1 <?xml version="1.0" encoding="utf-8"?>
2 <chapter xmlns="http://docbook.org/ns/docbook"
3 xmlns:xl="http://www.w3.org/1999/xlink" version="5.0" xml:lang="en-US"
5 xmlns:xi="http://www.w3.org/2001/XInclude">
6 <title xml:id="userutilities.title">User Utilities</title>
7 <para>This chapter describes user utilities.</para>
8 <section xml:id="dbdoclet.50438206_94597">
11 <primary>lfs</primary>
13 <literal>lfs</literal>
16 <literal>lfs</literal> utility can be used for user configuration routines
17 and monitoring.</para>
19 <title>Synopsis</title>
22 lfs changelog [--follow] <replaceable>mdt_name</replaceable> [startrec [endrec]]
23 lfs changelog_clear <replaceable>mdt_name id endrec</replaceable>
24 lfs check <replaceable>mds|osts|servers</replaceable>
25 lfs data_version [-nrw] <replaceable>filename</replaceable>
26 lfs df [-i] [-h] [--pool]-p <replaceable>fsname</replaceable>[.<replaceable>pool</replaceable>] [<replaceable>path</replaceable>] [--lazy]
27 lfs find [[!] --atime|-A [-+]N] [[!] --mtime|-M [-+]N]
28 [[!] --ctime|-C [-+]N] [--maxdepth|-D N] [--name|-n <replaceable>pattern</replaceable>]
29 [--print|-p] [--print0|-P] [[!] --obd|-O <replaceable>ost_name</replaceable>[,<replaceable>ost_name...</replaceable>]]
30 [[!] --size|-S [+-]N[kMGTPE]] --type |-t {bcdflpsD}]
31 [[!] --gid|-g|--group|-G <replaceable>gname|gid</replaceable>]
32 [[!] --uid|-u|--user|-U <replaceable>uname|uid</replaceable>]
33 <replaceable>dirname|filename</replaceable>
34 lfs getname [-h]|[path...]
35 lfs getstripe [--obd|-O <replaceable>ost_name</replaceable>] [--quiet|-q] [--verbose|-v]
36 [--stripe-count|-c] [--stripe-index|-i]
37 [--stripe-size|-s] [--pool|-p] [--directory|-d]
38 [--mdt-index|-M] [--recursive|-r] [--raw|-R]
40 <replaceable>dirname|filename</replaceable> ...
41 lfs setstripe [--size|-s stripe_size] [--count|-c <replaceable>stripe_count</replaceable>]
42 [--stripe-index|-i <replaceable>start_ost_index</replaceable>]
43 [--ost-list|-o <replaceable>ost_indicies</replaceable>]
44 [--pool|-p <replaceable>pool</replaceable>]
45 <replaceable>dirname|filename</replaceable>
46 lfs setstripe -d <replaceable>dir</replaceable>
48 lfs pool_list <replaceable>filesystem</replaceable>[.<replaceable>pool</replaceable>]| <replaceable>pathname</replaceable>
49 lfs quota [-q] [-v] [-h] [-o <replaceable>obd_uuid</replaceable>|-I <replaceable>ost_idx</replaceable>|-i <replaceable>mdt_idx</replaceable>]
50 [-u <replaceable>username|uid|-g</replaceable> <replaceable>group|gid</replaceable>|-p <replaceable>projid</replaceable>] <replaceable>/mount_point</replaceable>
51 lfs quota -t -u|-g|-p <replaceable>/mount_point</replaceable>
52 lfs quotacheck [-ug] <replaceable>/mount_point</replaceable>
53 lfs quotachown [-i] <replaceable>/mount_point</replaceable>
54 lfs quotainv [-ug] [-f] <replaceable>/mount_point</replaceable>
55 lfs quotaon [-ugf] <replaceable>/mount_point</replaceable>
56 lfs quotaoff [-ug] <replaceable>/mount_point</replaceable>
57 lfs setquota {-u|--user|-g|--group|-p|--project} <replaceable>uname|uid|gname|gid|projid</replaceable>
58 [--block-softlimit <replaceable>block_softlimit</replaceable>]
59 [--block-hardlimit <replaceable>block_hardlimit</replaceable>]
60 [--inode-softlimit <replaceable>inode_softlimit</replaceable>]
61 [--inode-hardlimit <replaceable>inode_hardlimit</replaceable>]
62 <replaceable>/mount_point</replaceable>
63 lfs setquota -u|--user|-g|--group|-p|--project <replaceable>uname|uid|gname|gid|projid</replaceable>
64 [-b <replaceable>block_softlimit</replaceable>] [-B <replaceable>block_hardlimit</replaceable>]
65 [-i <replaceable>inode-softlimit</replaceable>] [-I <replaceable>inode_hardlimit</replaceable>]
66 <replaceable>/mount_point</replaceable>
67 lfs setquota -t -u|-g|-p [--block-grace <replaceable>block_grace</replaceable>]
68 [--inode-grace <replaceable>inode_grace</replaceable>]
69 <replaceable>/mount_point</replaceable>
70 lfs setquota -t -u|-g|-p [-b <replaceable>block_grace</replaceable>] [-i <replaceable>inode_grace</replaceable>]
71 <replaceable>/mount_point</replaceable>
