1xfs_repair(8) System Manager's Manual xfs_repair(8)
2
6 xfs_repair - repair an XFS filesystem
7
9 xfs_repair [ -dfLPv ] [ -n | -e ] [ -m maxmem ] [ -c subopt=value ] [
10 -o subopt[=value] ] [ -t interval ] [ -l logdev ] [ -r rtdev ] device
11 xfs_repair -V
12
14 xfs_repair repairs corrupt or damaged XFS filesystems (see xfs(5)).
15 The filesystem is specified using the device argument which should be
16 the device name of the disk partition or volume containing the filesys‐
17 tem. If given the name of a block device, xfs_repair will attempt to
18 find the raw device associated with the specified block device and will
19 use the raw device instead.
20 Regardless, the filesystem to be repaired must be unmounted, otherwise,
22 the resulting filesystem may be inconsistent or corrupt.
23
25 -f Specifies that the filesystem image to be processed is stored in
26 a regular file at device (see the mkfs.xfs -d file option). This
27 might happen if an image copy of a filesystem has been copied or
28 written into an ordinary file. This option implies that any
29 external log or realtime section is also in an ordinary file.
30 -L Force Log Zeroing. Forces xfs_repair to zero the log even if it
32 is dirty (contains metadata changes). When using this option
33 the filesystem will likely appear to be corrupt, and can cause
34 the loss of user files and/or data. See the DIRTY LOGS section
35 for more information.
36 -l logdev
38 Specifies the device special file where the filesystem's exter‐
39 nal log resides. Only for those filesystems which use an exter‐
40 nal log. See the mkfs.xfs -l option, and refer to xfs(5) for a
41 detailed description of the XFS log.
42 -r rtdev
44 Specifies the device special file where the filesystem's real‐
45 time section resides. Only for those filesystems which use a
46 realtime section. See the mkfs.xfs -r option, and refer to
47 xfs(5) for a detailed description of the XFS realtime section.
48 -n No modify mode. Specifies that xfs_repair should not modify the
50 filesystem but should only scan the filesystem and indicate what
51 repairs would have been made. This option cannot be used
52 together with -e.
53 -P Disable prefetching of inode and directory blocks. Use this
55 option if you find xfs_repair gets stuck and stops proceeding.
56 Interrupting a stuck xfs_repair is safe.
57 -m maxmem
59 Specifies the approximate maximum amount of memory, in
60 megabytes, to use for xfs_repair. xfs_repair has its own inter‐
61 nal block cache which will scale out up to the lesser of the
62 process's virtual address limit or about 75% of the system's
63 physical RAM. This option overrides these limits.
64 NOTE: These memory limits are only approximate and may use more
66 than the specified limit.
67 -c subopt=value
69 Change filesystem parameters. Refer to xfs_admin(8) for informa‐
70 tion on changing filesystem parameters.
71 -o subopt[=value]
73 Override what the program might conclude about the filesystem if
74 left to its own devices.
75 The suboptions supported are:
77 bhash=bhashsize
79 overrides the default buffer cache hash size. The
80 total number of buffer cache entries are limited to 8
81 times this amount. The default size is set to use up
82 the remainder of 75% of the system's physical RAM
83 size.
84 ag_stride=ags_per_concat_unit
86 This creates additional processing threads to parallel
87 process AGs that span multiple concat units. This can
88 significantly reduce repair times on concat based
89 filesystems.
90 force_geometry
92 Check the filesystem even if geometry information
93 could not be validated. Geometry information can not
94 be validated if only a single allocation group exists
95 and thus we do not have a backup superblock available,
96 or if there are two allocation groups and the two
97 superblocks do not agree on the filesystem geometry.
98 Only use this option if you validated the geometry
99 yourself and know what you are doing. If In doubt run
100 in no modify mode first.
101 -t interval
103 Modify reporting interval, specified in seconds. During long
104 runs xfs_repair outputs its progress every 15 minutes. Reporting
105 is only activated when ag_stride is enabled.
106 -v Verbose output. May be specified multiple times to increase
108 verbosity.
109 -d Repair dangerously. Allow xfs_repair to repair an XFS filesystem
111 mounted read only. This is typically done on a root filesystem
112 from single user mode, immediately followed by a reboot.
113 -e If any metadata corruption was repaired, the status returned is
115 4 instead of the usual 0. This option cannot be used together
116 with -n.
117 -V Prints the version number and exits.
119 Checks Performed
121 Inconsistencies corrected include the following:
122 1. Inode and inode blockmap (addressing) checks: bad magic number
124 in inode, bad magic numbers in inode blockmap blocks, extents
125 out of order, incorrect number of records in inode blockmap
126 blocks, blocks claimed that are not in a legal data area of the
127 filesystem, blocks that are claimed by more than one inode.
128 2. Inode allocation map checks: bad magic number in inode map
130 blocks, inode state as indicated by map (free or in-use) incon‐
131 sistent with state indicated by the inode, inodes referenced by
132 the filesystem that do not appear in the inode allocation map,
133 inode allocation map referencing blocks that do not appear to
134 contain inodes.
