Documentation/filesystems/ext3.txt

   1
   2 Ext3 Filesystem
   3 ===============
   4
   5 Ext3 was originally released in September 1999. Written by Stephen Tweedie
   6 for the 2.2 branch, and ported to 2.4 kernels by Peter Braam, Andreas Dilger,
   7 Andrew Morton, Alexander Viro, Ted Ts'o and Stephen Tweedie.
   8
   9 Ext3 is the ext2 filesystem enhanced with journalling capabilities.
  10
  11 Options
  12 =======
  13
  14 When mounting an ext3 filesystem, the following option are accepted:
  15 (*) == default
  16
  17 ro                      Mount filesystem read only. Note that ext3 will replay
  18                         the journal (and thus write to the partition) even when
  19                         mounted "read only". Mount options "ro,noload" can be
  20                         used to prevent writes to the filesystem.
  21
  22 journal=update          Update the ext3 file system's journal to the current
  23                         format.
  24
  25 journal=inum            When a journal already exists, this option is ignored.
  26                         Otherwise, it specifies the number of the inode which
  27                         will represent the ext3 file system's journal file.
  28
  29 journal_path=path
  30 journal_dev=devnum      When the external journal device's major/minor numbers
  31                         have changed, these options allow the user to specify
  32                         the new journal location.  The journal device is
  33                         identified through either its new major/minor numbers
  34                         encoded in devnum, or via a path to the device.
  35
  36 norecovery              Don't load the journal on mounting. Note that this forces
  37 noload                  mount of inconsistent filesystem, which can lead to
  38                         various problems.
  39
  40 data=journal            All data are committed into the journal prior to being
  41                         written into the main file system.
  42
  43 data=ordered    (*)     All data are forced directly out to the main file
  44                         system prior to its metadata being committed to the
  45                         journal.
  46
  47 data=writeback          Data ordering is not preserved, data may be written
  48                         into the main file system after its metadata has been
  49                         committed to the journal.
  50
  51 commit=nrsec    (*)     Ext3 can be told to sync all its data and metadata
  52                         every 'nrsec' seconds. The default value is 5 seconds.
  53                         This means that if you lose your power, you will lose
  54                         as much as the latest 5 seconds of work (your
  55                         filesystem will not be damaged though, thanks to the
  56                         journaling).  This default value (or any low value)
  57                         will hurt performance, but it's good for data-safety.
  58                         Setting it to 0 will have the same effect as leaving
  59                         it at the default (5 seconds).
  60                         Setting it to very large values will improve
  61                         performance.
  62
  63 barrier=<0|1(*)>        This enables/disables the use of write barriers in
  64 barrier (*)             the jbd code.  barrier=0 disables, barrier=1 enables.
  65 nobarrier               This also requires an IO stack which can support
  66                         barriers, and if jbd gets an error on a barrier
  67                         write, it will disable again with a warning.
  68                         Write barriers enforce proper on-disk ordering
  69                         of journal commits, making volatile disk write caches
  70                         safe to use, at some performance penalty.  If
  71                         your disks are battery-backed in one way or another,
  72                         disabling barriers may safely improve performance.
  73                         The mount options "barrier" and "nobarrier" can
  74                         also be used to enable or disable barriers, for
  75                         consistency with other ext3 mount options.
  76
  77 user_xattr              Enables Extended User Attributes.  Additionally, you
  78                         need to have extended attribute support enabled in the
  79                         kernel configuration (CONFIG_EXT3_FS_XATTR).  See the
  80                         attr(5) manual page and http://acl.bestbits.at/ to
  81                         learn more about extended attributes.
  82
  83 nouser_xattr            Disables Extended User Attributes.
  84
  85 acl                     Enables POSIX Access Control Lists support.
  86                         Additionally, you need to have ACL support enabled in
  87                         the kernel configuration (CONFIG_EXT3_FS_POSIX_ACL).
  88                         See the acl(5) manual page and http://acl.bestbits.at/
  89                         for more information.
  90
  91 noacl                   This option disables POSIX Access Control List
  92                         support.
  93
  94 reservation
  95
  96 noreservation
  97
  98 bsddf           (*)     Make 'df' act like BSD.
  99 minixdf                 Make 'df' act like Minix.
 100
 101 check=none              Don't do extra checking of bitmaps on mount.
 102 nocheck
 103
 104 debug                   Extra debugging information is sent to syslog.
 105
 106 errors=remount-ro       Remount the filesystem read-only on an error.
 107 errors=continue         Keep going on a filesystem error.