75 <para>In the above example, the
77 <replaceable>/mount_point</replaceable>
78 </literal> parameter refers to the mount point of the Lustre file
82 <para>The old lfs quota output was very detailed and contained
83 cluster-wide quota statistics (including cluster-wide limits for a
84 user/group and cluster-wide usage for a user/group), as well as
85 statistics for each MDS/OST. Now,
86 <literal>lfs quota</literal> has been updated to provide only
87 cluster-wide statistics, by default. To obtain the full report of
88 cluster-wide limits, usage and statistics, use the
89 <literal>-v</literal> option with
90 <literal>lfs quota</literal>.</para>
94 <title>Description</title>
96 <literal>lfs</literal> utility is used to create a new file with a
97 specific striping pattern, determine the default striping pattern, gather
98 the extended attributes (object numbers and location) for a specific
99 file, find files with specific attributes, list OST information or set
100 quota limits. It can be invoked interactively without any arguments or in
101 a non-interactive mode with one of the supported arguments.</para>
104 <title>Options</title>
106 <literal>lfs</literal> options are listed and described below. For a
107 complete list of available options, type help at the
108 <literal>lfs</literal> prompt.</para>
109 <informaltable frame="all">
111 <colspec colname="c1" colwidth="20*" />
112 <colspec colname="c2" colwidth="30*" />
113 <colspec colname="c3" colwidth="50*" />
116 <entry nameend="c2" namest="c1">
118 <emphasis role="bold">Option</emphasis>
123 <emphasis role="bold">Description</emphasis>
130 <entry nameend="c2" namest="c1">
132 <literal>changelog</literal>
136 <para>Shows the metadata changes on an MDT. Start and end
137 points are optional. The
138 <literal>--follow</literal> option blocks on new changes; this
139 option is only valid when run directly on the MDT node.</para>
143 <entry nameend="c2" namest="c1">
145 <literal>changelog_clear</literal>
149 <para>Indicates that changelog records previous to
151 <replaceable>endrec</replaceable>
152 </literal> are no longer of interest to a particular consumer
154 <replaceable>id</replaceable>
155 </literal>, potentially allowing the MDT to free up disk space.
158 <replaceable>endrec</replaceable>
159 </literal> of 0 indicates the current last record. Changelog
160 consumers must be registered on the MDT node using
161 <literal>lctl</literal>.</para>
165 <entry nameend="c2" namest="c1">
166 <literal>check</literal>
169 <para>Displays the status of MDS or OSTs (as specified in the
170 command) or all servers (MDS and OSTs).</para>
174 <entry nameend="c2" namest="c1">
176 <literal>data_version [-nrw]
177 <replaceable>filename</replaceable></literal>
181 <para>Displays the current version of file data. If
182 <literal>-n</literal> is specified, the data version is read
183 without taking a lock. As a consequence, the data version could
184 be outdated if there are dirty caches on filesystem clients, but
185 this option will not force data flushes and has less of an
186 impact on the filesystem. If <literal>-r</literal> is specified,
187 the data version is read after dirty pages on clients are
188 flushed. If <literal>-w</literal> is specified, the data version
189 is read after all caching pages on clients are flushed.
192 Even with <literal>-r</literal> or <literal>-w</literal>, race
193 conditions are possible and the data version should be checked
194 before and after an operation to be confident the data did not
198 The data version is the sum of the last committed transaction
199 numbers of all data objects of a file. It is used by HSM policy
200 engines for verifying that file data has not been changed during
201 an archive operation or before a release operation, and by OST
202 migration, primarily for verifying that file data has not been
203 changed during a data copy, when done in non-blocking mode.
208 <entry nameend="c2" namest="c1">
209 <literal>df [-i] [-h] [--pool|-p
210 <replaceable>fsname</replaceable>[.
