| 1 |
.\" Man page generated from reStructuredText |
| 2 |
.\" by the Docutils 0.22.4 manpage writer. |
| 3 |
. |
| 4 |
. |
| 5 |
.nr rst2man-indent-level 0 |
| 6 |
. |
| 7 |
.de1 rstReportMargin |
| 8 |
\\$1 \\n[an-margin] |
| 9 |
level \\n[rst2man-indent-level] |
| 10 |
level margin: \\n[rst2man-indent\\n[rst2man-indent-level]] |
| 11 |
- |
| 12 |
\\n[rst2man-indent0] |
| 13 |
\\n[rst2man-indent1] |
| 14 |
\\n[rst2man-indent2] |
| 15 |
.. |
| 16 |
.de1 INDENT |
| 17 |
.\" .rstReportMargin pre: |
| 18 |
. RS \\$1 |
| 19 |
. nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin] |
| 20 |
. nr rst2man-indent-level +1 |
| 21 |
.\" .rstReportMargin post: |
| 22 |
.. |
| 23 |
.de UNINDENT |
| 24 |
. RE |
| 25 |
.\" indent \\n[an-margin] |
| 26 |
.\" old: \\n[rst2man-indent\\n[rst2man-indent-level]] |
| 27 |
.nr rst2man-indent-level -1 |
| 28 |
.\" new: \\n[rst2man-indent\\n[rst2man-indent-level]] |
| 29 |
.in \\n[rst2man-indent\\n[rst2man-indent-level]]u |
| 30 |
.. |
| 31 |
.TH "BTRFSTUNE" "8" "May 10, 2026" "7.0" "BTRFS" |
| 32 |
.SH NAME |
| 33 |
btrfstune \- tune various filesystem parameters |
| 34 |
.SH SYNOPSIS |
| 35 |
.sp |
| 36 |
\fBbtrfstune\fP [options] <device> [<device>...] |
| 37 |
.SH DESCRIPTION |
| 38 |
.sp |
| 39 |
\fBbtrfstune\fP can be used to enable, disable, or set various filesystem |
| 40 |
parameters. The filesystem must be unmounted. |
| 41 |
.sp |
| 42 |
The common use case is to enable features that were not enabled at mkfs time. |
| 43 |
Please make sure that you have kernel support for the features. You can find a |
| 44 |
complete list of features and kernel version of their introduction at |
| 45 |
Feature by version \%<> page. Also, the manual page |
| 46 |
mkfs.btrfs(8) \%<> contains more details about the features. |
| 47 |
.sp |
| 48 |
Some of the features could be also enabled on a mounted filesystem by other |
| 49 |
means. Please refer to the \fIFILESYSTEM FEATURES\fP in btrfs(5) \%<>\&. |
| 50 |
.sp |
| 51 |
\fBNote:\fP |
| 52 |
.INDENT 0.0 |
| 53 |
.INDENT 3.5 |
| 54 |
\fBbtrfstune\fP requires the target filesystem to have no running replace |
| 55 |
nor balance. Please finish/cancel the running replace/balance before using |
| 56 |
\fBbtrfstune\fP\&. |
| 57 |
.UNINDENT |
| 58 |
.UNINDENT |
| 59 |
.SH OPTIONS |
| 60 |
.INDENT 0.0 |
| 61 |
.TP |
| 62 |
.B \-\-convert\-to\-block\-group\-tree |
| 63 |
(since kernel 6.1) |
| 64 |
.sp |
| 65 |
Convert portions of extent tree that tracks block groups to a separate |
| 66 |
block group tree. This greatly reduces mount time. Can be also enabled |
| 67 |
at mkfs time. |
| 68 |
.TP |
| 69 |
.B \-\-convert\-from\-block\-group\-tree |
| 70 |
(since kernel 6.1) |
| 71 |
.sp |
| 72 |
Convert block groups tracked in standalone block group tree back to |
| 73 |
extent tree and remove \fIblock\-group\-tree\fP feature bit from the filesystem. |
| 74 |
.TP |
| 75 |
.B \-\-convert\-to\-free\-space\-tree |
| 76 |
(since kernel 4.5) |
| 77 |
.sp |
| 78 |
Convert to free\-space\-tree feature (v2 of space cache). |
| 79 |
.TP |
| 80 |
.B \-\-enable\-simple\-quota |
| 81 |
(since kernel 6.7) |
| 82 |
.sp |
| 83 |
Enable simple quota accounting (\fIsquota\fP mkfs feature). This is an |
| 84 |
alternative to qgroups with a smaller performance impact but no notion |
| 85 |
of shared vs. exclusive usage. |
| 86 |
.TP |
| 87 |
.B \-\-remove\-simple\-quota |
| 88 |
Remove simple quota accounting related structures. |
| 89 |
.TP |
| 90 |
.B \-f |
| 91 |
Allow dangerous changes, e.g. clear the seeding flag or change fsid. |
| 92 |
Make sure that you are aware of the dangers. |
| 93 |
.TP |
| 94 |
.B \-m |
| 95 |
(since kernel: 5.0) |
| 96 |
.sp |
| 97 |
change fsid stored as \fImetadata_uuid\fP to a randomly generated UUID, |
| 98 |
see also \fI\-U\fP |
| 99 |
.UNINDENT |
| 100 |
.INDENT 0.0 |
| 101 |
.TP |
| 102 |
.BI \-M \ <UUID> |
| 103 |
(since kernel: 5.0) |
| 104 |
.sp |
| 105 |
change fsid stored as \fImetadata_uuid\fP to a given UUID, see also \fI\-U\fP |
| 106 |
.sp |
| 107 |
The metadata_uuid is stored only in the superblock and is a backward |
| 108 |