 108 errors=panic            Panic and halt the machine if an error occurs.
 109                         (These mount options override the errors behavior
 110                         specified in the superblock, which can be
 111                         configured using tune2fs.)
 112
 113 data_err=ignore(*)      Just print an error message if an error occurs
 114                         in a file data buffer in ordered mode.
 115 data_err=abort          Abort the journal if an error occurs in a file
 116                         data buffer in ordered mode.
 117
 118 grpid                   Give objects the same group ID as their creator.
 119 bsdgroups
 120
 121 nogrpid         (*)     New objects have the group ID of their creator.
 122 sysvgroups
 123
 124 resgid=n                The group ID which may use the reserved blocks.
 125
 126 resuid=n                The user ID which may use the reserved blocks.
 127
 128 sb=n                    Use alternate superblock at this location.
 129
 130 quota                   These options are ignored by the filesystem. They
 131 noquota                 are used only by quota tools to recognize volumes
 132 grpquota                where quota should be turned on. See documentation
 133 usrquota                in the quota-tools package for more details
 134                         (http://sourceforge.net/projects/linuxquota).
 135
 136 jqfmt=<quota type>      These options tell filesystem details about quota
 137 usrjquota=<file>        so that quota information can be properly updated
 138 grpjquota=<file>        during journal replay. They replace the above
 139                         quota options. See documentation in the quota-tools
 140                         package for more details
 141                         (http://sourceforge.net/projects/linuxquota).
 142
 143 Specification
 144 =============
 145 Ext3 shares all disk implementation with the ext2 filesystem, and adds
 146 transactions capabilities to ext2.  Journaling is done by the Journaling Block
 147 Device layer.
 148
 149 Journaling Block Device layer
 150 -----------------------------
 151 The Journaling Block Device layer (JBD) isn't ext3 specific.  It was designed
 152 to add journaling capabilities to a block device.  The ext3 filesystem code
 153 will inform the JBD of modifications it is performing (called a transaction).
 154 The journal supports the transactions start and stop, and in case of a crash,
 155 the journal can replay the transactions to quickly put the partition back into
 156 a consistent state.
 157
 158 Handles represent a single atomic update to a filesystem.  JBD can handle an
 159 external journal on a block device.
 160
 161 Data Mode
 162 ---------
 163 There are 3 different data modes:
 164
 165 * writeback mode
 166 In data=writeback mode, ext3 does not journal data at all.  This mode provides
 167 a similar level of journaling as that of XFS, JFS, and ReiserFS in its default
 168 mode - metadata journaling.  A crash+recovery can cause incorrect data to
 169 appear in files which were written shortly before the crash.  This mode will
 170 typically provide the best ext3 performance.
 171
 172 * ordered mode
 173 In data=ordered mode, ext3 only officially journals metadata, but it logically
 174 groups metadata and data blocks into a single unit called a transaction.  When
 175 it's time to write the new metadata out to disk, the associated data blocks
 176 are written first.  In general, this mode performs slightly slower than
 177 writeback but significantly faster than journal mode.
 178
 179 * journal mode
 180 data=journal mode provides full data and metadata journaling.  All new data is
 181 written to the journal first, and then to its final location.
 182 In the event of a crash, the journal can be replayed, bringing both data and
 183 metadata into a consistent state.  This mode is the slowest except when data
 184 needs to be read from and written to disk at the same time where it
 185 outperforms all other modes.
 186
 187 Compatibility
 188 -------------
 189
 190 Ext2 partitions can be easily convert to ext3, with `tune2fs -j <dev>`.
 191 Ext3 is fully compatible with Ext2.  Ext3 partitions can easily be mounted as
 192 Ext2.
 193
 194
 195 External Tools
 196 ==============
 197 See manual pages to learn more.
 198
 199 tune2fs:        create a ext3 journal on a ext2 partition with the -j flag.
 200 mke2fs:         create a ext3 partition with the -j flag.
 201 debugfs:        ext2 and ext3 file system debugger.
 202 ext2online:     online (mounted) ext2 and ext3 filesystem resizer
 203
 204
 205 References
 206 ==========
 207
 208 kernel source:  <file:fs/ext3/>
 209                 <file:fs/jbd/>
 210
 211 programs:       http://e2fsprogs.sourceforge.net/
 212                 http://ext2resize.sourceforge.net
 213
 214 useful links:   http://www.ibm.com/developerworks/library/l-fs7/index.html
 215         http://www.ibm.com/developerworks/library/l-fs8/index.html