211 <replaceable>pool</replaceable>] [
212 <replaceable>path</replaceable>] [--lazy]</literal>
216 <literal>-i</literal> to report file system disk space usage or
217 inode usage of each MDT or OST or, if a pool is specified with
219 <literal>-p</literal> option, a subset of OSTs.</para>
220 <para>By default, the usage of all mounted Lustre file systems
222 <literal>path</literal> option is included, only the usage for
223 the specified file system is reported. If the
224 <literal>-h</literal> option is included, the output is printed
225 in human-readable format, using SI base-2 suffixes for
226 <emphasis role="bold">M</emphasis>ega-,
227 <emphasis role="bold">G</emphasis>iga-,
228 <emphasis role="bold">T</emphasis>era-,
229 <emphasis role="bold">P</emphasis>eta-, or
230 <emphasis role="bold">E</emphasis>xabytes.</para>
232 <literal>--lazy</literal> option is specified, any OST that is
233 currently disconnected from the client will be skipped. Using
235 <literal>--lazy</literal> option prevents the
236 <literal>df</literal> output from being blocked when an OST is
237 offline. Only the space on the OSTs that can currently be
238 accessed are returned. The
239 <literal>llite.*.lazystatfs</literal> tunable can be enabled to
240 make this the default behaviour for all
241 <literal>statfs()</literal> operations.</para>
245 <entry nameend="c2" namest="c1">
247 <literal>find</literal>
251 <para>Searches the directory tree rooted at the given
252 directory/filename for files that match the given
255 <literal>!</literal> before an option negates its meaning (files
256 NOT matching the parameter). Using
257 <literal>+</literal> before a numeric value means files with the
258 parameter OR MORE. Using
259 <literal>-</literal> before a numeric value means files with the
260 parameter OR LESS.</para>
266 <literal>--atime</literal>
269 <para>File was last accessed N*24 hours ago. (There is no
271 <literal>atime</literal> is kept coherent across the
273 <para>OSTs store a transient
274 <literal>atime</literal> that is updated when clients do read
276 <literal>atime</literal> is written to the MDS when the file is
277 closed. However, on-disk atime is only updated if it is more
278 than 60 seconds old (
279 <literal>/proc/fs/lustre/mds/*/max_atime_diff</literal>). The
280 Lustre software considers the latest
281 <literal>atime</literal> from all OSTs. If a
282 <literal>setattr</literal> is set by user, then it is updated on
283 both the MDS and OST, allowing the
284 <literal>atime</literal> to go backward.</para>
293 <literal>--ctime</literal>
297 <para>File status was last changed N*24 hours ago.</para>
306 <literal>--mtime</literal>
310 <para>File data was last modified N*24 hours ago.</para>
319 <literal>--obd</literal>
323 <para>File has an object on a specific OST(s).</para>
332 <literal>--size</literal>
336 <para>File has a size in bytes, or kilo-, Mega-, Giga-, Tera-,
337 Peta- or Exabytes if a suffix is given.</para>
346 <literal>--type</literal>
350 <para>File has the type - block, character, directory, pipe,
351 file, symlink, socket or door (used in Solaris operating
361 <literal>--uid</literal>
365 <para>File has a specific numeric user ID.</para>
374 <literal>--user</literal>
378 <para>File owned by a specific user (numeric user ID
388 <literal>--gid</literal>
392 <para>File has a specific group ID.</para>
401 <literal>--group</literal>
405 <para>File belongs to a specific group (numeric group ID
415 <literal>-maxdepth</literal></para>
418 <para>Limits find to descend at most N levels of the directory
428 <literal>--print</literal>/
429 <literal>--print0</literal></para>
432 <para>Prints the full filename, followed by a new line or NULL
433 character correspondingly.</para>
437 <entry nameend="c2" namest="c1">
439 <literal>osts [path]</literal>
443 <para>Lists all OSTs for the file system. If a path located on
444 a mounted Lustre file system is specified, then only OSTs
445 belonging to this file system are displayed.</para>
449 <entry nameend="c2" namest="c1">
451 <literal>getname [path...]</literal>
455 <para>List each Lustre file system instance associated with
456 each Lustre mount point. If no path is specified, all Lustre
457 mount points are interrogated. If a list of paths is provided,
458 the instance of each path is provided. If the path is not a
459 Lustre instance 'No such device' is returned.</para>
463 <entry nameend="c2" namest="c1">
465 <literal>getstripe</literal>
469 <para>Lists striping information for a given filename or
470 directory. By default, the stripe count, stripe size and offset
472 <para>If you only want specific striping information, then the
474 <literal>--stripe-count</literal>,
475 <literal>--stripe-size</literal>,
476 <literal>--stripe-index</literal>,
477 <literal>--layout</literal>, or
478 <literal>--pool</literal> plus various combinations of these
479 options can be used to retrieve specific information.</para>
481 <literal>--raw</literal> option is specified, the stripe
482 information is printed without substituting the file system
483 default values for unspecified fields. If the striping EA is
484 not set, 0, 0, and -1 will be printed for the stripe count,
485 size, and offset respectively.</para>
486 <para condition="l24">The
487 <literal>--mdt-index</literal> prints the index of the MDT for a given
489 <xref linkend="dbdoclet.rmremotedir" />.</para>
499 <replaceable>ost_name</replaceable></literal>
503 <para>Lists files that have an object on a specific OST.</para>
512 <literal>--quiet</literal>
516 <para>Lists details about the file's object ID
526 <literal>--verbose</literal>
530 <para>Prints additional striping information.</para>
539 <literal>--count</literal>
543 <para>Lists the stripe count (how many OSTs to use).</para>
552 <literal>--index</literal>
556 <para>Lists the index for each OST in the file system.</para>
565 <literal>--offset</literal>
569 <para>Lists the OST index on which file striping starts.</para>
578 <literal>--pool</literal>
582 <para>Lists the pools to which a file belongs.</para>
591 <literal>--size</literal>
595 <para>Lists the stripe size (how much data to write to one OST
596 before moving to the next OST).</para>
605 <literal>--directory</literal>
609 <para>Lists entries about a specified directory instead of its
610 contents (in the same manner as
611 <literal>ls -d</literal>).</para>
620 <literal>--recursive</literal>
624 <para>Recurses into all sub-directories.</para>
628 <entry nameend="c2" namest="c1">
630 <literal>setstripe</literal>
634 <para>Create new files with a specific file layout (stripe
635 pattern) configuration.
