[fs/lustre-release.git] / lustre / doc / lfs.1

.TH lfs 1 "2009 Jan 29" Lustre "user utilities"
.SH NAME
lfs \- Lustre utility to create a file with specific striping pattern, find the striping pattern of existing files, do certain quota operations, and manage distributed namespace options for directories
.SH SYNOPSIS
.br
.B lfs
.br
.B lfs changelog [--follow] <mdtname> [startrec [endrec]]
.br
.B lfs changelog_clear <mdtname> <id> <endrec>
.br
.B lfs check <mds|osts|servers>
.br
.B lfs df [-i] [-h] [--lazy] [--pool|-p <fsname>[.<pool>] [path]
.br
.B lfs fid2path [--link <linkno>] <fsname|rootpath> <fid> ...
.br
.B lfs find <directory|filename>
        \fB[[!] --atime|-A [-+]N] [[!] --mtime|-M [-+]N] [[!] --ctime|-C [+-]N]
        \fB[--maxdepth|-D N] [[!] --mdt|-m <uuid|index,...>] [--name|-n pattern]
        \fB[[!] --ost|-O <uuid|index,...>] [--print|-p] [--print0|-P]
        \fB[[!] --size|-s [-+]N[kMGTPE]]
        \fB[[!] --stripe-count|-c [+-]<stripes>]
        \fB[[!] --stripe-index|-i <index,...>]
        \fB[[!] --stripe-size|-S [+-]N[kMG]]
        \fB[[!] --layout|-L raid0,released]
        \fB[--type |-t {bcdflpsD}] [[!] --gid|-g|--group|-G <gname>|<gid>]
        \fB[[!] --uid|-u|--user|-U <uname>|<uid>] [[!] --pool <pool>]\fR
.br
.B lfs getname [-h]|[path ...]
.br
.B lfs getstripe [--obd|-O <uuid>] [--quiet|-q] [--verbose|-v] 
        \fB[--stripe-count|-c ] [--stripe-index|-i] [--mdt-index|-M]
        \fB[--stripe-size|-S] [--directory|-d]
        \fB[--layout|-L]
        \fB[--pool|-p] [--recursive|-r] [--raw|-R] <dirname|filename> ...\fR
.br
.B lfs setstripe [--stripe-size|-S stripe_size] [--stripe-count|-c stripe_count]
        \fB[--stripe-index|-i start_ost_index ] [--pool|-p <poolname>]
        \fB<directory|filename>\fR
.br
.B lfs setstripe -d <dir>
.br
.B lfs osts
.RB [ path ]
.br
.B path2fid <path> ...
.br
.B lfs pool_list <filesystem>[.<pool>] | <pathname>
.br
.B lfs quota [-q] [-v] [-o obd_uuid|-I ost_idx|-i mdt_idx] [-u <uname>| -u <uid>|-g <gname>| -g <gid>] <filesystem>
.br
.B lfs quota -t <-u|-g> <filesystem>
.br
.B lfs quotacheck [-ug] <filesystem>
.br
.B lfs quotaon [-ugf] <filesystem>
.br
.B lfs quotaoff [-ug] <filesystem>
.br
.B lfs setquota <-u|--user|-g|--group> <uname|uid|gname|gid>
             \fB[--block-softlimit <block-softlimit>]
             \fB[--block-hardlimit <block-hardlimit>]
             \fB[--inode-softlimit <inode-softlimit>]
             \fB[--inode-hardlimit <inode-hardlimit>]
             \fB<filesystem>\fR
.br
.B lfs setquota <-u|--user|-g|--group> <uname|uid|gname|gid>
             \fB[-b <block-softlimit>] [-B <block-hardlimit>]
             \fB[-i <inode-softlimit>] [-I <inode-hardlimit>]
             \fB<filesystem>\fR
.br
.B lfs setquota -t <-u|-g>
             \fB[--block-grace <block-grace>]
             \fB[--inode-grace <inode-grace>]
             \fB<filesystem>\fR
.br
.B lfs setquota -t <-u|-g>
             \fB[-b <block-grace>] [-i <inode-grace>]
             \fB<filesystem>\fR
.br
.br
.B lfs swap_layouts <filename1> <filename2>
.br
.B lfs data_version [-n] \fB<filename>\fR
.br
.B lfs help
.SH DESCRIPTION
.B lfs
can be used to create a new file with a specific striping pattern, determine the default striping pattern, gather the extended attributes (object numbers and location) for a specific file. It can be invoked interactively without any arguments or in a non-interactive mode with one of the arguements supported. 
.SH OPTIONS
The various options supported by lctl are listed and explained below:
.TP
.B changelog
Show the metadata changes on an MDT.  Start and end points are optional.  The --follow option will block on new changes; this option is only valid when run direclty on the MDT node.
