1xfs_repair(8) System Manager's Manual xfs_repair(8)
2
6 xfs_repair - repair an XFS filesystem
7
9 xfs_repair [ -dfLnPv ] [ -o subopt[=value] ] [ -t interval ] [ -l
10 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.
35 -l logdev
37 Specifies the device special file where the filesystem's exter‐
38 nal log resides. Only for those filesystems which use an exter‐
39 nal log. See the mkfs.xfs -l option, and refer to xfs(5) for a
40 detailed description of the XFS log.
41 -r rtdev
43 Specifies the device special file where the filesystem's real‐
44 time section resides. Only for those filesystems which use a
45 realtime section. See the mkfs.xfs -r option, and refer to
46 xfs(5) for a detailed description of the XFS realtime section.
47 -n No modify mode. Specifies that xfs_repair should not modify the
49 filesystem but should only scan the filesystem and indicate what
50 repairs would have been made.
51 -o subopt[=value]
53 Override what the program might conclude about the filesystem if
54 left to its own devices.
55 The suboptions supported are:
57 ihash=ihashsize
59 overrides the default inode cache hash size. The total
60 number of inode cache entries are limited to 8 times
61 this amount. The default ihashsize is 1024 (for a
62 total of 8192 entries).
63 bhash=bhashsize
65 overrides the default buffer cache hash size. The
66 total number of buffer cache entries are limited to 8
67 times this amount. The default size is set to use up
68 the remainder of 75% of the system's physical RAM.
69 size
70 ag_stride=ags_per_concat_unit
72 This creates additional processing threads to parallel
73 process AGs that span multiple concat units. This can
74 significantly reduce repair times on concat based
75 filesystems.
76 -t interval
78 Modify reporting interval. During long runs xfs_repair outputs
79 its progress every 15 minutes. Reporting is only activated when
80 ag_stride is enabled.
81 -P Disable prefetching of inode and directory blocks.
83 -v Verbose output.
85 -d Repair dangerously. Allow xfs_repair to repair an XFS filesystem
87 mounted read only. This is typically done on a root fileystem
88 from single user mode, immediately followed by a reboot.
89 -V Prints out the current version number and exits.
91 Checks Performed
93 Inconsistencies corrected include the following:
94 1. Inode and inode blockmap (addressing) checks: bad magic number
96 in inode, bad magic numbers in inode blockmap blocks, extents
97 out of order, incorrect number of records in inode blockmap
98 blocks, blocks claimed that are not in a legal data area of the
99 filesystem, blocks that are claimed by more than one inode.
100 2. Inode allocation map checks: bad magic number in inode map
102 blocks, inode state as indicated by map (free or in-use) incon‐
103 sistent with state indicated by the inode, inodes referenced by
104 the filesystem that do not appear in the inode allocation map,
105 inode allocation map referencing blocks that do not appear to
106 contain inodes.
107 3. Size checks: number of blocks claimed by inode inconsistent with
109 inode size, directory size not block aligned, inode size not
110 consistent with inode format.
111 4. Directory checks: bad magic numbers in directory blocks, incor‐
113 rect number of entries in a directory block, bad freespace
114 information in a directory leaf block, entry pointing to an
115 unallocated (free) or out of range inode, overlapping entries,
116 missing or incorrect dot and dotdot entries, entries out of
117 hashvalue order, incorrect internal directory pointers, direc‐
118 tory type not consistent with inode format and size.
119 5. Pathname checks: files or directories not referenced by a path‐
121 name starting from the filesystem root, illegal pathname compo‐
122 nents.
123 6. Link count checks: link counts that do not agree with the number
125 of directory references to the inode.
126 7. Freemap checks: blocks claimed free by the freemap but also
128 claimed by an inode, blocks unclaimed by any inode but not
129 appearing in the freemap.
130 8. Super Block checks: total free block and/or free i-node count
132 incorrect, filesystem geometry inconsistent, secondary and pri‐
133 mary superblocks contradictory.