637 <para>The file cannot exist prior to using
638 <literal>setstripe</literal>. A directory must exist prior to
640 <literal>setstripe</literal>.</para>
650 <literal>--count stripe_cnt</literal>
654 <para>Number of OSTs over which to stripe a file. A
655 <literal>stripe_cnt</literal> of 0 uses the file system-wide
656 default stripe count (default is 1). A
657 <literal>stripe_cnt</literal> of -1 stripes over all available
667 <literal>--size stripe_size</literal>
669 <para>The default stripe-size is 0. The default start-ost is
670 -1. Do NOT confuse them! If you set start-ost to 0, all new
671 file creations occur on OST 0 (seldom a good idea).</para>
672 </footnote> </para>
675 <para>Number of bytes to store on an OST before moving to the
676 next OST. A stripe_size of 0 uses the file system's default
677 stripe size, (default is 1 MB). Can be specified with
678 <emphasis role="bold">k</emphasis>(KB),
679 <emphasis role="bold">m</emphasis>(MB), or
680 <emphasis role="bold">g</emphasis>(GB), respectively.</para>
689 <literal>--stripe-index start_ost_index</literal>
693 <para>The OST index (base 10, starting at 0) on which to start
694 striping for this file. A start_ost_index value of -1 allows
695 the MDS to choose the starting index. This is the default
696 value, and it means that the MDS selects the starting OST as it
697 wants. We strongly recommend selecting this default, as it
698 allows space and load balancing to be done by the MDS as
700 <literal>start_ost_index</literal> value has no relevance on
701 whether the MDS will use round-robin or QoS weighted allocation
702 for the remaining stripes in the file.</para>
711 <literal>--ost-index ost_indices</literal>
715 <para>This option is used to specify the exact stripe
716 layout on the the file system. <literal>ost_indices</literal>
717 is a list of OSTs referenced by their indices and index ranges
718 separated by commas, e.g. <literal>1,2-4,7</literal>.</para>
728 <replaceable>pool</replaceable></literal>
732 <para>Name of the pre-defined pool of OSTs (see
733 <xref linkend="dbdoclet.50438219_38274" />) that will be used
735 <literal>stripe_cnt</literal>,
736 <literal>stripe_size</literal> and
737 <literal>start_ost</literal> values are used as well. The
738 start-ost value must be part of the pool or an error is
743 <entry nameend="c2" namest="c1">
745 <literal>setstripe -d</literal>
749 <para>Deletes default striping on the specified
754 <entry nameend="c2" namest="c1">
756 <literal>pool_list {filesystem}[.poolname]|{pathname}</literal>
760 <para>Lists pools in the file system or pathname, or OSTs in
761 the file system's pool.</para>
765 <entry nameend="c2" namest="c1">
767 <literal>quota [-q] [-v] [-o
768 <replaceable>obd_uuid</replaceable>|-i
769 <replaceable>mdt_idx</replaceable>|-I
770 <replaceable>ost_idx</replaceable>] [-u|-g|-p
771 <replaceable>uname|uid|gname|gid|projid]</replaceable>
772 <replaceable>/mount_point</replaceable></literal>
777 <para>Displays disk usage and limits, either for the full file
778 system or for objects on a specific OBD. A user or group name
779 or an usr, group and project ID can be specified. If all user,
780 group project ID are omitted, quotas for the current UID/GID
781 are shown. The <literal>-q</literal> option disables printing
782 of additional descriptions (including column titles). It fills
783 in blank spaces in the
784 <literal>grace</literal> column with zeros (when there is no
785 grace period set), to ensure that the number of columns is
787 <literal>-v</literal> option provides more verbose (per-OBD
788 statistics) output.</para>
792 <entry nameend="c2" namest="c1">
795 <replaceable>-u|-g|-p</replaceable>
796 <replaceable>/mount_point</replaceable></literal>
800 <para>Displays block and inode grace times for user (
801 <literal>-u</literal>) or group (
802 <literal>-g</literal>) or project (
803 <literal>-p</literal>) quotas.</para>
807 <entry nameend="c2" namest="c1">
809 <literal>quotachown</literal>
813 <para>Changes the file's owner and group on OSTs of the
814 specified file system.</para>
818 <entry nameend="c2" namest="c1">
820 <literal>quotacheck [-ugf]
821 <replaceable>/mount_point</replaceable></literal>
825 <para>Scans the specified file system for disk usage, and
826 creates or updates quota files. Options specify quota for users
828 <literal>-u</literal>), groups (
829 <literal>-g</literal>), and force (
830 <literal>-f</literal>).</para>
834 <entry nameend="c2" namest="c1">
836 <literal>quotaon [-ugf]
837 <replaceable>/mount_point</replaceable></literal>
841 <para>Turns on file system quotas. Options specify quota for
843 <literal>-u</literal>), groups (
844 <literal>-g</literal>), and force (
845 <literal>-f</literal>).</para>
849 <entry nameend="c2" namest="c1">
851 <literal>quotaoff [-ugf]
852 <replaceable>/mount_point</replaceable></literal>
856 <para>Turns off file system quotas. Options specify quota for
858 <literal>-u</literal>), groups (
859 <literal>-g</literal>), and force (
860 <literal>-f</literal>).</para>
864 <entry nameend="c2" namest="c1">
866 <literal>quotainv [-ug] [-f]
867 <replaceable>/mount_point</replaceable></literal>
871 <para>Clears quota files (administrative quota files if used
873 <literal>-f</literal>, operational quota files otherwise), all
874 of their quota entries for users (
875 <literal>-u</literal>) or groups (
876 <literal>-g</literal>). After running
877 <literal>quotainv</literal>, you must run
878 <literal>quotacheck</literal> before using quotas.</para>
880 <para>Use extreme caution when using this command; its
881 results cannot be undone.</para>
886 <entry nameend="c2" namest="c1">
888 <literal>setquota {-u|-g|-p
889 <replaceable>uname|uid|gname|gid|projid}</replaceable>
891 <replaceable>block_softlimit</replaceable>]
893 <replaceable>block_hardlimit</replaceable>]
895 <replaceable>inode_softlimit</replaceable>]
897 <replaceable>inode_hardlimit</replaceable>]
898 <replaceable>/mount_point</replaceable></literal>
902 <para>Sets file system quotas for users, groups or one project.
