Documentation/filesystems/xfs.txt

   1
   2 The SGI XFS Filesystem
   3 ======================
   4
   5 XFS is a high performance journaling filesystem which originated
   6 on the SGI IRIX platform.  It is completely multi-threaded, can
   7 support large files and large filesystems, extended attributes,
   8 variable block sizes, is extent based, and makes extensive use of
   9 Btrees (directories, extents, free space) to aid both performance
  10 and scalability.
  11
  12 Refer to the documentation at http://oss.sgi.com/projects/xfs/
  13 for further details.  This implementation is on-disk compatible
  14 with the IRIX version of XFS.
  15
  16
  17 Mount Options
  18 =============
  19
  20 When mounting an XFS filesystem, the following options are accepted.
  21
  22   allocsize=size
  23         Sets the buffered I/O end-of-file preallocation size when
  24         doing delayed allocation writeout (default size is 64KiB).
  25         Valid values for this option are page size (typically 4KiB)
  26         through to 1GiB, inclusive, in power-of-2 increments.
  27
  28   attr2/noattr2
  29         The options enable/disable (default is disabled for backward
  30         compatibility on-disk) an "opportunistic" improvement to be
  31         made in the way inline extended attributes are stored on-disk.
  32         When the new form is used for the first time (by setting or
  33         removing extended attributes) the on-disk superblock feature
  34         bit field will be updated to reflect this format being in use.
  35
  36   barrier
  37         Enables the use of block layer write barriers for writes into
  38         the journal and unwritten extent conversion.  This allows for
  39         drive level write caching to be enabled, for devices that
  40         support write barriers.
  41
  42   dmapi
  43         Enable the DMAPI (Data Management API) event callouts.
  44         Use with the "mtpt" option.
  45
  46   grpid/bsdgroups and nogrpid/sysvgroups
  47         These options define what group ID a newly created file gets.
  48         When grpid is set, it takes the group ID of the directory in
  49         which it is created; otherwise (the default) it takes the fsgid
  50         of the current process, unless the directory has the setgid bit
  51         set, in which case it takes the gid from the parent directory,
  52         and also gets the setgid bit set if it is a directory itself.
  53
  54   ihashsize=value
  55         In memory inode hashes have been removed, so this option has
  56         no function as of August 2007. Option is deprecated.
  57
  58   ikeep/noikeep
  59         When ikeep is specified, XFS does not delete empty inode clusters
  60         and keeps them around on disk. ikeep is the traditional XFS
  61         behaviour. When noikeep is specified, empty inode clusters
  62         are returned to the free space pool. The default is noikeep for
  63         non-DMAPI mounts, while ikeep is the default when DMAPI is in use.
  64
  65   inode64
  66         Indicates that XFS is allowed to create inodes at any location
  67         in the filesystem, including those which will result in inode
  68         numbers occupying more than 32 bits of significance.  This is
  69         provided for backwards compatibility, but causes problems for
  70         backup applications that cannot handle large inode numbers.
  71
  72   largeio/nolargeio
  73         If "nolargeio" is specified, the optimal I/O reported in
  74         st_blksize by stat(2) will be as small as possible to allow user
  75         applications to avoid inefficient read/modify/write I/O.
  76         If "largeio" specified, a filesystem that has a "swidth" specified
  77         will return the "swidth" value (in bytes) in st_blksize. If the
  78         filesystem does not have a "swidth" specified but does specify
  79         an "allocsize" then "allocsize" (in bytes) will be returned
  80         instead.
  81         If neither of these two options are specified, then filesystem
  82         will behave as if "nolargeio" was specified.
  83
  84   logbufs=value
  85         Set the number of in-memory log buffers.  Valid numbers range
  86         from 2-8 inclusive.
  87         The default value is 8 buffers for filesystems with a
  88         blocksize of 64KiB, 4 buffers for filesystems with a blocksize
  89         of 32KiB, 3 buffers for filesystems with a blocksize of 16KiB
  90         and 2 buffers for all other configurations.  Increasing the
  91         number of buffers may increase performance on some workloads
  92         at the cost of the memory used for the additional log buffers
  93         and their associated control structures.
  94
  95   logbsize=value
  96         Set the size of each in-memory log buffer.
  97         Size may be specified in bytes, or in kilobytes with a "k" suffix.
  98         Valid sizes for version 1 and version 2 logs are 16384 (16k) and
  99         32768 (32k).  Valid sizes for version 2 logs also include
 100         65536 (64k), 131072 (128k) and 262144 (256k).
 101         The default value for machines with more than 32MiB of memory
 102         is 32768, machines with less memory use 16384 by default.
 103
 104   logdev=device and rtdev=device
 105         Use an external log (metadata journal) and/or real-time device.
 106         An XFS filesystem has up to three parts: a data section, a log
 107         section, and a real-time section.  The real-time section is
 108         optional, and the log section can be separate from the data
 109         section or contained within it.
 110
 111   mtpt=mountpoint
 112         Use with the "dmapi" option.  The value specified here will be
 113         included in the DMAPI mount event, and should be the path of
 114         the actual mountpoint that is used.
 115
 116   noalign
 117         Data allocations will not be aligned at stripe unit boundaries.
 118
 119   noatime
 120         Access timestamps are not updated when a file is read.
 121
 122   norecovery
 123         The filesystem will be mounted without running log recovery.
 124         If the filesystem was not cleanly unmounted, it is likely to
 125         be inconsistent when mounted in "norecovery" mode.
 126         Some files or directories may not be accessible because of this.
 127         Filesystems mounted "norecovery" must be mounted read-only or
 128         the mount will fail.
