252
252
to create the file system.
255
This section describes mount options which are specific to ext2, ext3,
256
and ext4. Other generic mount options may be used as well; see
259
.SH "Mount options for ext2"
260
The `ext2' filesystem is the standard Linux filesystem.
261
Since Linux 2.5.46, for most mount options the default
262
is determined by the filesystem superblock. Set them with
266
Support POSIX Access Control Lists (or not).
269
Set the behavior for the
273
behavior is to return in the
275
field the total number of blocks of the filesystem, while the
277
behavior (which is the default) is to subtract the overhead blocks
278
used by the ext2 filesystem and not available for file storage. Thus
280
% mount /k \-o minixdf; df /k; umount /k
285
Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
286
/dev/sda6#2630655#86954#2412169#3%#/k
289
% mount /k \-o bsddf; df /k; umount /k
294
Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
295
/dev/sda6#2543714#13#2412169#0%#/k
298
(Note that this example shows that one can add command line options
299
to the options given in
302
.BR check=none " or " nocheck
303
No checking is done at mount time. This is the default. This is fast.
306
every now and then, e.g.\& at boot time. The non-default behavior is unsupported
307
(check=normal and check=strict options have been removed). Note that these mount options
308
don't have to be supported if ext4 kernel driver is used for ext2 and ext3 filesystems.
311
Print debugging info upon each (re)mount.
313
.BR errors= { continue | remount-ro | panic }
314
Define the behavior when an error is encountered.
315
(Either ignore errors and just mark the filesystem erroneous and continue,
316
or remount the filesystem read-only, or panic and halt the system.)
317
The default is set in the filesystem superblock, and can be
321
.BR grpid | bsdgroups " and " nogrpid | sysvgroups
322
These options define what group id a newly created file gets.
325
is set, it takes the group id of the directory in which it is created;
326
otherwise (the default) it takes the fsgid of the current process, unless
327
the directory has the setgid bit set, in which case it takes the gid
328
from the parent directory, and also gets the setgid bit set
329
if it is a directory itself.
331
.BR grpquota | noquota | quota | usrquota
332
The usrquota (same as quota) mount option enables user quota support on the
333
filesystem. grpquota enables group quotas support. You need the quota utilities
334
to actually enable and manage the quota system.
337
Disables 32-bit UIDs and GIDs. This is for interoperability with older
338
kernels which only store and expect 16-bit values.
340
.BR oldalloc " or " orlov
341
Use old allocator or Orlov allocator for new inodes. Orlov is default.
343
\fBresgid=\fP\,\fIn\fP and \fBresuid=\fP\,\fIn\fP
344
The ext2 filesystem reserves a certain percentage of the available
345
space (by default 5%, see
349
These options determine who can use the reserved blocks.
350
(Roughly: whoever has the specified uid, or belongs to the specified group.)
353
Instead of block 1, use block
355
as superblock. This could be useful when the filesystem has been damaged.
356
(Earlier, copies of the superblock would be made every 8192 blocks: in
357
block 1, 8193, 16385, \&...\& (and one got thousands of copies on
358
a big filesystem). Since version 1.08,
360
has a \-s (sparse superblock) option to reduce the number of backup
361
superblocks, and since version 1.15 this is the default. Note
362
that this may mean that ext2 filesystems created by a recent
364
cannot be mounted r/w under Linux 2.0.*.)
365
The block number here uses 1\ k units. Thus, if you want to use logical
366
block 32768 on a filesystem with 4\ k blocks, use "sb=131072".
368
.BR user_xattr | nouser_xattr
369
Support "user." extended attributes (or not).
372
.SH "Mount options for ext3"
373
The ext3 filesystem is a version of the ext2 filesystem which has been
374
enhanced with journaling. It supports the same options as ext2 as
375
well as the following additions:
378
Update the ext3 filesystem's journal to the current format.
381
When a journal already exists, this option is ignored. Otherwise, it
382
specifies the number of the inode which will represent the ext3 filesystem's
383
journal file; ext3 will create a new journal, overwriting the old contents
384
of the file whose inode number is
387
.BR journal_dev=devnum / journal_path=path
388
When the external journal device's major/minor numbers
389
have changed, these options allow the user to specify
390
the new journal location. The journal device is
391
identified either through its new major/minor numbers encoded
392
in devnum, or via a path to the device.
394
.BR norecovery / noload
395
Don't load the journal on mounting. Note that
396
if the filesystem was not unmounted cleanly,
397
skipping the journal replay will lead to the
398
filesystem containing inconsistencies that can
399
lead to any number of problems.