903 Limits can be specified with
904 <literal>--{block|inode}-{softlimit|hardlimit}</literal> or
905 their short equivalents
906 <literal>-b</literal>,
907 <literal>-B</literal>,
908 <literal>-i</literal>,
909 <literal>-I</literal>. Users can set 1, 2, 3 or 4 limits.
912 <literal>setquota</literal> interface is supported, but it may
913 be removed in a future Lustre software release.</para>
914 </footnote>Also, limits can be specified with special suffixes,
915 -b, -k, -m, -g, -t, and -p to indicate units of 1, 2^10, 2^20,
916 2^30, 2^40 and 2^50, respectively. By default, the block limits
917 unit is 1 kilobyte (1,024), and block limits are always
918 kilobyte-grained (even if specified in bytes). See
919 <xref linkend="dbdoclet.50438206_11903" />.</para>
923 <entry nameend="c2" namest="c1">
925 <literal>setquota -t -u|-g|-p [--block-grace
926 <replaceable>block_grace</replaceable>] [--inode-grace
927 <replaceable>inode_grace</replaceable>]
928 <replaceable>/mount_point</replaceable></literal>
932 <para>Sets the file system quota grace times for users or
933 groups. Grace time is specified in '
934 <literal>XXwXXdXXhXXmXXs</literal>' format or as an integer
936 <xref linkend="dbdoclet.50438206_11903" />.</para>
940 <entry nameend="c2" namest="c1">
942 <literal>help</literal>
946 <para>Provides brief help on various
947 <literal>lfs</literal> arguments.</para>
951 <entry nameend="c2" namest="c1">
953 <literal>exit/quit</literal>
957 <para>Quits the interactive
958 <literal>lfs</literal> session.</para>
965 <section remap="h5" xml:id="dbdoclet.50438206_11903">
966 <title>Examples</title>
967 <para>Creates a file striped on two OSTs with 128 KB on each
970 $ lfs setstripe -s 128k -c 2 /mnt/lustre/file1
972 <para>Deletes a default stripe pattern on a given directory. New files
973 use the default striping pattern.</para>
975 $ lfs setstripe -d /mnt/lustre/dir
977 <para>Lists the detailed object allocation of a given file.</para>
979 $ lfs getstripe -v /mnt/lustre/file1
981 <para>List all the mounted Lustre file systems and corresponding Lustre
986 <para>Efficiently lists all files in a given directory and its
987 subdirectories.</para>
989 $ lfs find /mnt/lustre
991 <para>Recursively lists all regular files in a given directory more than
994 $ lfs find /mnt/lustre -mtime +30 -type f -print
996 <para>Recursively lists all files in a given directory that have objects
997 on OST2-UUID. The lfs check servers command checks the status of all
998 servers (MDT and OSTs).</para>
1000 $ lfs find --obd OST2-UUID /mnt/lustre/
1002 <para>Lists all OSTs in the file system.</para>
1006 <para>Lists space usage per OST and MDT in human-readable format.</para>
1010 <para>Lists inode usage per OST and MDT.</para>
1014 <para>List space or inode usage for a specific OST pool.</para>
1017 <replaceable>filesystem</replaceable>[.
