1MKE2FS(8) System Manager's Manual MKE2FS(8)
2
6 mke2fs - create an ext2/ext3/ext4 filesystem
7
9 mke2fs [ -c | -l filename ] [ -b block-size ] [ -C cluster-size ] [ -d
10 root-directory ] [ -D ] [ -g blocks-per-group ] [ -G number-of-groups ]
11 [ -i bytes-per-inode ] [ -I inode-size ] [ -j ] [ -J journal-options ]
12 [ -N number-of-inodes ] [ -n ] [ -m reserved-blocks-percentage ] [ -o
13 creator-os ] [ -O [^]feature[,...] ] [ -q ] [ -r fs-revision-level ] [
14 -E extended-options ] [ -v ] [ -F ] [ -L volume-label ] [ -M last-
15 mounted-directory ] [ -S ] [ -t fs-type ] [ -T usage-type ] [ -U UUID ]
16 [ -V ] [ -e errors-behavior ] [ -z undo_file ] device [ fs-size ]
17 mke2fs -O journal_dev [ -b block-size ] [ -L volume-label ] [ -n ] [ -q
19 ] [ -v ] external-journal [ fs-size ]
20
22 mke2fs is used to create an ext2, ext3, or ext4 filesystem, usually in
23 a disk partition (or file) named by device.
24 The file system size is specified by fs-size. If fs-size does not have
26 a suffix, it is interpreted as power-of-two kilobytes, unless the -b
27 blocksize option is specified, in which case fs-size is interpreted as
28 the number of blocksize blocks. If the fs-size is suffixed by 'k',
29 'm', 'g', 't' (either upper-case or lower-case), then it is interpreted
30 in power-of-two kilobytes, megabytes, gigabytes, terabytes, etc. If
31 fs-size is omitted, mke2fs will create the file system based on the
32 device size.
33 If mke2fs is run as mkfs.XXX (i.e., mkfs.ext2, mkfs.ext3, or mkfs.ext4)
35 the option -t XXX is implied; so mkfs.ext3 will create a file system
36 for use with ext3, mkfs.ext4 will create a file system for use with
37 ext4, and so on.
38 The defaults of the parameters for the newly created filesystem, if not
40 overridden by the options listed below, are controlled by the
41 /etc/mke2fs.conf configuration file. See the mke2fs.conf(5) manual
42 page for more details.
43
45 -b block-size
46 Specify the size of blocks in bytes. Valid block-size values
47 are 1024, 2048 and 4096 bytes per block. If omitted, block-size
48 is heuristically determined by the filesystem size and the
49 expected usage of the filesystem (see the -T option). If block-
50 size is preceded by a negative sign ('-'), then mke2fs will use
51 heuristics to determine the appropriate block size, with the
52 constraint that the block size will be at least block-size
53 bytes. This is useful for certain hardware devices which
54 require that the blocksize be a multiple of 2k.
55 -c Check the device for bad blocks before creating the file system.
57 If this option is specified twice, then a slower read-write test
58 is used instead of a fast read-only test.
59 -C cluster-size
61 Specify the size of cluster in bytes for filesystems using the
62 bigalloc feature. Valid cluster-size values are from 2048 to
63 256M bytes per cluster. This can only be specified if the
64 bigalloc feature is enabled. (See the ext4 (5) man page for
65 more details about bigalloc.) The default cluster size if
66 bigalloc is enabled is 16 times the block size.
67 -d root-directory
69 Copy the contents of the given directory into the root directory
70 of the filesystem.
71 -D Use direct I/O when writing to the disk. This avoids mke2fs
73 dirtying a lot of buffer cache memory, which may impact other
74 applications running on a busy server. This option will cause
75 mke2fs to run much more slowly, however, so there is a tradeoff
76 to using direct I/O.
77 -e error-behavior
79 Change the behavior of the kernel code when errors are detected.
80 In all cases, a filesystem error will cause e2fsck(8) to check
81 the filesystem on the next boot. error-behavior can be one of
82 the following:
83 continue Continue normal execution.
85 remount-ro Remount filesystem read-only.
87 panic Cause a kernel panic.
89 -E extended-options
91 Set extended options for the filesystem. Extended options are
92 comma separated, and may take an argument using the equals ('=')
93 sign. The -E option used to be -R in earlier versions of
94 mke2fs. The -R option is still accepted for backwards compati‐
95 bility, but is deprecated. The following extended options are
96 supported:
97 mmp_update_interval=interval
99 Adjust the initial MMP update interval to interval
100 seconds. Specifying an interval of 0 means to use
101 the default interval. The specified interval must
102 be less than 300 seconds. Requires that the mmp
103 feature be enabled.