.TP
.B changelog_clear
Indicate that changelog records previous to <endrec> are no longer of
interest to a particular consumer <id>, potentially allowing the MDT to
free up disk space. An <endrec> of 0 indicates the current last record.
Changelog consumers must be registered on the MDT node using \fBlctl\fR.
.TP
.B check 
Display the status of MDS or OSTs (as specified in the command) or all the servers (MDS and OSTs)
.TP
.B df [-i] [-h] [--lazy] [--pool|-p <fsname>[.<pool>] [path]
Report filesystem disk space usage or inodes usage (with \fB-i\fR) of each
MDT/OST, or a subset of OSTs if a pool is specified with \fB-p\fR.  By default
print the usage of all mounted Lustre filesystems, otherwise if \fBpath\fR is
specified print only the usage of that filesystem.  If \fB-h\fR is given, the
output is printed in \fIh\fRuman readable format, using SI base-2 suffixes
for \fBM\fRega-, \fBG\fRiga-, \fBT\fRera-, \fBP\fReta-, or \fBE\fRxabytes.
The \fB--lazy\fR/\fB-l\fR option skips any OST that is currently disconnected
from the client.  This avoids blocking the \fBdf\fR output if an OST is down,
and only returns the space on the OSTs that can currently be accessed.
.TP
.B find 
To search the directory tree rooted at the given dir/file name for the files that match the given parameters: \fB--atime\fR (file was last accessed N*24 hours ago), \fB--ctime\fR (file's status was last changed N*24 hours ago), \fB--mtime\fR (file's data was last modified N*24 hours ago), \fB--obd\fR (file has an object on a specific OST or OSTs), \fB--size\fR (file has size in bytes, or \fBk\fRilo-, \fBM\fRega-, \fBG\fRiga-, \fBT\fRera-, \fBP\fReta-, or \fBE\fRxabytes if a suffix is given), \fB--type\fR (file has the type: \fBb\fRlock, \fBc\fRharacter, \fBd\fRirectory, \fBp\fRipe, \fBf\fRile, sym\fBl\fRink, \fBs\fRocket, or \fBD\fRoor (Solaris)), \fB--uid\fR (file has specific numeric user ID), \fB--user\fR (file owned by specific user, numeric user ID allowed), \fB--gid\fR (file has specific group ID), \fB--group\fR (file belongs to specific group, numeric group ID allowed), \fB--layout\fR (file has a raid0 layout or is released). The option \fB--maxdepth\fR limits find to decend at most N levels of directory tree. The options \fB--print\fR and \fB--print0\fR print full file name, followed by a newline or NUL character correspondingly.  Using \fB!\fR before an option negates its meaning (\fIfiles NOT matching the parameter\fR).  Using \fB+\fR before a numeric value means \fIfiles with the parameter OR MORE\fR, while \fB-\fR before a numeric value means \fIfiles with the parameter OR LESS\fR.
.TP
.B getname [-h]|[path ...]
Report all the Lustre mount points and the corresponding Lustre filesystem
instance. If one or more \fBpath\fR entries are provided, then only the
Lustre instance for these mount points is returned. If the path given is not on
a Lustre instance 'No such device' is reported.
.TP
.B osts 
.RB [ path ]
List all the OSTs for all mounted filesystems. If a \fBpath\fR is provided
that is located on a lustre mounted file system then only the OSTs belonging
to that filesystem are displayed.
.TP
.B getstripe [--obd|-O <uuid>] [--quiet|-q] [--verbose|-v] 
        \fB[--count | -c ] [--index | -i | --offset | -o  ]
        \fB[--pool | -p ] [--size | -s ] [--directory | -d ]
        \fB[--layout | -L]
        \fB[--recursive | -r ] [--raw | -R ] <dirname|filename>\fR
.br
List the striping information for a given filename or directory tree.
By default the stripe count, size, and offset will be returned. If you
only want specific striping information then the options of
.BR --count ,
.BR --size ,
.BR --index ,
.BR --offset ,
.BR --layout ,
or
.B --pool  
can be used to return only the specific fields.
.br
If the
.B --raw
option is specified, the stripe information is printed without substituting the
filesystem's default values for unspecified fields. If the striping EA is not
set, 0, 0, and -1 will be printed for the stripe count, size, and offset
respectively.
In the case where you only want details about the files' object id
information then the
.B --quiet
option is used. Additional information available about striping can be
displayed with
.BR --verbose .
The default behavior when a directory is specified is to list the striping
information for all files within the specified directory (like
.RB ' "ls -l" ') .
