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

DESCRIPTION

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

OPTIONS

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

DIAGNOSTICS

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

NOTES

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

CAVEATS

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