104 stride=stride-size
106 Configure the filesystem for a RAID array with
107 stride-size filesystem blocks. This is the number of
108 blocks read or written to disk before moving to the
109 next disk, which is sometimes referred to as the
110 chunk size. This mostly affects placement of
111 filesystem metadata like bitmaps at mke2fs time to
112 avoid placing them on a single disk, which can hurt
113 performance. It may also be used by the block allo‐
114 cator.
115 stripe_width=stripe-width
117 Configure the filesystem for a RAID array with
118 stripe-width filesystem blocks per stripe. This is
119 typically stride-size * N, where N is the number of
120 data-bearing disks in the RAID (e.g. for RAID 5
121 there is one parity disk, so N will be the number of
122 disks in the array minus 1). This allows the block
123 allocator to prevent read-modify-write of the parity
124 in a RAID stripe if possible when the data is writ‐
125 ten.
126 offset=offset
128 Create the filesystem at an offset from the begin‐
129 ning of the device or file. This can be useful when
130 creating disk images for virtual machines.
131 resize=max-online-resize
133 Reserve enough space so that the block group
134 descriptor table can grow to support a filesystem
135 that has max-online-resize blocks.
136 lazy_itable_init[= <0 to disable, 1 to enable>]
138 If enabled and the uninit_bg feature is enabled, the
139 inode table will not be fully initialized by mke2fs.
140 This speeds up filesystem initialization noticeably,
141 but it requires the kernel to finish initializing
142 the filesystem in the background when the filesystem
143 is first mounted. If the option value is omitted,
144 it defaults to 1 to enable lazy inode table zeroing.
145 lazy_journal_init[= <0 to disable, 1 to enable>]
147 If enabled, the journal inode will not be fully
148 zeroed out by mke2fs. This speeds up filesystem
149 initialization noticeably, but carries some small
150 risk if the system crashes before the journal has
151 been overwritten entirely one time. If the option
152 value is omitted, it defaults to 1 to enable lazy
153 journal inode zeroing.
154 no_copy_xattrs
156 Normally mke2fs will copy the extended attributes of
157 the files in the directory hierarchy specified via
158 the (optional) -d option. This will disable the
159 copy and leaves the files in the newly created file
160 system without any extended attributes.
161 num_backup_sb=<0|1|2>
163 If the sparse_super2 file system feature is enabled
164 this option controls whether there will be 0, 1, or
165 2 backup superblocks created in the file system.
166 packed_meta_blocks[= <0 to disable, 1 to enable>]
168 Place the allocation bitmaps and the inode table at
169 the beginning of the disk. This option requires
170 that the flex_bg file system feature to be enabled
171 in order for it to have effect, and will also create
172 the journal at the beginning of the file system.
173 This option is useful for flash devices that use SLC
174 flash at the beginning of the disk. It also maxi‐
175 mizes the range of contiguous data blocks, which can
176 be useful for certain specialized use cases, such as
177 supported Shingled Drives.
178 root_owner[=uid:gid]
180 Specify the numeric user and group ID of the root
181 directory. If no UID:GID is specified, use the user
182 and group ID of the user running mke2fs. In mke2fs
183 1.42 and earlier the UID and GID of the root direc‐
184 tory were set by default to the UID and GID of the
185 user running the mke2fs command. The root_owner=
186 option allows explicitly specifying these values,
187 and avoid side-effects for users that do not expect
188 the contents of the filesystem to change based on
189 the user running mke2fs.
190 test_fs
192 Set a flag in the filesystem superblock indicating
193 that it may be mounted using experimental kernel
194 code, such as the ext4dev filesystem.
195 discard
197 Attempt to discard blocks at mkfs time (discarding
198 blocks initially is useful on solid state devices
199 and sparse / thin-provisioned storage). When the
200 device advertises that discard also zeroes data (any
201 subsequent read after the discard and before write
202 returns zero), then mark all not-yet-zeroed inode
203 tables as zeroed. This significantly speeds up
204 filesystem initialization. This is set as default.
205 nodiscard
207 Do not attempt to discard blocks at mkfs time.
208 quotatype
210 Specify the which quota types (usrquota, grpquota,
211 prjquota) which should be enabled in the created
212 file system. The argument of this extended option
213 should be a colon separated list. This option has
214 effect only if the quota feature is set. The
215 default quota types to be initialized if this option
216 is not specified is both user and group quotas. If
217 the project feature is enabled that project quotas
218 will be initialized as well.