This can be expanded with
.B --recursive
which will recurse into all subdirectories.
If you wish to get striping information for only the specified directory, then
.B --directory
can be used to limit the information, like
.RB ' "ls -d" ').
You can limit the returned files to those with objects on a specific OST with
.BR --obd .
.TP
.B setstripe [--stripe-count|-c stripe_count] [--stripe-size|-S stripe_size]
        \fB[--stripe-index|-i start_ost_index] [--pool <poolname>]
        \fB<dirname|filename>\fR
.br
To create a new file, or set the directory default, with the specified striping parameters.  The
.I stripe_count
is the number of OSTs to stripe a file over. A
.I stripe_count
of 0 means to use the filesystem-wide default stripe count (default 1), and a
.I stripe_count
of -1 means to stripe over all available OSTs.  The
.I stripe_size
is the number of bytes to store on each OST before moving to the next OST.  A
.I stripe_size
of 0 means to use the filesystem-wide default stripe_size (default 1MB).  The
.I start_ost_index
is the OST index (starting at 0) on which to start striping for this file.  A
.I start_ost_index
of -1 allows the MDS to choose the starting index and it is strongly recommended, as this allows space and load balancing to be done by the MDS as needed.  The
.I poolname
is the name of a predefined pool of OSTs (see 
.B lctl
) that will be used for striping. The 
.IR stripe_count ,
.IR stripe_size ,
and
.I start_ost_index
will be used as well; the 
.I start_ost_index
must be part of the pool or an error will be returned. 
.TP
.B setstripe -d
Delete the default striping on the specified directory.
.TP
.B fid2path [--link <linkno>] <fsname|rootpath> <fid> ...
Print out the pathname(s) for the specified \fIfid\fR(s) from the filesystem
mounted at \fBrootpath\fR or named \fBfsname\fR.  If a file has multiple
hard links, then all of the pathnames for that file are printed, unless
\fB--link\fR limits the printing to only the specified link number (starting
at 0, in no particular order).  If multiple fids are specified, but only a
single pathname is needed for each file, use \fB--link 0\fR.
.TP
.B path2fid <path> ...
Print out the FIDs for the specified \fBpath(s)\fR.  If multiple pathnames
are given, then they will be printed one per line with the path as prefix.
.TP
.B pool_list
.RI { filesystem }[ .poolname "] | {" pathname }
List the pools in 
.I filesystem
or
.IR pathname ,
or the OSTs in
.IR filesystem.pool .
.TP
.B quota [-q] [-v] [-o obd_uuid|-i mdt_idx|-I ost_idx] [-u|-g <uname>|<uid>|<gname>|<gid>] <filesystem>
To display disk usage and limits, either for the full filesystem, or for objects on a specific obd. A user or group name or an ID can be specified. If both user and group are omitted quotas for current uid/gid are shown. -v provides more verbose (with per-obd statistics) output. -q disables printing of additional descriptions (including column titles).
.TP
.B quota -t <-u|-g> <filesystem>
To display block and inode grace times for user (-u) or group (-g) quotas
.TP
.B quotacheck [-ugf] <filesystem> (deprecated as of 2.4.0)
To scan the specified filesystem for disk usage, and create or update quota files. Options specify quota for users (-u) groups (-g) and force (-f). Not useful anymore with servers >= 2.4.0 since space accounting is always turned on.
.TP
.B quotaon [-ugf] <filesystem> (deprecated as of 2.4.0)
To turn filesystem quotas on. Options specify quota for users (-u) groups (-g) and force (-f). Not used anymore in lustre 2.4.0 where quota enforcement must be enabled via conf_param (e.g. lctl conf_param ${FSNAME}.quota.<ost|mdt>=<u|g|ug>)
.TP
.B quotaoff [-ugf] <filesystem> (deprecated as of 2.4.0)
To turn filesystem quotas off.  Options specify quota for users (-u) groups (-g) and force (-f). Not used anymore in lustre 2.4.0 where quota enforcement can be turned off (for inode or block) by running the following command on the MGS: lctl conf_param ${FSNAME}.quota.<ost|mdt>=""
.TP
.B setquota  <-u|-g> <uname>|<uid>|<gname>|<gid> [--block-softlimit <block-softlimit>] [--block-hardlimit <block-hardlimit>] [--inode-softlimit <inode-softlimit>] [--inode-hardlimit <inode-hardlimit>] <filesystem>
To set filesystem quotas for users or groups. Limits can be specified with -b, -k, -m, -g, -t, -p suffixes which specify units of 1, 2^10, 2^20, 2^30, 2^40 and 2^50 accordingly. Block limits unit is kilobyte (1024) by default and block limits are always kilobyte-grained (even if specified in bytes), see EXAMPLES
.TP
.B setquota -t [-u|-g] [--block-grace <block-grace>] [--inode-grace <inode-grace>] <filesystem>
To set filesystem quota grace times for users or groups. Grace time is specified in "XXwXXdXXhXXmXXs" format or as an integer seconds value, see EXAMPLES
.TP
.B swap_layouts <filename1> <filename2>
Swap the data (layout and OST objects) of two regular files. The
two files have to be in the same filesystem, owned by the same user,
reside on the same MDT and writable by the user.

