Switch to SpeakEasy.net DSL

The Modular Manual Browser

Home Page
Manual: (FreeBSD-5.4-RELEASE)
Apropos / Subsearch:
optional field

FFS(7)               BSD Miscellaneous Information Manual               FFS(7)

     ffs -- Berkeley fast file system

     In the kernel configuration file:
     options FFS
     options QUOTA
     options SOFTUPDATES
     options SUIDDIR
     options UFS_ACL
     options UFS_DIRHASH
     options UFS_EXTATTR

     In fstab(5):
     /dev/disk0a     /mnt ufs rw 1 1

     The Berkeley fast file system provides facilities to store file system
     data onto a disk device.  ffs has been optimized over the years for speed
     and reliability and is the default FreeBSD file system.

     options QUOTA
         This option allows system administrators to set limits on disk usage
         on a per-user basis.  Quotas can be used only on file systems mounted
         with the quota option; see quota(1) and edquota(8).

   Soft Updates
     options SOFTUPDATES
         The soft updates feature tracks writes to the disk and enforces meta-
         data update dependencies (e.g., updating free block maps) to ensure
         that the file system remains consistent.

         To enable soft updates on an unmounted file system, use the following

               tunefs -n enable fs

         fs can be either a mount point listed in fstab(5) (e.g., /usr), or a
         disk device (e.g., /dev/da0a).

   File Ownership Inheritance
     options SUIDDIR
         For use in file sharing environments on networks including Microsoft
         Windows and Apple Macintosh computers, this option allows files on
         file systems mounted with the suiddir option to inherit the ownership
         of its directory, i.e., ``if it's my directory, it must be my file.''

   Access Control Lists
     options UFS_ACL
         Access control lists allow the association of fine-grained discre-
         tionary access control information with files and directories.  This
         option requires the presence of the UFS_EXTATTR option, and it is
         recommended that UFS_EXTATTR_AUTOSTART is included as well, so that
         ACLs are enabled atomically upon mounting the file system.

     In order to enable support for ACLs, two extended attributes must be
     available in the EXTATTR_NAMESPACE_SYSTEM namespace: posix1e.acl_access,
     which holds the access ACL, and posix1e.acl_default, which holds the
     default ACL for directories.  If you are using file system extended
     attributes, the following commands may be used to allocate space for and
     create the necessary EA backing files for ACLs in the root of each file
     system.  In these examples, the root file system is used; see Extended
     Attributes for more details.

           mkdir -p /.attribute/system
           cd /.attribute/system
           extattrctl initattr -p / 388 posix1e.acl_access
           extattrctl initattr -p / 388 posix1e.acl_default

     On the next mount of the root file system, the attributes will be auto-
     matically started (if UFS_EXTATTR_AUTOSTART is included in the kernel
     configuration), and ACLs will be enabled.

   Directory Hashing
     options UFS_DIRHASH
         Implements a hash-based lookup scheme for directories in order to
         speed up accesses to very large directories.

   Extended Attributes
     options UFS_EXTATTR
         Extended attributes allow the association of additional arbitrary
         metadata with files and directories, which can be assigned and
         retrieved from userland as well as from within the kernel; see

         If this option is defined, ffs will search for a .attribute subdirec-
         tory of the file system root during the mount operation.  If found,
         extended attribute support will be automatically started for that
         file system.

     The following sysctl(8) MIBs are defined for use with ffs:

     vfs.ffs.doasyncfree   Asynchronously write out modified i-node and indi-
                           rect blocks upon reallocating file system blocks to
                           be contiguous.  (Default: 1.)

     vfs.ffs.doreallocblks Enable support for the rearrangement of blocks to
                           be contiguous.  (Default: 1.)

     quota(1), acl(3), extattr(3), edquota(8), extattrctl(8), sysctl(8)

     M. McKusick, W. Joy, S. Lefler, and R. Fabry, "A Fast File System for
     UNIX", ACM Transactions on Computer Systems, 2, 3, 181-197, August 1984.

     M. McKusick, "Soft Updates: A Technique for Eliminating Most Synchronous
     Writes in the Fast Filesystem", Proceedings of the Freenix Track at the
     1999 Usenix Annual Technical Conference, 71-84, June 2000.

BSD                            December 26, 2001                           BSD