219 -F Force mke2fs to create a filesystem, even if the specified
221 device is not a partition on a block special device, or if other
222 parameters do not make sense. In order to force mke2fs to cre‐
223 ate a filesystem even if the filesystem appears to be in use or
224 is mounted (a truly dangerous thing to do), this option must be
225 specified twice.
226 -g blocks-per-group
228 Specify the number of blocks in a block group. There is gener‐
229 ally no reason for the user to ever set this parameter, as the
230 default is optimal for the filesystem. (For administrators who
231 are creating filesystems on RAID arrays, it is preferable to use
232 the stride RAID parameter as part of the -E option rather than
233 manipulating the number of blocks per group.) This option is
234 generally used by developers who are developing test cases.
235 If the bigalloc feature is enabled, the -g option will specify
237 the number of clusters in a block group.
238 -G number-of-groups
240 Specify the number of block groups that will be packed together
241 to create a larger virtual block group (or "flex_bg group") in
242 an ext4 filesystem. This improves meta-data locality and per‐
243 formance on meta-data heavy workloads. The number of groups
244 must be a power of 2 and may only be specified if the flex_bg
245 filesystem feature is enabled.
246 -i bytes-per-inode
248 Specify the bytes/inode ratio. mke2fs creates an inode for
249 every bytes-per-inode bytes of space on the disk. The larger
250 the bytes-per-inode ratio, the fewer inodes will be created.
251 This value generally shouldn't be smaller than the blocksize of
252 the filesystem, since in that case more inodes would be made
253 than can ever be used. Be warned that it is not possible to
254 change this ratio on a filesystem after it is created, so be
255 careful deciding the correct value for this parameter. Note
256 that resizing a filesystem changes the number of inodes to main‐
257 tain this ratio.
258 -I inode-size
260 Specify the size of each inode in bytes. The inode-size value
261 must be a power of 2 larger or equal to 128. The larger the
262 inode-size the more space the inode table will consume, and this
263 reduces the usable space in the filesystem and can also nega‐
264 tively impact performance. It is not possible to change this
265 value after the filesystem is created.
266 In kernels after 2.6.10 and some earlier vendor kernels it is
268 possible to utilize inodes larger than 128 bytes to store
269 extended attributes for improved performance. Extended
270 attributes stored in large inodes are not visible with older
271 kernels, and such filesystems will not be mountable with 2.4
272 kernels at all.
273 The default inode size is controlled by the mke2fs.conf(5) file.
275 In the mke2fs.conf file shipped with e2fsprogs, the default
276 inode size is 256 bytes for most file systems, except for small
277 file systems where the inode size will be 128 bytes.
278 -j Create the filesystem with an ext3 journal. If the -J option is
280 not specified, the default journal parameters will be used to
281 create an appropriately sized journal (given the size of the
282 filesystem) stored within the filesystem. Note that you must be
283 using a kernel which has ext3 support in order to actually make
284 use of the journal.
285 -J journal-options
287 Create the ext3 journal using options specified on the command-
288 line. Journal options are comma separated, and may take an
289 argument using the equals ('=') sign. The following journal
290 options are supported:
291 size=journal-size
293 Create an internal journal (i.e., stored inside the
294 filesystem) of size journal-size megabytes. The
295 size of the journal must be at least 1024 filesystem
296 blocks (i.e., 1MB if using 1k blocks, 4MB if using
297 4k blocks, etc.) and may be no more than 10,240,000
298 filesystem blocks or half the total file system size
299 (whichever is smaller)
300 location=journal-location
302 Specify the location of the journal. The argument
303 journal-location can either be specified as a block
304 number, or if the number has a units suffix (e.g.,
305 'M', 'G', etc.) interpret it as the offset from the
306 beginning of the file system.
307 device=external-journal
309 Attach the filesystem to the journal block device
310 located on external-journal. The external journal
311 must already have been created using the command
312 mke2fs -O journal_dev external-journal
314 Note that external-journal must have been created
316 with the same block size as the new filesystem. In
317 addition, while there is support for attaching mul‐
318 tiple filesystems to a single external journal, the
319 Linux kernel and e2fsck(8) do not currently support
320 shared external journals yet.