134 Orphaned files and directories (allocated, in-use but unreferenced) are
136 reconnected by placing them in the lost+found directory. The name
137 assigned is the inode number.
138 Disk Errors
140 xfs_repair aborts on most disk I/O errors. Therefore, if you are trying
141 to repair a filesystem that was damaged due to a disk drive failure,
142 steps should be taken to ensure that all blocks in the filesystem are
143 readable and writeable before attempting to use xfs_repair to repair
144 the filesystem. A possible method is using dd(8) to copy the data onto
145 a good disk.
146 lost+found
148 The directory lost+found does not have to already exist in the filesys‐
149 tem being repaired. If the directory does not exist, it is automati‐
150 cally created if required. If it already exists, it will be checked
151 for consistency and if valid will be used for additional orphaned
152 files. Invalid lost+found directories are removed and recreated. Exist‐
153 ing files in a valid lost+found are not removed or renamed.
154 Corrupted Superblocks
156 XFS has both primary and secondary superblocks. xfs_repair uses infor‐
157 mation in the primary superblock to automatically find and validate the
158 primary superblock against the secondary superblocks before proceeding.
159 Should the primary be too corrupted to be useful in locating the sec‐
160 ondary superblocks, the program scans the filesystem until it finds and
161 validates some secondary superblocks. At that point, it generates a
162 primary superblock.
163 Quotas
165 If quotas are in use, it is possible that xfs_repair will clear some or
166 all of the filesystem quota information. If so, the program issues a
167 warning just before it terminates. If all quota information is lost,
168 quotas are disabled and the program issues a warning to that effect.
169 Note that xfs_repair does not check the validity of quota limits. It is
171 recommended that you check the quota limit information manually after
172 xfs_repair. Also, space usage information is automatically regenerated
173 the next time the filesystem is mounted with quotas turned on, so the
174 next quota mount of the filesystem may take some time.
175
177 xfs_repair issues informative messages as it proceeds indicating what
178 it has found that is abnormal or any corrective action that it has
179 taken. Most of the messages are completely understandable only to
180 those who are knowledgeable about the structure of the filesystem.
181 Some of the more common messages are explained here. Note that the
182 language of the messages is slightly different if xfs_repair is run in
183 no-modify mode because the program is not changing anything on disk.
184 No-modify mode indicates what it would do to repair the filesystem if
185 run without the no-modify flag.
186 disconnected inode ino, moving to lost+found
188 An inode numbered ino was not connected to the filesystem direc‐
190 tory tree and was reconnected to the lost+found directory. The
191 inode is assigned the name of its inode number (ino). If a
192 lost+found directory does not exist, it is automatically cre‐
193 ated.
194 disconnected dir inode ino, moving to lost+found
196 As above only the inode is a directory inode. If a directory
198 inode is attached to lost+found, all of its children (if any)
199 stay attached to the directory and therefore get automatically
200 reconnected when the directory is reconnected.
201 imap claims in-use inode ino is free, correcting imap
203 The inode allocation map thinks that inode ino is free whereas
205 examination of the inode indicates that the inode may be in use
206 (although it may be disconnected). The program updates the
207 inode allocation map.
208 imap claims free inode ino is in use, correcting imap
210 The inode allocation map thinks that inode ino is in use whereas
212 examination of the inode indicates that the inode is not in use
213 and therefore is free. The program updates the inode allocation
214 map.
215 resetting inode ino nlinks from x to y
217 The program detected a mismatch between the number of valid
219 directory entries referencing inode ino and the number of refer‐
220 ences recorded in the inode and corrected the the number in the
221 inode.
222 fork-type fork in ino ino claims used block bno
224 Inode ino claims a block bno that is used (claimed) by either
226 another inode or the filesystem itself for metadata storage. The
227 fork-type is either data or attr indicating whether the problem
228 lies in the portion of the inode that tracks regular data or the
229 portion of the inode that stores XFS attributes. If the inode
230 is a real-time (rt) inode, the message says so. Any inode that
231 claims blocks used by the filesystem is deleted. If two or more
232 inodes claim the same block, they are both deleted.