135 3. Size checks: number of blocks claimed by inode inconsistent with
137 inode size, directory size not block aligned, inode size not
138 consistent with inode format.
139 4. Directory checks: bad magic numbers in directory blocks, incor‐
141 rect number of entries in a directory block, bad freespace
142 information in a directory leaf block, entry pointing to an
143 unallocated (free) or out of range inode, overlapping entries,
144 missing or incorrect dot and dotdot entries, entries out of
145 hashvalue order, incorrect internal directory pointers, direc‐
146 tory type not consistent with inode format and size.
147 5. Pathname checks: files or directories not referenced by a path‐
149 name starting from the filesystem root, illegal pathname compo‐
150 nents.
151 6. Link count checks: link counts that do not agree with the number
153 of directory references to the inode.
154 7. Freemap checks: blocks claimed free by the freemap but also
156 claimed by an inode, blocks unclaimed by any inode but not
157 appearing in the freemap.
158 8. Super Block checks: total free block and/or free i-node count
160 incorrect, filesystem geometry inconsistent, secondary and pri‐
161 mary superblocks contradictory.
162 Orphaned files and directories (allocated, in-use but unreferenced) are
164 reconnected by placing them in the lost+found directory. The name
165 assigned is the inode number.
166 Disk Errors
168 xfs_repair aborts on most disk I/O errors. Therefore, if you are trying
169 to repair a filesystem that was damaged due to a disk drive failure,
170 steps should be taken to ensure that all blocks in the filesystem are
171 readable and writable before attempting to use xfs_repair to repair the
172 filesystem. A possible method is using dd(8) to copy the data onto a
173 good disk.
174 lost+found
176 The directory lost+found does not have to already exist in the filesys‐
177 tem being repaired. If the directory does not exist, it is automati‐
178 cally created if required. If it already exists, it will be checked
179 for consistency and if valid will be used for additional orphaned
180 files. Invalid lost+found directories are removed and recreated. Exist‐
181 ing files in a valid lost+found are not removed or renamed.
182 Corrupted Superblocks
184 XFS has both primary and secondary superblocks. xfs_repair uses infor‐
185 mation in the primary superblock to automatically find and validate the
186 primary superblock against the secondary superblocks before proceeding.
187 Should the primary be too corrupted to be useful in locating the sec‐
188 ondary superblocks, the program scans the filesystem until it finds and
189 validates some secondary superblocks. At that point, it generates a
190 primary superblock.
191 Quotas
193 If quotas are in use, it is possible that xfs_repair will clear some or
194 all of the filesystem quota information. If so, the program issues a
195 warning just before it terminates. If all quota information is lost,
196 quotas are disabled and the program issues a warning to that effect.
197 Note that xfs_repair does not check the validity of quota limits. It is
199 recommended that you check the quota limit information manually after
200 xfs_repair. Also, space usage information is automatically regenerated
201 the next time the filesystem is mounted with quotas turned on, so the
202 next quota mount of the filesystem may take some time.
203
205 xfs_repair issues informative messages as it proceeds indicating what
206 it has found that is abnormal or any corrective action that it has
207 taken. Most of the messages are completely understandable only to
208 those who are knowledgeable about the structure of the filesystem.
209 Some of the more common messages are explained here. Note that the
210 language of the messages is slightly different if xfs_repair is run in
211 no-modify mode because the program is not changing anything on disk.
212 No-modify mode indicates what it would do to repair the filesystem if
213 run without the no-modify flag.
214 disconnected inode ino, moving to lost+found
216 An inode numbered ino was not connected to the filesystem direc‐
218 tory tree and was reconnected to the lost+found directory. The
219 inode is assigned the name of its inode number (ino). If a
220 lost+found directory does not exist, it is automatically cre‐
221 ated.
222 disconnected dir inode ino, moving to lost+found
224 As above only the inode is a directory inode. If a directory
226 inode is attached to lost+found, all of its children (if any)
227 stay attached to the directory and therefore get automatically
228 reconnected when the directory is reconnected.
229 imap claims in-use inode ino is free, correcting imap
231 The inode allocation map thinks that inode ino is free whereas
233 examination of the inode indicates that the inode may be in use
234 (although it may be disconnected). The program updates the
235 inode allocation map.
236 imap claims free inode ino is in use, correcting imap
238 The inode allocation map thinks that inode ino is in use whereas
240 examination of the inode indicates that the inode is not in use
241 and therefore is free. The program updates the inode allocation
242 map.
243 resetting inode ino nlinks from x to y
245 The program detected a mismatch between the number of valid
247 directory entries referencing inode ino and the number of refer‐
248 ences recorded in the inode and corrected the the number in the
249 inode.
250 fork-type fork in ino ino claims used block bno
252 Inode ino claims a block bno that is used (claimed) by either
254 another inode or the filesystem itself for metadata storage. The
255 fork-type is either data or attr indicating whether the problem
256 lies in the portion of the inode that tracks regular data or the
257 portion of the inode that stores XFS attributes. If the inode
258 is a real-time (rt) inode, the message says so. Any inode that
259 claims blocks used by the filesystem is deleted. If two or more
260 inodes claim the same block, they are both deleted.
