2 .\" Copyright 1993, 1994, 1995 by Theodore Ts'o. All Rights Reserved.
3 .\" This file may be copied under the terms of the GNU Public License.
5 .TH E2FSCK 8 "@E2FSPROGS_MONTH@ @E2FSPROGS_YEAR@" "E2fsprogs version @E2FSPROGS_VERSION@"
7 e2fsck \- check a Linux ext2/ext3/ext4 file system
31 @JDEV@.I external-journal
44 is used to check the ext2/ext3/ext4 family of file systems.
45 For ext3 and ext4 file systems that use a journal, if the system has been
46 shut down uncleanly without any errors, normally, after replaying the
47 committed transactions in the journal, the file system should be
49 Hence, for file systems that use journaling,
51 will normally replay the journal and exit, unless its superblock
52 indicates that further checking is required.
55 is a block device (e.g.,
57 or file containing the file system.
59 Note that in general it is not safe to run
61 on mounted file systems. The only exception is if the
63 option is specified, and
71 However, even if it is safe to do so, the results printed by
73 are not valid if the file system is mounted.
76 asks whether or not you should check a file system which is mounted,
77 the only correct answer is ``no''. Only experts who really know what
78 they are doing should consider answering this question in any other way.
82 is run in interactive mode (meaning that none of
87 are specified), the program will ask the user to fix each problem found in the
88 file system. A response of \&'y' will fix the error; \&'n' will leave the error
89 unfixed; and \&'a' will fix the problem and all subsequent problems; pressing
90 Enter will proceed with the default response, which is printed before the
91 question mark. Pressing Control-C terminates e2fsck immediately.
95 This option does the same thing as the
97 option. It is provided for backwards compatibility only; it is
98 suggested that people use
100 option whenever possible.
102 .BI \-b " superblock"
103 Instead of using the normal superblock, use an alternative superblock
106 This option is normally used when the primary superblock has been
107 corrupted. The location of backup superblocks is dependent on the
108 file system's blocksize, the number of blocks per group, and features
112 Additional backup superblocks can be determined by using the
116 option to print out where the superblocks exist, supposing
118 is supplied with arguments that are consistent with the file system's layout
119 (e.g.\& blocksize, blocks per group,
123 If an alternative superblock is specified and
124 the file system is not opened read-only, e2fsck will make sure that the
125 primary superblock is updated appropriately upon completion of the
131 will search for the superblock at various different
132 block sizes in an attempt to find the appropriate block size.
133 This search can be fooled in some cases. This option forces
135 to only try locating the superblock at a particular blocksize.
136 If the superblock is not found,
138 will terminate with a fatal error.
145 program to do a read-only scan of the device in order to find any bad
146 blocks. If any bad blocks are found, they are added to the bad block
147 inode to prevent them from being allocated to a file or directory. If
148 this option is specified twice, then the bad block scan will be done
149 using a non-destructive read-write test.
154 to write completion information to the specified file descriptor
155 so that the progress of the file system
156 check can be monitored. This option is typically used by programs
159 If the file descriptor number is negative, then absolute value of
160 the file descriptor will be used, and the progress information will be
161 suppressed initially. It can later be enabled by sending the
163 process a SIGUSR1 signal.
164 If the file descriptor specified is 0,
166 will print a completion bar as it goes about its business. This requires
167 that e2fsck is running on a video console or terminal.
170 Print debugging output (useless unless you are debugging
174 Optimize directories in file system. This option causes e2fsck to
175 try to optimize all directories, either by re-indexing them if the
176 file system supports directory indexing, or by sorting and compressing
177 directories for smaller directories, or for file systems using
178 traditional linear directories.
184 may sometimes optimize a few directories --- for example, if
185 directory indexing is enabled and a directory is not indexed and would
186 benefit from being indexed, or if the index structures are corrupted
187 and need to be rebuilt. The
189 option forces all directories in the file system to be optimized. This can
190 sometimes make them a little smaller and slightly faster to search, but
191 in practice, you should rarely need to use this option.
195 option will detect directory entries with duplicate names in a single
196 directory, which e2fsck normally does not enforce for performance reasons.
198 .BI \-E " extended_options"
199 Set e2fsck extended options. Extended options are comma
200 separated, and may take an argument using the equals ('=') sign. The
201 following options are supported:
204 .BI ea_ver= extended_attribute_version
205 Set the version of the extended attribute blocks which
207 will require while checking the file system. The version number may
208 be 1 or 2. The default extended attribute version format is 2.
211 Only replay the journal if required, but do not perform any further checks
215 During pass 1, print a detailed report of any discontiguous blocks for
216 files in the file system.
219 Attempt to discard free blocks and unused inode blocks after the full
220 file system check (discarding blocks is useful on solid state devices and sparse
221 / thin-provisioned storage).
222 Note that discard is done in pass 5 AFTER the
223 file system has been fully checked and only if it does not contain recognizable
225 However there might be cases where
227 does not fully recognize a problem and hence in this case this
228 option may prevent you from further manual data recovery.
231 Do not attempt to discard free blocks and unused inode blocks.
232 This option is exactly the opposite of discard option.
233 This is set as default.
235 .B no_optimize_extents
236 Do not offer to optimize the extent tree by eliminating unnecessary
237 width or depth. This can also be enabled in the options section of
238 .BR /etc/e2fsck.conf .
241 Offer to optimize the extent tree by eliminating unnecessary
242 width or depth. This is the default unless otherwise specified in
243 .BR /etc/e2fsck.conf .
245 .B inode_count_fullmap
246 Trade off using memory for speed when checking a file system with a
247 large number of hard-linked files. The amount of memory required is
248 proportional to the number of inodes in the file system. For large file
249 systems, this can be gigabytes of memory. (For example, a 40\ TB file system
250 with 2.8 billion inodes will consume an additional 5.7 GB memory if this
251 optimization is enabled.) This optimization can also be enabled in the
253 .BR /etc/e2fsck.conf .