233 fork-type fork in ino ino claims dup extent ...
235 Inode ino claims a block in an extent known to be claimed more
237 than once. The offset in the inode, start and length of the
238 extent is given. The message is slightly different if the inode
239 is a real-time (rt) inode and the extent is therefore a real-
240 time (rt) extent.
241 inode ino - bad extent ...
243 An extent record in the blockmap of inode ino claims blocks that
245 are out of the legal range of the filesystem. The message sup‐
246 plies the start, end, and file offset of the extent. The mes‐
247 sage is slightly different if the extent is a real-time (rt)
248 extent.
249 bad fork-type fork in inode ino
251 There was something structurally wrong or inconsistent with the
253 data structures that map offsets to filesystem blocks.
254 cleared inode ino
256 There was something wrong with the inode that was uncorrectable
258 so the program freed the inode. This usually happens because
259 the inode claims blocks that are used by something else or the
260 inode itself is badly corrupted. Typically, this message is pre‐
261 ceded by one or more messages indicating why the inode needed to
262 be cleared.
263 bad attribute fork in inode ino, clearing attr fork
265 There was something wrong with the portion of the inode that
267 stores XFS attributes (the attribute fork) so the program reset
268 the attribute fork. As a result of this, all attributes on that
269 inode are lost.
270 correcting nextents for inode ino, was x - counted y
272 The program found that the number of extents used to store the
274 data in the inode is wrong and corrected the number. The mes‐
275 sage refers to nextents if the count is wrong on the number of
276 extents used to store attribute information.
277 entry name in dir dir_ino not consistent with .. value (xxxx) in dir
279 ino ino, junking entry name in directory inode dir_ino
280 The entry name in directory inode dir_ino references a directory
282 inode ino. However, the .. entry in directory ino does not
283 point back to directory dir_ino, so the program deletes the
284 entry name in directory inode dir_ino. If the directory inode
285 ino winds up becoming a disconnected inode as a result of this,
286 it is moved to lost+found later.
287 entry name in dir dir_ino references already connected dir ino ino,
289 junking entry name in directory inode dir_ino
290 The entry name in directory inode dir_ino points to a directory
292 inode ino that is known to be a child of another directory.
293 Therefore, the entry is invalid and is deleted. This message
294 refers to an entry in a small directory. If this were a large
295 directory, the last phrase would read "will clear entry".
296 entry references free inode ino in directory dir_ino, will clear entry
298 An entry in directory inode dir_ino references an inode ino that
300 is known to be free. The entry is therefore invalid and is
301 deleted. This message refers to a large directory. If the
302 directory were small, the message would read "junking entry
303 ...".
304
306 xfs_repair -n (no modify node) will return a status of 1 if filesystem
307 corruption was detected and 0 if no filesystem corruption was detected.
308 xfs_repair run without the -n option will always return a status code
309 of 0.
310
312 The filesystem to be checked and repaired must have been unmounted
313 cleanly using normal system administration procedures (the umount(8)
314 command or system shutdown), not as a result of a crash or system
315 reset. If the filesystem has not been unmounted cleanly, mount it and
316 unmount it cleanly before running xfs_repair.
317 xfs_repair does not do a thorough job on XFS extended attributes. The
319 structure of the attribute fork will be consistent, but only the con‐
320 tents of attribute forks that will fit into an inode are checked. This
321 limitation will be fixed in the future.
322 The no-modify mode (-n option) is not completely accurate. It does not
324 catch inconsistencies in the freespace and inode maps, particularly
325 lost blocks or subtly corrupted maps (trees).
326 The no-modify mode can generate repeated warnings about the same prob‐
328 lems because it cannot fix the problems as they are encountered.
329 If a filesystem fails to be repaired, a metadump image can be generated
331 with xfs_metadump(8) and be sent to an XFS maintainer to be analysed
332 and xfs_repair fixed and/or improved.
333
335 dd(1), mkfs.xfs(8), umount(8), xfs_check(8), xfs_metadump(8), xfs(5).
336 xfs_repair(8)