261 fork-type fork in ino ino claims dup extent ...
263 Inode ino claims a block in an extent known to be claimed more
265 than once. The offset in the inode, start and length of the
266 extent is given. The message is slightly different if the inode
267 is a real-time (rt) inode and the extent is therefore a real-
268 time (rt) extent.
269 inode ino - bad extent ...
271 An extent record in the blockmap of inode ino claims blocks that
273 are out of the legal range of the filesystem. The message sup‐
274 plies the start, end, and file offset of the extent. The mes‐
275 sage is slightly different if the extent is a real-time (rt)
276 extent.
277 bad fork-type fork in inode ino
279 There was something structurally wrong or inconsistent with the
281 data structures that map offsets to filesystem blocks.
282 cleared inode ino
284 There was something wrong with the inode that was uncorrectable
286 so the program freed the inode. This usually happens because
287 the inode claims blocks that are used by something else or the
288 inode itself is badly corrupted. Typically, this message is pre‐
289 ceded by one or more messages indicating why the inode needed to
290 be cleared.
291 bad attribute fork in inode ino, clearing attr fork
293 There was something wrong with the portion of the inode that
295 stores XFS attributes (the attribute fork) so the program reset
296 the attribute fork. As a result of this, all attributes on that
297 inode are lost.
298 correcting nextents for inode ino, was x - counted y
300 The program found that the number of extents used to store the
302 data in the inode is wrong and corrected the number. The mes‐
303 sage refers to nextents if the count is wrong on the number of
304 extents used to store attribute information.
305 entry name in dir dir_ino not consistent with .. value (xxxx) in dir
307 ino ino, junking entry name in directory inode dir_ino
308 The entry name in directory inode dir_ino references a directory
310 inode ino. However, the .. entry in directory ino does not
311 point back to directory dir_ino, so the program deletes the
312 entry name in directory inode dir_ino. If the directory inode
313 ino winds up becoming a disconnected inode as a result of this,
314 it is moved to lost+found later.
315 entry name in dir dir_ino references already connected dir ino ino,
317 junking entry name in directory inode dir_ino
318 The entry name in directory inode dir_ino points to a directory
320 inode ino that is known to be a child of another directory.
321 Therefore, the entry is invalid and is deleted. This message
322 refers to an entry in a small directory. If this were a large
323 directory, the last phrase would read "will clear entry".
324 entry references free inode ino in directory dir_ino, will clear entry
326 An entry in directory inode dir_ino references an inode ino that
328 is known to be free. The entry is therefore invalid and is
329 deleted. This message refers to a large directory. If the
330 directory were small, the message would read "junking entry
331 ...".
332
334 xfs_repair -n (no modify mode) will return a status of 1 if filesystem
335 corruption was detected and 0 if no filesystem corruption was detected.
336 xfs_repair run without the -n option will always return a status code
337 of 0 if it completes without problems, unless the flag -e is used. If
338 it is used, then status 4 is reported when any issue with the filesys‐
339 tem was found, but could be fixed. If a runtime error is encountered
340 during operation, it will return a status of 1. In this case,
341 xfs_repair should be restarted. If xfs_repair is unable to proceed due
342 to a dirty log, it will return a status of 2. See below.
343
345 Due to the design of the XFS log, a dirty log can only be replayed by
346 the kernel, on a machine having the same CPU architecture as the
347 machine which was writing to the log. xfs_repair cannot replay a dirty
348 log and will exit with a status code of 2 when it detects a dirty log.
349 In this situation, the log can be replayed by mounting and immediately
351 unmounting the filesystem on the same class of machine that crashed.
352 Please make sure that the machine's hardware is reliable before replay‐
353 ing to avoid compounding the problems.
354 If mounting fails, the log can be erased by running xfs_repair with the
356 -L option. All metadata updates in progress at the time of the crash
357 will be lost, which may cause significant filesystem damage. This
358 should only be used as a last resort.
359
361 The filesystem to be checked and repaired must have been unmounted
362 cleanly using normal system administration procedures (the umount(8)
363 command or system shutdown), not as a result of a crash or system
364 reset. If the filesystem has not been unmounted cleanly, mount it and
365 unmount it cleanly before running xfs_repair.
366 xfs_repair does not do a thorough job on XFS extended attributes. The
368 structure of the attribute fork will be consistent, but only the con‐
369 tents of attribute forks that will fit into an inode are checked. This
370 limitation will be fixed in the future.
371 The no-modify mode (-n option) is not completely accurate. It does not
373 catch inconsistencies in the freespace and inode maps, particularly
374 lost blocks or subtly corrupted maps (trees).
375 The no-modify mode can generate repeated warnings about the same prob‐
377 lems because it cannot fix the problems as they are encountered.
378 If a filesystem fails to be repaired, a metadump image can be generated
380 with xfs_metadump(8) and be sent to an XFS maintainer to be analysed
381 and xfs_repair fixed and/or improved.
382
384 dd(1), mkfs.xfs(8), umount(8), xfs_admin(8), xfs_metadump(8), xfs(5).
385 xfs_repair(8)