1 2The SGI XFS Filesystem 3====================== 4 5XFS is a high performance journaling filesystem which originated 6on the SGI IRIX platform. It is completely multi-threaded, can 7support large files and large filesystems, extended attributes, 8variable block sizes, is extent based, and makes extensive use of 9Btrees (directories, extents, free space) to aid both performance 10and scalability. 11 12Refer to the documentation at http://oss.sgi.com/projects/xfs/ 13for further details. This implementation is on-disk compatible 14with the IRIX version of XFS. 15 16 17Mount Options 18============= 19 20When 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 discard 43 Issue command to let the block device reclaim space freed by the 44 filesystem. This is useful for SSD devices, thinly provisioned 45 LUNs and virtual machine images, but may have a performance 46 impact. This option is incompatible with the nodelaylog option. 47 48 dmapi 49 Enable the DMAPI (Data Management API) event callouts. 50 Use with the "mtpt" option. 51 52 grpid/bsdgroups and nogrpid/sysvgroups 53 These options define what group ID a newly created file gets. 54 When grpid is set, it takes the group ID of the directory in 55 which it is created; otherwise (the default) it takes the fsgid 56 of the current process, unless the directory has the setgid bit 57 set, in which case it takes the gid from the parent directory, 58 and also gets the setgid bit set if it is a directory itself. 59 60 ihashsize=value 61 In memory inode hashes have been removed, so this option has 62 no function as of August 2007. Option is deprecated. 63 64 ikeep/noikeep 65 When ikeep is specified, XFS does not delete empty inode clusters 66 and keeps them around on disk. ikeep is the traditional XFS 67 behaviour. When noikeep is specified, empty inode clusters 68 are returned to the free space pool. The default is noikeep for 69 non-DMAPI mounts, while ikeep is the default when DMAPI is in use. 70 71 inode64 72 Indicates that XFS is allowed to create inodes at any location 73 in the filesystem, including those which will result in inode 74 numbers occupying more than 32 bits of significance. This is 75 provided for backwards compatibility, but causes problems for 76 backup applications that cannot handle large inode numbers. 77 78 largeio/nolargeio 79 If "nolargeio" is specified, the optimal I/O reported in 80 st_blksize by stat(2) will be as small as possible to allow user 81 applications to avoid inefficient read/modify/write I/O. 82 If "largeio" specified, a filesystem that has a "swidth" specified 83 will return the "swidth" value (in bytes) in st_blksize. If the 84 filesystem does not have a "swidth" specified but does specify 85 an "allocsize" then "allocsize" (in bytes) will be returned 86 instead. 87 If neither of these two options are specified, then filesystem 88 will behave as if "nolargeio" was specified. 89 90 logbufs=value 91 Set the number of in-memory log buffers. Valid numbers range 92 from 2-8 inclusive. 93 The default value is 8 buffers for filesystems with a 94 blocksize of 64KiB, 4 buffers for filesystems with a blocksize 95 of 32KiB, 3 buffers for filesystems with a blocksize of 16KiB 96 and 2 buffers for all other configurations. Increasing the 97 number of buffers may increase performance on some workloads 98 at the cost of the memory used for the additional log buffers 99 and their associated control structures. 100 101 logbsize=value 102 Set the size of each in-memory log buffer. 103 Size may be specified in bytes, or in kilobytes with a "k" suffix. 104 Valid sizes for version 1 and version 2 logs are 16384 (16k) and 105 32768 (32k). Valid sizes for version 2 logs also include 106 65536 (64k), 131072 (128k) and 262144 (256k). 107 The default value for machines with more than 32MiB of memory 108 is 32768, machines with less memory use 16384 by default. 109 110 logdev=device and rtdev=device 111 Use an external log (metadata journal) and/or real-time device. 112 An XFS filesystem has up to three parts: a data section, a log 113 section, and a real-time section. The real-time section is 114 optional, and the log section can be separate from the data 115 section or contained within it. 116 117 mtpt=mountpoint 118 Use with the "dmapi" option. The value specified here will be 119 included in the DMAPI mount event, and should be the path of 120 the actual mountpoint that is used. 121 122 noalign 123 Data allocations will not be aligned at stripe unit boundaries. 124 125 noatime 126 Access timestamps are not updated when a file is read. 127 128 norecovery 129 The filesystem will be mounted without running log recovery. 130 If the filesystem was not cleanly unmounted, it is likely to 131 be inconsistent when mounted in "norecovery" mode. 132 Some files or directories may not be accessible because of this. 133 Filesystems mounted "norecovery" must be mounted read-only or 134 the mount will fail. 135 136 nouuid 137 Don't check for double mounted file systems using the file system uuid. 138 This is useful to mount LVM snapshot volumes. 