401
.BR data= { journal | ordered | writeback }
402
Specifies the journaling mode for file data. Metadata is always journaled.
403
To use modes other than
405
on the root filesystem, pass the mode to the kernel as boot parameter, e.g.\&
406
.IR rootflags=data=journal .
410
All data is committed into the journal prior to being written into the
414
This is the default mode. All data is forced directly out to the main file
415
system prior to its metadata being committed to the journal.
418
Data ordering is not preserved \(en data may be written into the main
419
filesystem after its metadata has been committed to the journal.
420
This is rumoured to be the highest-throughput option. It guarantees
421
internal filesystem integrity, however it can allow old data to appear
422
in files after a crash and journal recovery.
426
Just print an error message if an error occurs in a file data buffer in
430
Abort the journal if an error occurs in a file data buffer in ordered mode.
432
.BR barrier=0 " / " barrier=1 "
433
This disables / enables the use of write barriers in the jbd code. barrier=0
434
disables, barrier=1 enables (default). This also requires an IO stack which can
435
support barriers, and if jbd gets an error on a barrier write, it will disable
436
barriers again with a warning. Write barriers enforce proper on-disk ordering
437
of journal commits, making volatile disk write caches safe to use, at some
438
performance penalty. If your disks are battery-backed in one way or another,
439
disabling barriers may safely improve performance.
442
Sync all data and metadata every
444
seconds. The default value is 5 seconds. Zero means default.
447
Enable Extended User Attributes. See the
452
Enable POSIX Access Control Lists. See the
456
.BR usrjquota=aquota.user | grpjquota=aquota.group | jqfmt=vfsv0
457
Apart from the old quota system (as in ext2, jqfmt=vfsold aka version 1 quota)
458
ext3 also supports journaled quotas (version 2 quota). jqfmt=vfsv0
459
enables journaled quotas. For journaled quotas the mount options
460
usrjquota=aquota.user and grpjquota=aquota.group are required to tell the
461
quota system which quota database files to use. Journaled quotas have the
462
advantage that even after a crash no quota check is required.
464
.SH "Mount options for ext4"
465
The ext4 filesystem is an advanced level of the ext3 filesystem which
466
incorporates scalability and reliability enhancements for supporting large
470
.B journal_dev, norecovery, noload, data, commit, orlov, oldalloc, [no]user_xattr
471
.B [no]acl, bsddf, minixdf, debug, errors, data_err, grpid, bsdgroups, nogrpid
472
.B sysvgroups, resgid, resuid, sb, quota, noquota, grpquota, usrquota
473
.B usrjquota, grpjquota and jqfmt
474
are backwardly compatible with ext3 or ext2.
477
Enable checksumming of the journal transactions. This will allow the recovery
478
code in e2fsck and the kernel to detect corruption in the kernel. It is a
479
compatible change and will be ignored by older kernels.
481
.B journal_async_commit
482
Commit block can be written to disk without waiting for descriptor blocks. If
483
enabled older kernels cannot mount the device.
484
This will enable 'journal_checksum' internally.
486
.BR barrier=0 " / " barrier=1 " / " barrier " / " nobarrier
487
These mount options have the same effect as in ext3. The mount options
488
"barrier" and "nobarrier" are added for consistency with other ext4 mount
491
The ext4 filesystem enables write barriers by default.
493
.BI inode_readahead_blks= n
494
This tuning parameter controls the maximum number of inode table blocks that
495
ext4's inode table readahead algorithm will pre-read into the buffer cache.
496
The value must be a power of 2. The default value is 32 blocks.
499
Number of filesystem blocks that mballoc will try to use for allocation size
500
and alignment. For RAID5/6 systems this should be the number of data disks *
501
RAID chunk size in filesystem blocks.
504
Deferring block allocation until write-out time.
507
Disable delayed allocation. Blocks are allocated when data is copied from user
510
.BI max_batch_time= usec
511
Maximum amount of time ext4 should wait for additional filesystem operations to
512
be batch together with a synchronous write operation. Since a synchronous
513
write operation is going to force a commit and then a wait for the I/O
514
complete, it doesn't cost much, and can be a huge throughput win, we wait for a
515
small amount of time to see if any other transactions can piggyback on the
516
synchronous write. The algorithm used is designed to automatically tune for
517
the speed of the disk, by measuring the amount of time (on average) that it
518
takes to finish committing a transaction. Call this time the "commit time".
