xfs_copy(8)

1xfs_copy(8)                 System Manager's Manual                xfs_copy(8)
2
3
4

NAME

6       xfs_copy - copy the contents of an XFS filesystem
7

SYNOPSIS

9       xfs_copy [ -bd ] [ -L log ] source target1 [ target2 ... ]
10       xfs_copy -V
11

DESCRIPTION

13       xfs_copy  copies  an  XFS filesystem to one or more targets in parallel
14       (see xfs(5)).  The first (source) argument must be the pathname of  the
15       device  or  file containing the XFS filesystem. The remaining arguments
16       specify one or more target devices or  file  names.  If  the  pathnames
17       specify devices, a copy of the source XFS filesystem is created on each
18       device. The target can also be the name of a  regular  file,  in  which
19       case  an image of the source XFS filesystem is created in that file. If
20       the file does not exist, xfs_copy creates the file. The length  of  the
21       resulting  file is equal to the size of the source filesystem. However,
22       if the file is created on an XFS filesystem, the file consumes  roughly
23       the  amount  of  space  actually  used  in the source filesystem by the
24       filesystem and the XFS log.  The space saving is because xfs_copy seeks
25       over  free  blocks  instead of copying them and the XFS filesystem sup‐
26       ports sparse files efficiently.
27
28       xfs_copy should only be used to copy unmounted  filesystems,  read-only
29       mounted filesystems, or frozen filesystems (see xfs_freeze(8)).  Other‐
30       wise, the generated filesystem(s) would be inconsistent or corrupt.
31
32       xfs_copy does not alter the source filesystem  in  any  way.  Each  new
33       (target) filesystem is identical to the original filesystem except that
34       new filesystems each have a new unique  filesystem  identifier  (UUID).
35       Therefore, if both the old and new filesystems will be used as separate
36       distinct filesystems, xfs_copy or  xfsdump(8)/xfsrestore(8)  should  be
37       used  to  generate the new filesystem(s) instead of dd(1) or other pro‐
38       grams that do block-by-block disk copying.
39
40       xfs_copy uses synchronous  writes  to  ensure  that  write  errors  are
41       detected.
42
43       xfs_copy  uses  pthreads(7)  to  perform  simultaneous parallel writes.
44       xfs_copy creates one additional thread for each target to  be  written.
45       All threads die if xfs_copy terminates or aborts.
46

OPTIONS

48       -d     Create  a duplicate (true clone) filesystem. This should be done
49              only if the new filesystem will be used as a replacement for the
50              original filesystem (such as in the case of disk replacement).
51
52       -b     The  buffered  option  can  be  used  to ensure direct IO is not
53              attempted to any of the target files. This is  useful  when  the
54              filesystem holding the target file does not support direct IO.
55
56       -L log Specifies  the  location  of  the log if the default location of
57              /var/tmp/xfs_copy.log.XXXXXX is not desired.
58
59       -V     Prints the version number and exits.
60

DIAGNOSTICS

62       xfs_copy reports errors to both stderr and in more detailed form  to  a
63       generated     log     file     whose    name    is    of    the    form
64       /var/tmp/xfs_copy.log.XXXXXX or a log file specified by the -L  option.
65       If  xfs_copy  detects  a  write error on a target, the copy of that one
66       target is aborted and an error message is issued to both stderr and the
67       log  file,  but  the  rest of the copies continue. When xfs_copy termi‐
68       nates, all aborted targets are reported to  both  stderr  and  the  log
69       file.
70
71       If  all  targets  abort  or  if  there  is  an error reading the source
72       filesystem, xfs_copy immediately aborts.
73
74       xfs_copy returns an exit code of 0  if  all  targets  are  successfully
75       copied and an exit code of 1 if any target fails.
76

NOTES

78       When  moving  filesystems  from  one  disk  to another, if the original
79       filesystem is significantly smaller than the new filesystem,  and  will
80       be  made  larger,  we  recommend that mkfs.xfs(8) and xfsdump(8)/xfsre‐
81       store(8) be used instead of  using  xfs_copy  and  xfs_growfs(8).   The
82       filesystem  layout  resulting  from using xfs_copy/xfs_growfs is almost
83       always worse than the result of using  mkfs.xfs/xfsdump/xfsrestore  but
84       in  the  case of small filesystems, the differences can have a signifi‐
85       cant performance impact. This is due to the  way  xfs_growfs(8)  works,
86       and not due to any shortcoming in xfs_copy itself.
87

CAVEATS

89       xfs_copy does not copy XFS filesystems that have a real-time section or
90       XFS filesystems with external logs. In both cases, xfs_copy aborts with
91       an error message.
92