 129
 130   nouuid
 131         Don't check for double mounted file systems using the file system uuid.
 132         This is useful to mount LVM snapshot volumes.
 133
 134   uquota/usrquota/uqnoenforce/quota
 135         User disk quota accounting enabled, and limits (optionally)
 136         enforced.  Refer to xfs_quota(8) for further details.
 137
 138   gquota/grpquota/gqnoenforce
 139         Group disk quota accounting enabled and limits (optionally)
 140         enforced.  Refer to xfs_quota(8) for further details.
 141
 142   pquota/prjquota/pqnoenforce
 143         Project disk quota accounting enabled and limits (optionally)
 144         enforced.  Refer to xfs_quota(8) for further details.
 145
 146   sunit=value and swidth=value
 147         Used to specify the stripe unit and width for a RAID device or
 148         a stripe volume.  "value" must be specified in 512-byte block
 149         units.
 150         If this option is not specified and the filesystem was made on
 151         a stripe volume or the stripe width or unit were specified for
 152         the RAID device at mkfs time, then the mount system call will
 153         restore the value from the superblock.  For filesystems that
 154         are made directly on RAID devices, these options can be used
 155         to override the information in the superblock if the underlying
 156         disk layout changes after the filesystem has been created.
 157         The "swidth" option is required if the "sunit" option has been
 158         specified, and must be a multiple of the "sunit" value.
 159
 160   swalloc
 161         Data allocations will be rounded up to stripe width boundaries
 162         when the current end of file is being extended and the file
 163         size is larger than the stripe width size.
 164
 165
 166 sysctls
 167 =======
 168
 169 The following sysctls are available for the XFS filesystem:
 170
 171   fs.xfs.stats_clear            (Min: 0  Default: 0  Max: 1)
 172         Setting this to "1" clears accumulated XFS statistics
 173         in /proc/fs/xfs/stat.  It then immediately resets to "0".
 174
 175   fs.xfs.xfssyncd_centisecs     (Min: 100  Default: 3000  Max: 720000)
 176         The interval at which the xfssyncd thread flushes metadata
 177         out to disk.  This thread will flush log activity out, and
 178         do some processing on unlinked inodes.
 179
 180   fs.xfs.xfsbufd_centisecs      (Min: 50  Default: 100  Max: 3000)
 181         The interval at which xfsbufd scans the dirty metadata buffers list.
 182
 183   fs.xfs.age_buffer_centisecs   (Min: 100  Default: 1500  Max: 720000)
 184         The age at which xfsbufd flushes dirty metadata buffers to disk.
 185
 186   fs.xfs.error_level            (Min: 0  Default: 3  Max: 11)
 187         A volume knob for error reporting when internal errors occur.
 188         This will generate detailed messages & backtraces for filesystem
 189         shutdowns, for example.  Current threshold values are:
 190
 191                 XFS_ERRLEVEL_OFF:       0
 192                 XFS_ERRLEVEL_LOW:       1
 193                 XFS_ERRLEVEL_HIGH:      5
 194
 195   fs.xfs.panic_mask             (Min: 0  Default: 0  Max: 127)
 196         Causes certain error conditions to call BUG(). Value is a bitmask;
 197         AND together the tags which represent errors which should cause panics:
 198
 199                 XFS_NO_PTAG                     0
 200                 XFS_PTAG_IFLUSH                 0x00000001
 201                 XFS_PTAG_LOGRES                 0x00000002
 202                 XFS_PTAG_AILDELETE              0x00000004
 203                 XFS_PTAG_ERROR_REPORT           0x00000008
 204                 XFS_PTAG_SHUTDOWN_CORRUPT       0x00000010
 205                 XFS_PTAG_SHUTDOWN_IOERROR       0x00000020
 206                 XFS_PTAG_SHUTDOWN_LOGERROR      0x00000040
 207
 208         This option is intended for debugging only.
 209
 210   fs.xfs.irix_symlink_mode      (Min: 0  Default: 0  Max: 1)
 211         Controls whether symlinks are created with mode 0777 (default)
 212         or whether their mode is affected by the umask (irix mode).
 213
 214   fs.xfs.irix_sgid_inherit      (Min: 0  Default: 0  Max: 1)
 215         Controls files created in SGID directories.
 216         If the group ID of the new file does not match the effective group
 217         ID or one of the supplementary group IDs of the parent dir, the
 218         ISGID bit is cleared if the irix_sgid_inherit compatibility sysctl
 219         is set.
 220
 221   fs.xfs.inherit_sync           (Min: 0  Default: 1  Max: 1)
 222         Setting this to "1" will cause the "sync" flag set
 223         by the xfs_io(8) chattr command on a directory to be
 224         inherited by files in that directory.
 225
 226   fs.xfs.inherit_nodump         (Min: 0  Default: 1  Max: 1)
 227         Setting this to "1" will cause the "nodump" flag set
 228         by the xfs_io(8) chattr command on a directory to be
 229         inherited by files in that directory.
 230
 231   fs.xfs.inherit_noatime        (Min: 0  Default: 1  Max: 1)
 232         Setting this to "1" will cause the "noatime" flag set
 233         by the xfs_io(8) chattr command on a directory to be
 234         inherited by files in that directory.
 235
 236   fs.xfs.inherit_nosymlinks     (Min: 0  Default: 1  Max: 1)
 237         Setting this to "1" will cause the "nosymlinks" flag set
 238         by the xfs_io(8) chattr command on a directory to be
 239         inherited by files in that directory.
 240
 241   fs.xfs.rotorstep              (Min: 1  Default: 1  Max: 256)
 242         In "inode32" allocation mode, this option determines how many
 243         files the allocator attempts to allocate in the same allocation
 244         group before moving to the next allocation group.  The intent
 245         is to control the rate at which the allocator moves between
 246         allocation groups when allocating extents for new files.