321 Instead of specifying a device name directly, exter‐
323 nal-journal can also be specified by either
324 LABEL=label or UUID=UUID to locate the external
325 journal by either the volume label or UUID stored in
326 the ext2 superblock at the start of the journal.
327 Use dumpe2fs(8) to display a journal device's volume
328 label and UUID. See also the -L option of
329 tune2fs(8).
330 Only one of the size or device options can be given for a
332 filesystem.
333 -l filename
335 Read the bad blocks list from filename. Note that the block
336 numbers in the bad block list must be generated using the same
337 block size as used by mke2fs. As a result, the -c option to
338 mke2fs is a much simpler and less error-prone method of checking
339 a disk for bad blocks before formatting it, as mke2fs will auto‐
340 matically pass the correct parameters to the badblocks program.
341 -L new-volume-label
343 Set the volume label for the filesystem to new-volume-label.
344 The maximum length of the volume label is 16 bytes.
345 -m reserved-blocks-percentage
347 Specify the percentage of the filesystem blocks reserved for the
348 super-user. This avoids fragmentation, and allows root-owned
349 daemons, such as syslogd(8), to continue to function correctly
350 after non-privileged processes are prevented from writing to the
351 filesystem. The default percentage is 5%.
352 -M last-mounted-directory
354 Set the last mounted directory for the filesystem. This might
355 be useful for the sake of utilities that key off of the last
356 mounted directory to determine where the filesystem should be
357 mounted.
358 -n Causes mke2fs to not actually create a filesystem, but display
360 what it would do if it were to create a filesystem. This can be
361 used to determine the location of the backup superblocks for a
362 particular filesystem, so long as the mke2fs parameters that
363 were passed when the filesystem was originally created are used
364 again. (With the -n option added, of course!)
365 -N number-of-inodes
367 Overrides the default calculation of the number of inodes that
368 should be reserved for the filesystem (which is based on the
369 number of blocks and the bytes-per-inode ratio). This allows
370 the user to specify the number of desired inodes directly.
371 -o creator-os
373 Overrides the default value of the "creator operating system"
374 field of the filesystem. The creator field is set by default to
375 the name of the OS the mke2fs executable was compiled for.
376 -O [^]feature[,...]
378 Create a filesystem with the given features (filesystem
379 options), overriding the default filesystem options. The fea‐
380 tures that are enabled by default are specified by the base_fea‐
381 tures relation, either in the [defaults] section in the
382 /etc/mke2fs.conf configuration file, or in the [fs_types] sub‐
383 sections for the usage types as specified by the -T option, fur‐
384 ther modified by the features relation found in the [fs_types]
385 subsections for the filesystem and usage types. See the
386 mke2fs.conf(5) manual page for more details. The filesystem
387 type-specific configuration setting found in the [fs_types] sec‐
388 tion will override the global default found in [defaults].
389 The filesystem feature set will be further edited using either
391 the feature set specified by this option, or if this option is
392 not given, by the default_features relation for the filesystem
393 type being created, or in the [defaults] section of the configu‐
394 ration file.
395 The filesystem feature set is comprised of a list of features,
397 separated by commas, that are to be enabled. To disable a fea‐
398 ture, simply prefix the feature name with a caret ('^') charac‐
399 ter. Features with dependencies will not be removed success‐
400 fully. The pseudo-filesystem feature "none" will clear all
401 filesystem features.
402 For more information about the features which can be set, please see
404 the manual page ext4(5).
405 -q Quiet execution. Useful if mke2fs is run in a script.
407 -r revision
409 Set the filesystem revision for the new filesystem. Note that
410 1.2 kernels only support revision 0 filesystems. The default is
411 to create revision 1 filesystems.
412 -S Write superblock and group descriptors only. This is an extreme
414 measure to be taken only in the very unlikely case that all of
415 the superblock and backup superblocks are corrupted, and a last-
416 ditch recovery method is desired by experienced users. It
417 causes mke2fs to reinitialize the superblock and group descrip‐
418 tors, while not touching the inode table and the block and inode
419 bitmaps. The e2fsck program should be run immediately after
420 this option is used, and there is no guarantee that any data
421 will be salvageable. Due to the wide variety of possible
422 options to mke2fs that affect the on-disk layout, it is critical
423 to specify exactly the same format options, such as blocksize,
424 fs-type, feature flags, and other tunables when using this
425 option, or the filesystem will be further corrupted. In some
426 cases, such as filesystems that have been resized, or have had
427 features enabled after format time, it is impossible to over‐
428 write all of the superblocks correctly, and at least some
429 filesystem corruption will occur. It is best to run this on a
430 full copy of the filesystem so other options can be tried if
431 this doesn't work.
