within the device size
77 This can be used to use a different starting point if some of
79 the primary superblock is damaged.
80 --clear-space-cache v1|v2
82 completely wipe all free space cache of given type
83 For free space cache v1, the clear_cache kernel mount option
85 only rebuilds the free space cache for block groups that are
86 modified while the filesystem is mounted with that option. Thus,
87 using this option with v1 makes it possible to actually clear
88 the entire free space cache.
89 For free space cache v2, the clear_cache kernel mount option de‐
91 stroys the entire free space cache. This option, with v2 pro‐
92 vides an alternative method of clearing the free space cache
93 that doesn't require mounting the filesystem.
94 --clear-ino-cache
96 remove leftover items pertaining to the deprecated inode map
97 feature
98
100 --repair
101 enable the repair mode and attempt to fix problems where possi‐
102 ble
103 NOTE:
105 There's a warning and 10 second delay when this option is run
106 without --force to give users a chance to think twice before
107 running repair, the warnings in documentation have shown to
108 be insufficient
109 --init-csum-tree
111 create a new checksum tree and recalculate checksums in all
112 files
113 WARNING:
115 Do not blindly use this option to fix checksum mismatch prob‐
116 lems.
117 --init-extent-tree
119 build the extent tree from scratch
120 WARNING:
122 Do not use unless you know what you're doing.
123 --mode <MODE>
125 select mode of operation regarding memory and IO
126 The MODE can be one of:
128 original
130 The metadata are read into memory and verified, thus the
131 requirements are high on large filesystems and can even
132 lead to out-of-memory conditions. The possible work‐
133 around is to export the block device over network to a
134 machine with enough memory.
135 lowmem This mode is supposed to address the high memory consump‐
137 tion at the cost of increased IO when it needs to re-read
138 blocks. This may increase run time.
139 NOTE:
141 lowmem mode does not work with --repair yet, and is still
142 considered experimental.
143 --force
145 allow work on a mounted filesystem. Note that this should work
146 fine on a quiescent or read-only mounted filesystem but may
147 crash if the device is changed externally, e.g. by the kernel
148 module. Repair without mount checks is not supported right now.
149 This option also skips the delay and warning in the repair mode
151 (see --repair).
152
154 btrfs check returns a zero exit status if it succeeds. Non zero is re‐
155 turned in case of failure.
156
158 btrfs is part of btrfs-progs. Please refer to the documentation at
159 https://btrfs.readthedocs.io or wiki http://btrfs.wiki.kernel.org for
160 further information.
161
163 mkfs.btrfs(8), btrfs-scrub(8), btrfs-rescue(8)
1646.1.3 Jan 25, 2023 BTRFS-CHECK(8)