Swapping the layout of two directories is not permitted.
.TP
.B data_version [-n] <filename>
Display current version of file data. If -n is specified, data version is read
without taking lock. As a consequence, data version could be outdated if there
is dirty caches on filesystem clients, but this will not force data flushes and
has less impact on filesystem.

Even without -n, race conditions are possible and data version should be
checked before and after an operation to be confident the data did not change
during it.
.TP
.B mkdir <--index|-i mdt_index> <dir>
Allocate a new directory on a specified MDT.  

The "lfs mkdir" command is only executable by root unless
"mdt.*.enable_remote_dir_gid" is set via "lctl set_param" to be either a
non-zero GID to limit it to a single group (e.g. "operator" or "admin"),
or "-1" to allow any group to create remote directories.

The root of the file system is on MDT0000, and directories and files inherit the
MDT of their parent directory unless a different MDT is specified with this
command.

By default, only directories on MDT0000 can contain directories that are not on
the same MDT.  However, if "mdt.*.enable_remote_dir" is set non-zero on an MDT
then it will allow creating remote directories that have parents other than
MDT0000. This is restricted to avoid creating directory trees that have
intermediate path components on a series different MDTs and become unavailable
if any of the intermediate MDTs are offline.
.TP
.B help 
Provides brief help on the various arguments
.TP
.B exit/quit 
Quit the interactive lfs session
.SH EXAMPLES
.TP
.B $ lfs setstripe -s 128k -c 2 /mnt/lustre/file1
This creates a file striped on two OSTs with 128kB on each stripe.
.TP
.B $ lfs setstripe -d /mnt/lustre/dir
This deletes a default stripe pattern on dir. New files will use the default striping pattern created therein.
.TP
.B $ lfs getstripe -v /mnt/lustre/file1
Lists the detailed object allocation of a given file
.TP
.B $ lfs find /mnt/lustre
Efficiently lists all files in a given directory and its subdirectories
.TP
.B $ lfs find /mnt/lustre -mtime +30 -type f -print
Recursively list all regular files in given directory more than 30 days old
.TP
.B $ lfs find --obd OST2-UUID /mnt/lustre/
Recursively list all files in a given directory that have objects on OST2-UUID.
.tP
.B $ lfs check servers 
Check the status of all servers (MDT, OST)
.TP
.B $ lfs osts
List all the OSTs
.TP
.B $ lfs df -h 
Lists space usage per OST and MDT in human readable format.
.TP
.B $ lfs df -i 
Lists inode usage per OST and MDT
.TP
.B $ lfs df --pool <filesystem>[.<pool>] | <pathname>
List space or inode usage for a specific OST pool
.TP
.B $ lfs quota -u bob /mnt/lustre
List quotas of user `bob'
.TP
.B $ lfs quota -t -u /mnt/lustre
Show grace times for user quotas on /mnt/lustre
.TP
.B $ lfs quotachown -i /mnt/lustre
Change file owner and group
.TP
.B $ lfs quotacheck -ug /mnt/lustre
Quotacheck for user and group - will turn on quotas after making the check.
.TP
.B $ lfs quotaon -ug /mnt/lustre
Turn quotas of user and group on
.TP
.B $ lfs quotaoff -ug /mnt/lustre
Turn quotas of user and group off
.TP
.B $ lfs setquota -u bob --block-softlimit 2000000 --block-hardlimit 1000000 /mnt/lustre
Set quotas of user `bob': 1GB block quota hardlimit and 2 GB block quota softlimit
.TP
.B $ lfs setquota -t -u --block-grace 1000 --inode-grace 1w4d /mnt/lustre
Set grace times for user quotas: 1000 seconds for block quotas, 1 week and 4 days for inode quotas
.TP
.SH BUGS
The \fBlfs find\fR command isn't as comprehensive as \fBfind\fR(1).
.SH AUTHOR
The lfs command is part of the Lustre filesystem.
.SH SEE ALSO
.BR lfs-hsm (1),
.BR lfs-setdirstripe (1),
.BR lfs-getdirstripe (1),
.BR lfs-mkdir (1),
.BR lctl (8),
.BR lustre (7)