432 -t fs-type
434 Specify the filesystem type (i.e., ext2, ext3, ext4, etc.) that
435 is to be created. If this option is not specified, mke2fs will
436 pick a default either via how the command was run (for example,
437 using a name of the form mkfs.ext2, mkfs.ext3, etc.) or via a
438 default as defined by the /etc/mke2fs.conf file. This option
439 controls which filesystem options are used by default, based on
440 the fstypes configuration stanza in /etc/mke2fs.conf.
441 If the -O option is used to explicitly add or remove filesystem
443 options that should be set in the newly created filesystem, the
444 resulting filesystem may not be supported by the requested fs-
445 type. (e.g., "mke2fs -t ext3 -O extent /dev/sdXX" will create a
446 filesystem that is not supported by the ext3 implementation as
447 found in the Linux kernel; and "mke2fs -t ext3 -O ^has_journal
448 /dev/hdXX" will create a filesystem that does not have a journal
449 and hence will not be supported by the ext3 filesystem code in
450 the Linux kernel.)
451 -T usage-type[,...]
453 Specify how the filesystem is going to be used, so that mke2fs
454 can choose optimal filesystem parameters for that use. The
455 usage types that are supported are defined in the configuration
456 file /etc/mke2fs.conf. The user may specify one or more usage
457 types using a comma separated list.
458 If this option is is not specified, mke2fs will pick a single
460 default usage type based on the size of the filesystem to be
461 created. If the filesystem size is less than 3 megabytes,
462 mke2fs will use the filesystem type floppy. If the filesystem
463 size is greater than or equal to 3 but less than 512 megabytes,
464 mke2fs(8) will use the filesystem type small. If the filesystem
465 size is greater than or equal to 4 terabytes but less than 16
466 terabytes, mke2fs(8) will use the filesystem type big. If the
467 filesystem size is greater than or equal to 16 terabytes,
468 mke2fs(8) will use the filesystem type huge. Otherwise,
469 mke2fs(8) will use the default filesystem type default.
470 -U UUID
472 Set the universally unique identifier (UUID) of the filesystem
473 to UUID. The format of the UUID is a series of hex digits sepa‐
474 rated by hyphens, like this:
475 "c1b9d5a2-f162-11cf-9ece-0020afc76f16". The UUID parameter may
476 also be one of the following:
477 clear clear the filesystem UUID
479 random generate a new randomly-generated UUID
481 time generate a new time-based UUID
483 -v Verbose execution.
485 -V Print the version number of mke2fs and exit.
487 -z undo_file
489 Before overwriting a file system block, write the old contents
490 of the block to an undo file. This undo file can be used with
491 e2undo(8) to restore the old contents of the file system should
492 something go wrong. If the empty string is passed as the
493 undo_file argument, the undo file will be written to a file
494 named mke2fs-device.e2undo in the directory specified via the
495 E2FSPROGS_UNDO_DIR environment variable or the undo_dir direc‐
496 tive in the configuration file.
497 WARNING: The undo file cannot be used to recover from a power or
499 system crash.
500
502 MKE2FS_SYNC
503 If set to non-zero integer value, its value is used to determine
504 how often sync(2) is called during inode table initialization.
505 MKE2FS_CONFIG
507 Determines the location of the configuration file (see
508 mke2fs.conf(5)).
509 MKE2FS_FIRST_META_BG
511 If set to non-zero integer value, its value is used to determine
512 first meta block group. This is mostly for debugging purposes.
513 MKE2FS_DEVICE_SECTSIZE
515 If set to non-zero integer value, its value is used to determine
516 logical sector size of the device.
517 MKE2FS_DEVICE_PHYS_SECTSIZE
519 If set to non-zero integer value, its value is used to determine
520 physical sector size of the device.
521 MKE2FS_SKIP_CHECK_MSG
523 If set, do not show the message of filesystem automatic check
524 caused by mount count or check interval.
525
527 This version of mke2fs has been written by Theodore Ts'o
528 <tytso@mit.edu>.
529
531 mke2fs is part of the e2fsprogs package and is available from
532 http://e2fsprogs.sourceforge.net.
533
535 mke2fs.conf(5), badblocks(8), dumpe2fs(8), e2fsck(8), tune2fs(8),
536 ext4(5)
537E2fsprogs version 1.44.6 March 2019 MKE2FS(8)