1018 <replaceable>pool</replaceable>] |
1019 <replaceable>pathname</replaceable>
1021 <para>List quotas of user 'bob'.</para>
1023 $ lfs quota -u bob /mnt/lustre
1025 <para>List quotas of project ID '1'.</para>
1027 $ lfs quota -p 1 /mnt/lustre
1029 <para>Show grace times for user quotas on
1030 <literal>/mnt/lustre</literal>.</para>
1032 $ lfs quota -t -u /mnt/lustre
1034 <para>Changes file owner and group.</para>
1036 $ lfs quotachown -i /mnt/lustre
1038 <para>Checks quotas for user and group. Turns on quotas after making the
1041 $ lfs quotacheck -ug /mnt/lustre
1043 <para>Turns on quotas of user and group.</para>
1045 $ lfs quotaon -ug /mnt/lustre
1047 <para>Turns off quotas of user and group.</para>
1049 $ lfs quotaoff -ug /mnt/lustre
1051 <para>Sets quotas of user 'bob', with a 1 GB block quota hardlimit and a
1052 2 GB block quota softlimit.</para>
1054 $ lfs setquota -u bob --block-softlimit 2000000 --block-hardlimit 1000000
1057 <para>Sets grace times for user quotas: 1000 seconds for block quotas, 1
1058 week and 4 days for inode quotas.</para>
1060 $ lfs setquota -t -u --block-grace 1000 --inode-grace 1w4d /mnt/lustre
1062 <para>Checks the status of all servers (MDT, OST)</para>
1066 <para>Creates a file striped on two OSTs from the pool
1067 <literal>my_pool</literal></para>
1069 $ lfs setstripe --pool my_pool -c 2 /mnt/lustre/file
1071 <para>Lists the pools defined for the mounted Lustre file system
1072 <literal>/mnt/lustre</literal></para>
1074 $ lfs pool_list /mnt/lustre/
1076 <para>Lists the OSTs which are members of the pool
1077 <literal>my_pool</literal> in file system
1078 <literal>my_fs</literal></para>
1080 $ lfs pool_list my_fs.my_pool
1082 <para>Finds all directories/files associated with
1083 <literal>poolA</literal>.</para>
1085 $ lfs find /mnt/lustre --pool poolA
1087 <para>Finds all directories/files not associated with a pool.</para>
1089 $ lfs find /mnt//lustre --pool ""
1091 <para>Finds all directories/files associated with pool.</para>
1093 $ lfs find /mnt/lustre ! --pool ""
1095 <para>Associates a directory with the pool
1096 <literal>my_pool</literal>, so all new files and directories are created
1099 $ lfs setstripe --pool my_pool /mnt/lustre/dir
1102 <section remap="h5">
1103 <title>See Also</title>
1105 <xref linkend="dbdoclet.50438219_38274" />
1109 <section xml:id="dbdoclet.lfs_migrate">
1112 <primary>lfs_migrate</primary>
1114 <literal>lfs_migrate</literal>
1117 <literal>lfs_migrate</literal> utility is a simple to migrate file
1118 <emphasis>data</emphasis> between OSTs.</para>
1119 <section remap="h5">
1120 <title>Synopsis</title>
1122 lfs_migrate [<replaceable>lfs_setstripe_options</replaceable>]
1123 [-h] [-n] [-q] [-R] [-s] [-y] [-0] [file|directory ...]
1126 <section remap="h5">
1127 <title>Description</title>
1129 <literal>lfs_migrate</literal> utility is a tool to assist migration
1130 of file data between Lustre OSTs. The utility copies each specified
1131 file to a temporary file using supplied <literal>lfs setstripe</literal>
1132 options, if any, optionally verifies the file contents have not changed,
1133 and then swaps the layout (OST objects) from the temporary file and the
1134 original file (for Lustre 2.5 and later), or renames the temporary file
1135 to the original filename. This allows the user/administrator to balance
1136 space usage between OSTs, or move files off OSTs that are starting to show
1137 hardware problems (though are still functional) or will be removed.</para>
1139 <para>For versions of Lustre before 2.5,
1140 <literal>lfs_migrate</literal> was not integrated with the MDS at all.
1141 That made it UNSAFE for use on files that were being modified by other
1142 applications, since the file was migrated through a copy and rename of
1143 the file. With Lustre 2.5 and later, the new file layout is swapped
1144 with the existing file layout, which ensures that the user-visible
1145 inode number is kept, and open file handles and locks on the file are
1148 <para>Files to be migrated can be specified as command-line arguments. If
1149 a directory is specified on the command-line then all files within the
1150 directory are migrated. If no files are specified on the command-line,
1151 then a list of files is read from the standard input, making
1152 <literal>lfs_migrate</literal> suitable for use with
1153 <literal>lfs find</literal> to locate files on specific OSTs and/or
1154 matching other file attributes, and other tools that generate a list
1155 of files on standard output.</para>
1156 <para>Unless otherwise specified through command-line options, the
1157 file allocation policies on the MDS dictate where the new files
1158 are placed, taking into account whether specific OSTs have been
1159 disabled on the MDS via <literal>lctl</literal> (preventing new
1160 files from being allocated there), whether some OSTs are overly full
1161 (reducing the number of files placed on those OSTs), or if there is
1162 a specific default file striping for the parent directory (potentially
1163 changing the stripe count, stripe size, OST pool, or OST index of a
1167 <literal>lfs_migrate</literal> utility can also be used in some cases to
1170 <primary>fragmentation</primary>
1171 </indexterm>fragmentation. File fragmentation will typically reduce
1172 Lustre file system performance. File fragmentation may be observed on
1173 an aged file system and will commonly occur if the file was written by
1174 many threads. Provided there is sufficient free space (or if it was
1175 written when the file system was nearly full) that is less fragmented
1176 than the file being copied, re-writing a file with
1177 <literal>lfs_migrate</literal> will result in a migrated file with
1178 reduced fragmentation. The tool
1179 <literal>filefrag</literal> can be used to report file fragmentation.
