6 btrfs-mount - mount options for the btrfs filesystem
10 This document describes mount options specific to the btrfs filesystem.
11 Other generic mount options are available,and are described in the
16 *alloc_start='bytes'*::
17 Debugging option to force all block allocations above a certain
18 byte threshold on each block device. The value is specified in
19 bytes, optionally with a K, M, or G suffix, case insensitive.
23 Disable/enable auto defragmentation.
24 Auto defragmentation detects small random writes into files and queue
25 them up for the defrag process. Works best for small files;
26 Not well suited for large database workloads.
28 *check_int*|*check_int_data*|*check_int_print_mask='value'*::
29 These debugging options control the behavior of the integrity checking
30 module (the BTRFS_FS_CHECK_INTEGRITY config option required). +
32 `check_int` enables the integrity checker module, which examines all
33 block write requests to ensure on-disk consistency, at a large
34 memory and CPU cost. +
36 `check_int_data` includes extent data in the integrity checks, and
37 implies the check_int option. +
39 `check_int_print_mask` takes a bitmask of BTRFSIC_PRINT_MASK_* values
40 as defined in 'fs/btrfs/check-integrity.c', to control the integrity
41 checker module behavior. +
43 See comments at the top of 'fs/btrfs/check-integrity.c'
47 Set the interval of periodic commit, 30 seconds by default. Higher
48 values defer data being synced to permanent storage with obvious
49 consequences when the system crashes. The upper bound is not forced,
50 but a warning is printed if it's more than 300 seconds (5 minutes).
52 *compress*|*compress='type'*|*compress-force*|*compress-force='type'*::
53 Control BTRFS file data compression. Type may be specified as "zlib"
54 "lzo" or "no" (for no compression, used for remounting). If no type
55 is specified, zlib is used. If compress-force is specified,
56 all files will be compressed, whether or not they compress well.
57 If compression is enabled, nodatacow and nodatasum are disabled.
60 Allow mounts to continue with missing devices. A read-write mount may
61 fail with too many devices missing, for example if a stripe member
62 is completely missing.
64 *device='devicepath'*::
65 Specify a device during mount so that ioctls on the control device
66 can be avoided. Especially useful when trying to mount a multi-device
67 setup as root. May be specified multiple times for multiple devices.
70 Disable/enable discard mount option.
71 Discard issues frequent commands to let the block device reclaim space
72 freed by the filesystem.
73 This is useful for SSD devices, thinly provisioned
74 LUNs and virtual machine images, but may have a significant
75 performance impact. (The fstrim command is also available to
76 initiate batch trims from userspace).
79 Disable/enable debugging option to be more verbose in some ENOSPC conditions.
81 *fatal_errors='action'*::
82 Action to take when encountering a fatal error. +
83 "bug" - BUG() on a fatal error. This is the default. +
84 "panic" - panic() on a fatal error.
87 The `flushoncommit` mount option forces any data dirtied by a write in a
88 prior transaction to commit as part of the current commit. This makes
89 the committed state a fully consistent view of the file system from the
90 application's perspective (i.e., it includes all completed file system
91 operations). This was previously the behavior only when a snapshot is
95 Enable free inode number caching. Defaults to off due to an overflow
96 problem when the free space crcs don't fit inside a single page.
98 *max_inline='bytes'*::
99 Specify the maximum amount of space, in bytes, that can be inlined in
100 a metadata B-tree leaf. The value is specified in bytes, optionally
101 with a K, M, or G suffix, case insensitive. In practice, this value
102 is limited by the root sector size, with some space unavailable due
103 to leaf headers. For a 4k sectorsize, max inline data is ~3900 bytes.
105 *metadata_ratio='value'*::
106 Specify that 1 metadata chunk should be allocated after every
107 'value' data chunks. Off by default.
110 Enable/disable support for Posix Access Control Lists (ACLs). See the
111 `acl`(5) manual page for more information about ACLs.
114 ensure that certain IOs make it through the device cache and are on
115 persistent storage. If disabled on a device with a volatile
116 (non-battery-backed) write-back cache, nobarrier option will lead to
117 filesystem corruption on a system crash or power loss.
120 Enable/disable data copy-on-write for newly created files.
121 Nodatacow implies nodatasum, and disables all compression.
124 Enable/disable data checksumming for newly created files.
125 Datasum implies datacow.
128 Enable/disable the tree logging used for fsync and O_SYNC writes.
131 Enable autorecovery attempts if a bad tree root is found at mount time.
132 Currently this scans a list of several previous tree roots and tries to
133 use the first readable.
136 Force check and rebuild procedure of the UUID tree. This should not
140 Skip automatic resume of interrupted balance operation after mount.
141 May be resumed with "btrfs balance resume."
144 Disable freespace cache loading without clearing the cache.
147 Force clearing and rebuilding of the disk space cache if something
150 *ssd*|*nossd*|*ssd_spread*::
151 Options to control ssd allocation schemes. By default, BTRFS will
152 enable or disable ssd allocation heuristics depending on whether a
153 rotational or nonrotational disk is in use. The ssd and nossd options
154 can override this autodetection. +
155 The ssd_spread mount option attempts to allocate into big chunks
156 of unused space, and may perform better on low-end ssds. ssd_spread
157 implies ssd, enabling all other ssd heuristics as well.
160 Mount subvolume at 'path' rather than the root subvolume. The
161 'path' is relative to the top level subvolume.
164 Mount subvolume specified by an ID number rather than the root subvolume.
165 This allows mounting of subvolumes which are not in the root of the mounted
167 You can use "btrfs subvolume list" to see subvolume ID numbers.
169 *subvolrootid='objectid' (deprecated)*::
170 Mount subvolume specified by 'objectid' rather than the root subvolume.
171 This allows mounting of subvolumes which are not in the root of the mounted
173 You can use "btrfs subvolume show" to see the object ID for a subvolume.
175 *thread_pool='number'*::
176 The number of worker threads to allocate. The default number is equal
177 to the number of CPUs + 2, or 8, whichever is smaller.
179 *user_subvol_rm_allowed*::
180 Allow subvolumes to be deleted by a non-root user. Use with caution.