118 lines · plain
1.. SPDX-License-Identifier: GPL-2.02 3================4OCFS2 filesystem5================6 7OCFS2 is a general purpose extent based shared disk cluster file8system with many similarities to ext3. It supports 64 bit inode9numbers, and has automatically extending metadata groups which may10also make it attractive for non-clustered use.11 12You'll want to install the ocfs2-tools package in order to at least13get "mount.ocfs2" and "ocfs2_hb_ctl".14 15Project web page: http://ocfs2.wiki.kernel.org16Tools git tree: https://github.com/markfasheh/ocfs2-tools17OCFS2 mailing lists: https://subspace.kernel.org/lists.linux.dev.html18 19All code copyright 2005 Oracle except when otherwise noted.20 21Credits22=======23 24Lots of code taken from ext3 and other projects.25 26Authors in alphabetical order:27 28- Joel Becker <joel.becker@oracle.com>29- Zach Brown <zach.brown@oracle.com>30- Mark Fasheh <mfasheh@suse.com>31- Kurt Hackel <kurt.hackel@oracle.com>32- Tao Ma <tao.ma@oracle.com>33- Sunil Mushran <sunil.mushran@oracle.com>34- Manish Singh <manish.singh@oracle.com>35- Tiger Yang <tiger.yang@oracle.com>36 37Caveats38=======39Features which OCFS2 does not support yet:40 41 - Directory change notification (F_NOTIFY)42 - Distributed Caching (F_SETLEASE/F_GETLEASE/break_lease)43 44Mount options45=============46 47OCFS2 supports the following mount options:48 49(*) == default50 51======================= ========================================================52barrier=1 This enables/disables barriers. barrier=0 disables it,53 barrier=1 enables it.54errors=remount-ro(*) Remount the filesystem read-only on an error.55errors=panic Panic and halt the machine if an error occurs.56intr (*) Allow signals to interrupt cluster operations.57nointr Do not allow signals to interrupt cluster58 operations.59noatime Do not update access time.60relatime(*) Update atime if the previous atime is older than61 mtime or ctime62strictatime Always update atime, but the minimum update interval63 is specified by atime_quantum.64atime_quantum=60(*) OCFS2 will not update atime unless this number65 of seconds has passed since the last update.66 Set to zero to always update atime. This option need67 work with strictatime.68data=ordered (*) All data are forced directly out to the main file69 system prior to its metadata being committed to the70 journal.71data=writeback Data ordering is not preserved, data may be written72 into the main file system after its metadata has been73 committed to the journal.74preferred_slot=0(*) During mount, try to use this filesystem slot first. If75 it is in use by another node, the first empty one found76 will be chosen. Invalid values will be ignored.77commit=nrsec (*) Ocfs2 can be told to sync all its data and metadata78 every 'nrsec' seconds. The default value is 5 seconds.79 This means that if you lose your power, you will lose80 as much as the latest 5 seconds of work (your81 filesystem will not be damaged though, thanks to the82 journaling). This default value (or any low value)83 will hurt performance, but it's good for data-safety.84 Setting it to 0 will have the same effect as leaving85 it at the default (5 seconds).86 Setting it to very large values will improve87 performance.88localalloc=8(*) Allows custom localalloc size in MB. If the value is too89 large, the fs will silently revert it to the default.90localflocks This disables cluster aware flock.91inode64 Indicates that Ocfs2 is allowed to create inodes at92 any location in the filesystem, including those which93 will result in inode numbers occupying more than 3294 bits of significance.95user_xattr (*) Enables Extended User Attributes.96nouser_xattr Disables Extended User Attributes.97acl Enables POSIX Access Control Lists support.98noacl (*) Disables POSIX Access Control Lists support.99resv_level=2 (*) Set how aggressive allocation reservations will be.100 Valid values are between 0 (reservations off) to 8101 (maximum space for reservations).102dir_resv_level= (*) By default, directory reservations will scale with file103 reservations - users should rarely need to change this104 value. If allocation reservations are turned off, this105 option will have no effect.106coherency=full (*) Disallow concurrent O_DIRECT writes, cluster inode107 lock will be taken to force other nodes drop cache,108 therefore full cluster coherency is guaranteed even109 for O_DIRECT writes.110coherency=buffered Allow concurrent O_DIRECT writes without EX lock among111 nodes, which gains high performance at risk of getting112 stale data on other nodes.113journal_async_commit Commit block can be written to disk without waiting114 for descriptor blocks. If enabled older kernels cannot115 mount the device. This will enable 'journal_checksum'116 internally.117======================= ========================================================118