1181 <xref linkend="dbdoclet.50438206_75125" /></para>
1184 <para>As long as a file has extent lengths of tens of megabytes (
1185 <replaceable>read_bandwidth * seek_time</replaceable>) or more, the
1186 read performance for the file will not be significantly impacted by
1187 fragmentation, since the read pipeline can be filled by large reads
1188 from disk even with an occasional disk seek.</para>
1191 <section remap="h5">
1192 <title>Options</title>
1193 <para>Options supporting
1194 <literal>lfs_migrate</literal> are described below.</para>
1195 <informaltable frame="all">
1197 <colspec colname="c1" colwidth="50*" />
1198 <colspec colname="c2" colwidth="50*" />
1203 <emphasis role="bold">Option</emphasis>
1208 <emphasis role="bold">Description</emphasis>
1218 <replaceable>stripecount</replaceable></literal>
1222 <para>Restripe file using the specified stripe count. This
1223 option may not be specified at the same time as the
1224 <literal>-R</literal> option.</para>
1230 <literal>-h</literal>
1234 <para>Display help information.</para>
1239 <literal>-l</literal>
1242 <para>Migrate files with hard links (skips, by default). Files
1243 with multiple hard links are split into multiple separate files
1245 <literal>lfs_migrate</literal>, so they are skipped, by
1246 default, to avoid breaking the hard links.</para>
1251 <literal>-n</literal>
1254 <para>Only print the names of files to be migrated.</para>
1260 <literal>-q</literal>
1264 <para>Run quietly (does not print filenames or status).</para>
1269 <literal>-R</literal>
1271 <entry>Restripe file using default directory striping instead of
1272 keeping striping. This option may not be specified at the same
1274 <literal>-c</literal> option.</entry>
1278 <literal>-s</literal>
1280 <entry>Skip file data comparison after migrate. Default is to
1281 compare migrated file against original to verify
1282 correctness.</entry>
1287 <literal>-y</literal>
1292 <literal>y</literal>' to usage warning without prompting (for
1293 scripts, use with caution).</para>
1298 <literal>-0</literal>
1300 <entry>Expect NUL-terminated filenames on standard input, as
1301 generated by <literal>lfs find -print0</literal> or
1302 <literal>find -print0</literal>. This allows filenames with
1303 embedded newlines to be handled correctly.
1310 <section remap="h5">
1311 <title>Examples</title>
1312 <para>Rebalance all files in
1313 <literal>/mnt/lustre/dir</literal>:</para>
1315 $ lfs_migrate /mnt/lustre/dir
1317 <para>Migrate files in /test filesystem on OST0004 larger than 4 GB in
1318 size and older than a day old:</para>
1320 $ lfs find /test -obd test-OST0004 -size +4G -mtime +1 | lfs_migrate -y
1323 <section remap="h5">
1324 <title>See Also</title>
1326 <xref linkend="dbdoclet.50438206_94597" />
1330 <section xml:id="dbdoclet.50438206_75125">
1333 <primary>filefrag</primary>
1335 <literal>filefrag</literal>
1338 <literal>e2fsprogs</literal> package contains the
1339 <literal>filefrag</literal> tool which reports the extent of file
1340 fragmentation.</para>
1341 <section remap="h5">
1342 <title>Synopsis</title>
1344 filefrag [ -belsv ] [ files... ]
1347 <section remap="h5">
1348 <title>Description</title>
1350 <literal>filefrag</literal> utility reports the extent of fragmentation in
1352 <literal>filefrag</literal> utility obtains the extent information from
1353 Lustre files using the
1354 <literal>FIEMAP ioctl</literal>, which is efficient and fast, even for
1355 very large files.</para>
1356 <para>In default mode
1358 <para>The default mode is faster than the verbose/extent mode since it
1359 only counts the number of extents.</para>
1361 <literal>filefrag</literal> prints the number of physically discontiguous
1362 extents in the file. In extent or verbose mode, each extent is printed
1363 with details such as the blocks allocated on each OST. For a Lustre file
1364 system, the extents are printed in device offset order (i.e. all of the
1365 extents for one OST first, then the next OST, etc.), not file logical
1366 offset order. If the file logical offset order was used, the Lustre
1367 striping would make the output very verbose and difficult to see if there
1368 was file fragmentation or not.</para>
1370 <para>Note that as long as a file has extent lengths of tens of
1371 megabytes or more (i.e.
