1xfs_fsr(8) System Manager's Manual xfs_fsr(8)
2
3
4
6 xfs_fsr - filesystem reorganizer for XFS
7
9 xfs_fsr [-v] [-t seconds] [-f leftoff] [-m mtab]
10 xfs_fsr [-v] [xfsdev | file] ...
11
13 xfs_fsr is applicable only to XFS filesystems.
14
15 xfs_fsr improves the organization of mounted filesystems. The reorga‐
16 nization algorithm operates on one file at a time, compacting or other‐
17 wise improving the layout of the file extents (contiguous blocks of
18 file data).
19
20 The following options are accepted by xfs_fsr. The -m, -t, and -f
21 options have no meaning if any filesystems or files are specified on
22 the command line.
23
24 -m mtab Use this file for the list of filesystems to reorganize.
25 The default is to use /etc/mtab.
26
27 -t seconds How long to reorganize. The default is 7200 (2 hours).
28
29 -f leftoff Use this file instead of /var/tmp/.fsrlast to read the
30 state of where to start and as the file to store the state
31 of where reorganization left off.
32
33 -v Verbose. Print cryptic information about each file being
34 reorganized.
35
36 When invoked with no arguments xfs_fsr reorganizes all regular files in
37 all mounted filesystems. xfs_fsr makes many cycles over /etc/mtab each
38 time making a single pass over each XFS filesystem. Each pass goes
39 through and selects files that have the largest number of extents. It
40 attempts to defragment the top 10% of these files on each pass.
41
42 It runs for up to two hours after which it records the filesystem where
43 it left off, so it can start there the next time. This information is
44 stored in the file /var/tmp/.fsrlast_xfs. If the information found
45 here is somehow inconsistent or out of date it is ignored and reorgani‐
46 zation starts at the beginning of the first filesystem found in
47 /etc/mtab.
48
49 xfs_fsr can be called with one or more arguments naming filesystems
50 (block device name), and files to reorganize. In this mode xfs_fsr
51 does not read or write /var/tmp/.fsrlast_xfs nor does it run for a
52 fixed time interval. It makes one pass through each specified regular
53 file and all regular files in each specified filesystem. A command
54 line name referring to a symbolic link (except to a file system
55 device), FIFO, or UNIX domain socket generates a warning message, but
56 is otherwise ignored. While traversing the filesystem these types of
57 files are silently skipped.
58
60 /etc/mtab contains default list of filesystems to reorga‐
61 nize.
62 /var/tmp/.fsrlast_xfs
63 records the state where reorganization left off.
64
66 xfs_fsr(8), mkfs.xfs(8), xfs_ncheck(8), xfs(5).
67
69 xfs_fsr improves the layout of extents for each file by copying the
70 entire file to a temporary location and then interchanging the data
71 extents of the target and temporary files in an atomic manner. This
72 method requires that enough free disk space be available to copy any
73 given file and that the space be less fragmented than the original
74 file. It also requires the owner of the file to have enough remaining
75 filespace quota to do the copy on systems running quotas. xfs_fsr gen‐
76 erates a warning message if space is not sufficient to improve the tar‐
77 get file.
78
79 A temporary file used in improving a file given on the command line is
80 created in the same parent directory of the target file and is prefixed
81 by the string '.fsr'. The temporary files used in improving an entire
82 XFS device are stored in a directory at the root of the target device
83 and use the same naming scheme. The temporary files are unlinked upon
84 creation so data will not be readable by any other process.
85
86 xfs_fsr does not operate on files that are currently mapped in memory.
87 A 'file busy' error can be seen for these files if the verbose flag
88 (-v) is set.
89
90 Files marked as no-defrag will be skipped. The xfs_io(8) chattr command
91 with the f attribute can be used to set or clear this flag. Files and
92 directories created in a directory with the no-defrag flag will inherit
93 the attribute.
94
95 An entry in /etc/mtab or the file specified using the -m option must
96 have the rw option specified for read and write access. If this option
97 is not present, then xfs_fsr skips the filesystem described by that
98 line. See the fstab(5) reference page for more details.
99
100 In general we do not foresee the need to run xfs_fsr on system parti‐
101 tions such as /, /boot and /usr as in general these will not suffer
102 from fragmentation. There are also issues with defragmenting files
103 lilo(8) uses to boot your system. It is recommended that these files
104 should be flagged as no-defrag with the xfs_io(8) chattr command.
105 Should these files be moved by xfs_fsr then you must rerun lilo before
106 you reboot or you may have an unbootable system.
107
108
109
110 xfs_fsr(8)