incompatible change. The fsid in metadata blocks remains unchanged and |
| 109 |
is not overwritten, thus the whole operation is significantly faster |
| 110 |
than \fI\-U\fP\&. |
| 111 |
.sp |
| 112 |
The new metadata_uuid can be used for mount by UUID and is also used to |
| 113 |
identify devices of a multi\-device filesystem. |
| 114 |
.TP |
| 115 |
.B \-n |
| 116 |
(since kernel: 3.14) |
| 117 |
.sp |
| 118 |
Enable no\-holes feature (more efficient representation of file holes), |
| 119 |
enabled by mkfs feature \fIno\-holes\fP\&. |
| 120 |
.TP |
| 121 |
.B \-r |
| 122 |
(since kernel: 3.7) |
| 123 |
.sp |
| 124 |
Enable extended inode refs (hardlink limit per file in a directory is |
| 125 |
65536), enabled by mkfs feature \fIextref\fP\&. |
| 126 |
.TP |
| 127 |
.BI \-S \ <0|1> |
| 128 |
Enable seeding on a given device. Value 1 will enable seeding, 0 will |
| 129 |
disable it. A seeding filesystem is forced to be mounted read\-only. A |
| 130 |
new device can be added to the filesystem and will capture all writes |
| 131 |
keeping the seeding device intact. See also section |
| 132 |
SEEDING DEVICE \%<#\:man-btrfs5-seeding-device> |
| 133 |
in btrfs(5) \%<>\&. |
| 134 |
.sp |
| 135 |
\fBWarning:\fP |
| 136 |
.INDENT 7.0 |
| 137 |
.INDENT 3.5 |
| 138 |
Clearing the seeding flag on a device may be dangerous. If a |
| 139 |
previously\-seeding device is changed, all filesystems that used |
| 140 |
that device will become unmountable. Setting the seeding flag |
| 141 |
back will not fix that. |
| 142 |
.sp |
| 143 |
A valid usecase is \fIseeding device as a base image\fP\&. Clear the |
| 144 |
seeding flag, update the filesystem and make it seeding again, |
| 145 |
provided that it\(aqs OK to throw away all filesystems built on |
| 146 |
top of the previous base. |
| 147 |
.UNINDENT |
| 148 |
.UNINDENT |
| 149 |
.TP |
| 150 |
.B \-u |
| 151 |
Change fsid to a randomly generated UUID or continue previous fsid |
| 152 |
change operation in case it was interrupted. |
| 153 |
.TP |
| 154 |
.BI \-U \ <UUID> |
| 155 |
Change fsid to \fIUUID\fP in all metadata blocks. |
| 156 |
.sp |
| 157 |
The \fIUUID\fP should be a 36 bytes string in printf(3) \%<https://\:man7\:.org/\:linux/\:man-pages/\:man3/\:printf\:.3\:.html> format |
| 158 |
\fB%08x\-%04x\-%04x\-%04x\-%012x\fP\&. |
| 159 |
If there is a previous unfinished fsid change, it will continue only if the |
| 160 |
\fIUUID\fP matches the unfinished one or if you use the option \fI\-u\fP\&. |
| 161 |
.sp |
| 162 |
All metadata blocks are rewritten, this may take some time, but the final |
| 163 |
filesystem compatibility is unaffected, unlike \fI\-M\fP\&. |
| 164 |
.sp |
| 165 |
\fBWarning:\fP |
| 166 |
.INDENT 7.0 |
| 167 |
.INDENT 3.5 |
| 168 |
Cancelling or interrupting a UUID change operation will make |
| 169 |
the filesystem temporarily unmountable. To fix it, rerun |
| 170 |
\fBbtrfstune \-u\fP and let it complete. |
| 171 |
.UNINDENT |
| 172 |
.UNINDENT |
| 173 |
.TP |
| 174 |
.B \-x |
| 175 |
(since kernel: 3.10) |
| 176 |
.sp |
| 177 |
Enable skinny metadata extent refs (more efficient representation of |
| 178 |
extents), enabled by mkfs feature \fIskinny\-metadata\fP\&. |
| 179 |
.sp |
| 180 |
All newly created extents will use the new representation. To |
| 181 |
completely switch the entire filesystem, run a full balance of the |
| 182 |
metadata. Please refer to btrfs\-balance(8) \%<>\&. |
| 183 |
.TP |
| 184 |
.B \-\-version |
| 185 |
Print the \fBbtrfstune\fP version, builtin features and exit. |
| 186 |
.UNINDENT |
| 187 |
.SH EXIT STATUS |
| 188 |
.sp |
| 189 |
\fBbtrfstune\fP returns 0 if no error happened, 1 otherwise. |
| 190 |
.SH COMPATIBILITY NOTE |
| 191 |
.sp |
| 192 |
This deprecated tool exists for historical reasons but is still in use today. |
| 193 |
Its functionality will be merged to the main tool, at which time \fBbtrfstune\fP |
| 194 |
will be declared obsolete and scheduled for removal. |
| 195 |
.SH SEE ALSO |
| 196 |
.sp |
| 197 |
btrfs(5) \%<>, |
| 198 |
btrfs\-balance(8) \%<>, |
| 199 |
mkfs.btrfs(8) \%<> |
| 200 |
.\" End of generated man page. |