139 140 uquota/usrquota/uqnoenforce/quota 141 User disk quota accounting enabled, and limits (optionally) 142 enforced. Refer to xfs_quota(8) for further details. 143 144 gquota/grpquota/gqnoenforce 145 Group disk quota accounting enabled and limits (optionally) 146 enforced. Refer to xfs_quota(8) for further details. 147 148 pquota/prjquota/pqnoenforce 149 Project disk quota accounting enabled and limits (optionally) 150 enforced. Refer to xfs_quota(8) for further details. 151 152 sunit=value and swidth=value 153 Used to specify the stripe unit and width for a RAID device or 154 a stripe volume. "value" must be specified in 512-byte block 155 units. 156 If this option is not specified and the filesystem was made on 157 a stripe volume or the stripe width or unit were specified for 158 the RAID device at mkfs time, then the mount system call will 159 restore the value from the superblock. For filesystems that 160 are made directly on RAID devices, these options can be used 161 to override the information in the superblock if the underlying 162 disk layout changes after the filesystem has been created. 163 The "swidth" option is required if the "sunit" option has been 164 specified, and must be a multiple of the "sunit" value. 165 166 swalloc 167 Data allocations will be rounded up to stripe width boundaries 168 when the current end of file is being extended and the file 169 size is larger than the stripe width size. 170 171 172sysctls 173======= 174 175The following sysctls are available for the XFS filesystem: 176 177 fs.xfs.stats_clear (Min: 0 Default: 0 Max: 1) 178 Setting this to "1" clears accumulated XFS statistics 179 in /proc/fs/xfs/stat. It then immediately resets to "0". 180 181 fs.xfs.xfssyncd_centisecs (Min: 100 Default: 3000 Max: 720000) 182 The interval at which the xfssyncd thread flushes metadata 183 out to disk. This thread will flush log activity out, and 184 do some processing on unlinked inodes. 185 186 fs.xfs.xfsbufd_centisecs (Min: 50 Default: 100 Max: 3000) 187 The interval at which xfsbufd scans the dirty metadata buffers list. 188 189 fs.xfs.age_buffer_centisecs (Min: 100 Default: 1500 Max: 720000) 190 The age at which xfsbufd flushes dirty metadata buffers to disk. 191 192 fs.xfs.error_level (Min: 0 Default: 3 Max: 11) 193 A volume knob for error reporting when internal errors occur. 194 This will generate detailed messages & backtraces for filesystem 195 shutdowns, for example. Current threshold values are: 196 197 XFS_ERRLEVEL_OFF: 0 198 XFS_ERRLEVEL_LOW: 1 199 XFS_ERRLEVEL_HIGH: 5 200 201 fs.xfs.panic_mask (Min: 0 Default: 0 Max: 127) 202 Causes certain error conditions to call BUG(). Value is a bitmask; 203 AND together the tags which represent errors which should cause panics: 204 205 XFS_NO_PTAG 0 206 XFS_PTAG_IFLUSH 0x00000001 207 XFS_PTAG_LOGRES 0x00000002 208 XFS_PTAG_AILDELETE 0x00000004 209 XFS_PTAG_ERROR_REPORT 0x00000008 210 XFS_PTAG_SHUTDOWN_CORRUPT 0x00000010 211 XFS_PTAG_SHUTDOWN_IOERROR 0x00000020 212 XFS_PTAG_SHUTDOWN_LOGERROR 0x00000040 213 214 This option is intended for debugging only. 215 216 fs.xfs.irix_symlink_mode (Min: 0 Default: 0 Max: 1) 217 Controls whether symlinks are created with mode 0777 (default) 218 or whether their mode is affected by the umask (irix mode). 219 220 fs.xfs.irix_sgid_inherit (Min: 0 Default: 0 Max: 1) 221 Controls files created in SGID directories. 222 If the group ID of the new file does not match the effective group 223 ID or one of the supplementary group IDs of the parent dir, the 224 ISGID bit is cleared if the irix_sgid_inherit compatibility sysctl 225 is set. 226 227 fs.xfs.inherit_sync (Min: 0 Default: 1 Max: 1) 228 Setting this to "1" will cause the "sync" flag set 229 by the xfs_io(8) chattr command on a directory to be 230 inherited by files in that directory. 231 232 fs.xfs.inherit_nodump (Min: 0 Default: 1 Max: 1) 233 Setting this to "1" will cause the "nodump" flag set 234 by the xfs_io(8) chattr command on a directory to be 235 inherited by files in that directory. 236 237 fs.xfs.inherit_noatime (Min: 0 Default: 1 Max: 1) 238 Setting this to "1" will cause the "noatime" flag set 239 by the xfs_io(8) chattr command on a directory to be 240 inherited by files in that directory. 241 242 fs.xfs.inherit_nosymlinks (Min: 0 Default: 1 Max: 1) 243 Setting this to "1" will cause the "nosymlinks" flag set 244 by the xfs_io(8) chattr command on a directory to be 245 inherited by files in that directory. 246 247 fs.xfs.rotorstep (Min: 1 Default: 1 Max: 256) 248 In "inode32" allocation mode, this option determines how many 249 files the allocator attempts to allocate in the same allocation 250 group before moving to the next allocation group. The intent 251 is to control the rate at which the allocator moves between 252 allocation groups when allocating extents for new files. 253