255 .B no_inode_count_fullmap
257 .B inode_count_fullmap
258 optimization. This is the default unless otherwise specified in
259 .BR /etc/e2fsck.conf .
262 Use this many KiB of memory to pre-fetch metadata in the hopes of reducing
263 e2fsck runtime. By default, this is set to the size of two block groups' inode
264 tables (typically 4\ MiB on a regular ext4 file system); if this amount is more
265 than 1/50th of total physical memory, readahead is disabled. Set this to zero
266 to disable readahead entirely.
269 Convert block-mapped files to extent-mapped files.
272 Only fix damaged metadata; do not optimize htree directories or compress
273 extent trees. This option is incompatible with the \-D and \-E bmap2extent
277 Force verification of encoded filenames in case-insensitive directories.
278 This is the default mode if the file system has the strict flag enabled.
281 If the file system has shared blocks, with the shared blocks read-only feature
282 enabled, then this will unshare all shared blocks and unset the read-only
284 If there is not enough free space then the operation will fail.
285 If the file system does not have the read-only feature bit, but has shared
286 blocks anyway, then this option will have no effect.
288 option, if there is no free space to clone blocks, there is no prompt to
289 delete files and instead the operation will fail.
291 Note that unshare_blocks implies the "\-f" option to ensure that all passes
293 Additionally, if "\-n" is also specified, e2fsck will simulate trying
294 to allocate enough space to deduplicate.
295 If this fails, the exit code will be non-zero.
299 Force checking even if the file system seems clean.
302 Flush the file system device's buffer caches before beginning. Only
303 really useful for doing
307 @JDEV@.BI \-j " external-journal"
308 @JDEV@Set the pathname where the external-journal for this file system can be
312 When combined with the
314 option, any existing bad blocks in the bad blocks list are preserved,
315 and any new bad blocks found by running
317 will be added to the existing bad blocks list.
320 Add the block numbers listed in the file specified by
322 to the list of bad blocks. The format of this file is the same as the
325 program. Note that the block numbers are based on the blocksize
326 of the file system. Hence,
328 must be given the blocksize of the file system in order to obtain correct
329 results. As a result, it is much simpler and safer to use the
333 since it will assure that the correct parameters are passed to the
338 Set the bad blocks list to be the list of blocks specified by
340 (This option is the same as the
342 option, except the bad blocks list is cleared before the blocks listed
343 in the file are added to the bad blocks list.)
346 Run e2fsck with up to the specified number of
348 The actual number of threads may be lower, if the filesystem does not
349 have enough block groups to effectively parallelize the workload.
352 Open the file system read-only, and assume an answer of `no' to all
355 to be used non-interactively. This option
356 may not be specified at the same time as the
363 Automatically repair ("preen") the file system. This option will cause
366 fix any file system problems that can be safely fixed without human
369 discovers a problem which may require the system administrator
370 to take additional corrective action,
372 will print a description of the problem and then exit with the value 4
373 logically or'ed into the exit code. (See the \fBEXIT CODE\fR section.)
374 This option is normally used by the system's boot scripts. It may not
375 be specified at the same time as the
382 This option does nothing at all; it is provided only for backwards
386 Print timing statistics for
388 If this option is used twice, additional timing statistics are printed
389 on a pass by pass basis.
395 Print version information and exit.
398 Assume an answer of `yes' to all questions; allows
400 to be used non-interactively. This option
401 may not be specified at the same time as the
408 Before overwriting a file system block, write the old contents of the block to
409 an undo file. This undo file can be used with
412 contents of the file system should something go wrong. If the empty string is
413 passed as the undo_file argument, the undo file will be written to a file named
414 e2fsck-\fIdevice\fR.e2undo in the directory specified via the
415 \fIE2FSPROGS_UNDO_DIR\fR environment variable.
417 WARNING: The undo file cannot be used to recover from a power or system crash.
419 The exit code returned by
421 is the sum of the following conditions:
425 1 \-\ File system errors corrected
427 2 \-\ File system errors corrected, system should
431 4 \-\ File system errors left uncorrected
433 8 \-\ Operational error
435 16 \-\ Usage or syntax error
437 32 \-\ E2fsck canceled by user request
439 128 \-\ Shared library error
441 The following signals have the following effect when sent to
447 to start displaying a completion bar or emitting progress information.
448 (See discussion of the
455 to stop displaying a completion bar or emitting progress information.
457 Almost any piece of software will have bugs. If you manage to find a
458 file system which causes
462 is unable to repair, please report it to the author.
464 Please include as much information as possible in your bug report.
465 Ideally, include a complete transcript of the
467 run, so I can see exactly what error messages are displayed. (Make sure
468 the messages printed by
470 are in English; if your system has been
473 messages have been translated into another language, please set the
475 environment variable to
477 so that the transcript of e2fsck's output will be useful to me.)
479 have a writable file system where the transcript can be stored, the
481 program is a handy way to save the output of
485 It is also useful to send the output of
487 If a specific inode or inodes seems to be giving
489 trouble, try running the
491 command and send the output of the
493 command run on the relevant inode(s). If the inode is a directory, the
496 command will allow you to extract the contents of the directory inode,
497 which can sent to me after being first run through
499 The most useful data you can send to help reproduce
500 the bug is a compressed raw image dump of the file system, generated using
504 man page for more details.
506 Always include the full version string which
508 displays when it is run, so I know which version you are running.
512 Determines the location of the configuration file (see
513 .BR e2fsck.conf (5)).
517 was written by Theodore Ts'o <tytso@mit.edu>.