1372 <replaceable>read_bandwidth * seek_time >
1373 extent_length</replaceable>), the read performance for the file will
1374 not be significantly impacted by fragmentation, since file readahead
1375 can fully utilize the disk disk bandwidth even with occasional
1378 <para>In default mode
1380 <para>The default mode is faster than the verbose/extent mode.</para>
1382 <literal>filefrag</literal> returns the number of physically discontiguous
1383 extents in the file. In extent or verbose mode, each extent is printed
1384 with details. For a Lustre file system, the extents are printed in device
1385 offset order, not logical offset order.</para>
1387 <section remap="h5">
1388 <title>Options</title>
1389 <para>The options and descriptions for the
1390 <literal>filefrag</literal> utility are listed below.</para>
1391 <informaltable frame="all">
1393 <colspec colname="c1" colwidth="50*" />
1394 <colspec colname="c2" colwidth="50*" />
1399 <emphasis role="bold">Option</emphasis>
1404 <emphasis role="bold">Description</emphasis>
1413 <literal>-b</literal>
1417 <para>Uses the 1024-byte blocksize for the output. By default,
1418 this blocksize is used by the Lustre file system, since OSTs
1419 may use different block sizes.</para>
1425 <literal>-e</literal>
1429 <para>Uses the extent mode when printing the output. This is
1430 the default for Lustre files in verbose mode.</para>
1436 <literal>-l</literal>
1440 <para>Displays extents in LUN offset order. This is the only
1441 available mode for Lustre.</para>
1447 <literal>-s</literal>
1451 <para>Synchronizes any unwritten file data to disk before
1452 requesting the mapping.</para>
1458 <literal>-v</literal>
1462 <para>Prints the file's layout in verbose mode when checking
1463 file fragmentation, including the logical to physical mapping
1464 for each extent in the file and the OST index.</para>
1471 <section remap="h5">
1472 <title>Examples</title>
1473 <para>Lists default output.</para>
1475 $ filefrag /mnt/lustre/foo
1476 /mnt/lustre/foo: 13 extents found
1478 <para>Lists verbose output in extent format.</para>
1480 $ filefrag -v /mnt/lustre/foo
1481 Filesystem type is: bd00bd0
1482 File size of /mnt/lustre/foo is 1468297786 (1433888 blocks of 1024 bytes)
1483 ext: device_logical: physical_offset: length: dev: flags:
1484 0: 0.. 122879: 2804679680..2804802559: 122880: 0002: network
1485 1: 122880.. 245759: 2804817920..2804940799: 122880: 0002: network
1486 2: 245760.. 278527: 2804948992..2804981759: 32768: 0002: network
1487 3: 278528.. 360447: 2804982784..2805064703: 81920: 0002: network
1488 4: 360448.. 483327: 2805080064..2805202943: 122880: 0002: network
1489 5: 483328.. 606207: 2805211136..2805334015: 122880: 0002: network
1490 6: 606208.. 729087: 2805342208..2805465087: 122880: 0002: network
1491 7: 729088.. 851967: 2805473280..2805596159: 122880: 0002: network
1492 8: 851968.. 974847: 2805604352..2805727231: 122880: 0002: network
1493 9: 974848.. 1097727: 2805735424..2805858303: 122880: 0002: network
1494 10: 1097728.. 1220607: 2805866496..2805989375: 122880: 0002: network
1495 11: 1220608.. 1343487: 2805997568..2806120447: 122880: 0002: network
1496 12: 1343488.. 1433599: 2806128640..2806218751: 90112: 0002: network
1497 /mnt/lustre/foo: 13 extents found
1501 <section xml:id="dbdoclet.50438206_86244">
1504 <primary>mount</primary>
1506 <literal>mount</literal>
1509 <literal>mount(8)</literal> Linux command is used to mount a Lustre file
1510 system. When mounting a Lustre file system, mount(8) executes the
1511 <literal>/sbin/mount.lustre</literal> command to complete the mount. The
1512 mount command supports these options specific to a Lustre file
1514 <informaltable frame="all">
1516 <colspec colname="c1" colwidth="50*" />
1517 <colspec colname="c2" colwidth="50*" />
1522 <emphasis role="bold">Server options</emphasis>
1527 <emphasis role="bold">Description</emphasis>
1536 <literal>abort_recov</literal>
1540 <para>Aborts recovery when starting a target</para>
1546 <literal>nosvc</literal>
1550 <para>Starts only MGS/MGC servers</para>
1556 <literal>nomgs</literal>
1560 <para>Start a MDT with a co-located MGS without starting the
1567 <literal>exclude</literal>
1571 <para>Starts with a dead OST</para>
1577 <literal>md_stripe_cache_size</literal>
1581 <para>Sets the stripe cache size for server side disk with a
1582 striped raid configuration</para>
1588 <informaltable frame="all">
1590 <colspec colname="c1" colwidth="50*" />
1591 <colspec colname="c2" colwidth="50*" />
1596 <emphasis role="bold">Client options</emphasis>
1601 <emphasis role="bold">Description</emphasis>
1610 <literal>flock/noflock/localflock</literal>
1614 <para>Enables/disables global flock or local flock support</para>
1620 <literal>user_xattr/nouser_xattr</literal>
1624 <para>Enables/disables user-extended attributes</para>
1630 <literal>user_fid2path/nouser_fid2path</literal>
1634 <para condition="l23">Enables/disables FID to path translation by
1635 regular users</para>
1641 <literal>retry=</literal>
1645 <para>Number of times a client will retry to mount the file
1653 <section xml:id="dbdoclet.50438206_56217">
1654 <title>Handling Timeouts</title>
1655 <para>Timeouts are the most common cause of hung applications. After a
1656 timeout involving an MDS or failover OST, applications attempting to access
1657 the disconnected resource wait until the connection gets
1659 <para>When a client performs any remote operation, it gives the server a
1660 reasonable amount of time to respond. If a server does not reply either due
1661 to a down network, hung server, or any other reason, a timeout occurs which
1662 requires a recovery.</para>
1663 <para>If a timeout occurs, a message (similar to this one), appears on the
1664 console of the client, and in
1665 <literal>/var/log/messages</literal>:</para>
1667 LustreError: 26597:(client.c:810:ptlrpc_expire_one_request()) @@@ timeout
1669 req@a2d45200 x5886/t0 o38->mds_svc_UUID@NID_mds_UUID:12 lens 168/64 ref 1 fl