519
If the time that the transaction has been running is less than the commit time,
520
ext4 will try sleeping for the commit time to see if other operations will join
521
the transaction. The commit time is capped by the max_batch_time, which
522
defaults to 15000\ \[mc]s (15\ ms). This optimization can be turned off entirely by
523
setting max_batch_time to 0.
525
.BI min_batch_time= usec
526
This parameter sets the commit time (as described above) to be at least
527
min_batch_time. It defaults to zero microseconds. Increasing this parameter
528
may improve the throughput of multi-threaded, synchronous workloads on very
529
fast disks, at the cost of increasing latency.
531
.BI journal_ioprio= prio
532
The I/O priority (from 0 to 7, where 0 is the highest priority) which should be
533
used for I/O operations submitted by kjournald2 during a commit operation.
534
This defaults to 3, which is a slightly higher priority than the default I/O
538
Simulate the effects of calling ext4_abort() for
539
debugging purposes. This is normally used while
540
remounting a filesystem which is already mounted.
542
.BR auto_da_alloc | noauto_da_alloc
543
Many broken applications don't use fsync() when
544
replacing existing files via patterns such as
546
fd = open("foo.new")/write(fd,...)/close(fd)/ rename("foo.new", "foo")
550
fd = open("foo", O_TRUNC)/write(fd,...)/close(fd).
552
If auto_da_alloc is enabled, ext4 will detect the replace-via-rename and
553
replace-via-truncate patterns and force that any delayed allocation blocks are
554
allocated such that at the next journal commit, in the default data=ordered
555
mode, the data blocks of the new file are forced to disk before the rename()
556
operation is committed. This provides roughly the same level of guarantees as
557
ext3, and avoids the "zero-length" problem that can happen when a system
558
crashes before the delayed allocation blocks are forced to disk.
561
Do not initialize any uninitialized inode table blocks in the background. This
562
feature may be used by installation CD's so that the install process can
563
complete as quickly as possible; the inode table initialization process would
564
then be deferred until the next time the filesystem is mounted.
567
The lazy itable init code will wait n times the number of milliseconds it took
568
to zero out the previous block group's inode table. This minimizes the impact on
569
system performance while the filesystem's inode table is being initialized.
571
.BR discard / nodiscard
572
Controls whether ext4 should issue discard/TRIM commands to the underlying
573
block device when blocks are freed. This is useful for SSD devices and
574
sparse/thinly-provisioned LUNs, but it is off by default until sufficient
575
testing has been done.
578
Disables 32-bit UIDs and GIDs. This is for
579
interoperability with older kernels which only
580
store and expect 16-bit values.
582
.BR block_validity / noblock_validity
583
This options allows to enables/disables the in-kernel facility for tracking
584
filesystem metadata blocks within internal data structures. This allows multi-\c
585
block allocator and other routines to quickly locate extents which might
586
overlap with filesystem metadata blocks. This option is intended for debugging
587
purposes and since it negatively affects the performance, it is off by default.
589
.BR dioread_lock / dioread_nolock
590
Controls whether or not ext4 should use the DIO read locking. If the
591
dioread_nolock option is specified ext4 will allocate uninitialized extent
592
before buffer write and convert the extent to initialized after IO completes.
593
This approach allows ext4 code to avoid using inode mutex, which improves
594
scalability on high speed storages. However this does not work with data
595
journaling and dioread_nolock option will be ignored with kernel warning.
596
Note that dioread_nolock code path is only used for extent-based files.
597
Because of the restrictions this options comprises it is off by default
598
(e.g.\& dioread_lock).
601
This limits the size of the directories so that any attempt to expand them
602
beyond the specified limit in kilobytes will cause an ENOSPC error. This is
603
useful in memory-constrained environments, where a very large directory can
604
cause severe performance problems or even provoke the Out Of Memory killer. (For
605
example, if there is only 512\ MB memory available, a 176\ MB directory may
606
seriously cramp the system's style.)
609
Enable 64-bit inode version support. This option is off by default.
612
The ext2, ext3, and ext4 filesystems support setting the following file
613
attributes on Linux systems using the
617
.BR a " - append only"
619
.BR A " - no atime updates"
623
.BR D " - synchronous directory updates"
627
.BR S " - synchronous updates"
629
.BR u " - undeletable"
631
In addition, the ext3 and ext4 filesystems support the following flag:
633
.BR j " - data journaling"
635
Finally, the ext4 filesystem also supports the following flag:
637
.BR e " - extents format"
639
For descriptions of these attribute flags, please refer to the
256
644
.BR mke2fs.conf (5),
258
646
.BR dumpe2fs (8),
added
removed
misc/ext4.5.in
