autofsd(1M)                                                        autofsd(1M)


NAME
     autofsd - autoFS mount/unmount daemon

SYNOPSIS
     autofsd [ -vTL ] [ -p priority ] [ -D name=value ]
             [ -tp duration ] [ -tm duration ] [ -tn duration ]
             [-m maxthreads ]  [-E user ] [ -P tcp|udp ] [ -u umask ]

DESCRIPTION
     autofsd is an RPC server that answers filesystem mount and unmount
     requests from the AutoFS filesystem.  It uses local files or name service
     maps to locate filesystems to be mounted.  These maps are described with
     the autofs(1M) command.

     The autofsd daemon is started during system initialization from the
     /etc/init.d/network script if the configuration flags nfs and autofs are
     set on.  Type the following commands as superuser to enable AutoFS:

          /etc/chkconfig nfs on
          /etc/chkconfig autofs on

     Rebooting invokes the autofs command with options from
     /etc/config/autofs.option and starts the autofsd daemon with the options
     contained in the file /etc/config/autofsd.options.  The autofs command
     passes mount point configuration information found in the file
     /etc/auto_master to autofsd.  Do not kill autofsd to change the AutoFS
     configuration.  Run the autofs command to cause changes in the
     configuration maps to take affect.

OPTIONS
     -v              Verbose.  Log status messages to the console.

     -T              Trace.  Expand each RPC call and display it on the
                     standard output.

     -p priority     Set the priority of the daemon process.  The daemon will
                     run at the priority specified, with a scheduling policy
                     of SCHED_TS. The default priority is 30.

     -tp duration    Specify the duration, in seconds, that the results of a
                     probe of a server's availability will remain cached.  The
                     default is 5 seconds.

     -tm duration    To assure robust compatability, autofsd if necessary will
                     attempt to contact a remote server using all supported
                     mount versions and protocols until the server
                     successfully responds, or the all attempts have been
                     exhausted.  This parameter affects the time out for each
                     attempt.  The default timeout per attempt is 20 seconds.


     -tn duration    Timeout to use when attempting to acquire NFS file handle
                     for the root of the mounted filesystem. Default is 2
                     seconds.

     -D name=value   Assign value to the indicated AutoFS map substitution
                     variable.  These assignments cannot be used to substitute
                     variables in the master map, auto_master.

     -m maxthreads   Set the maximum number of threads that autofsd will
                     create to handle autofs mount or unmount requests.
                     autofsd will create the threads only when needed. At this
                     time the user cannot specify a number greater than 16 for
                     maxthreads. The default is to run autofsd multi-threaded
                     with the maximum, 16 threads. Setting this option to the
                     value 1 will cause autofsd to run as a single threaded
                     process.

     -E user         enable processing of executable autofs maps and execute
                     the map's file as user.

                     For security reasons, unless the user is explicitly
                     specified by using this option, autofsd will not attempt
                     to execute a map file which has the executable bit set
                     but will process the map as standard file map.  If the
                     user is specified then autofsd will change effective user
                     ID of the process to the ID of the user before executing
                     the map.

                     The executable map file must be owned by root, must be
                     protected from being modified by users other then root,
                     must be executable by the user specified in -E option and
                     must exit with status of 0 on success. If the file is a
                     script then it must also be readble by the user under
                     whose ID it will be executed.

     -P tcp|udp      autofsd will default to do all nfs mounts with the IP
                     protocol specified here. The user can override the
                     protocol specified here by using the proto=xyz mount
                     option in the autofs maps. The currently allowed options
                     are only tcp and udp. If this (-P) option is not used,
                     the default is to do all nfs mounts with udp. If the tcp
                     option is specified and the remote nfs server does not
                     support nfs over tcp, autofsd will retry the mount over
                     udp.

     -u umask        Change umask of the autofsd' process to umask. Default is
                     0022.

     -L              Use symlinks instead of LoFS mounts when accessing
                     directories on the local host.


FILES
     /etc/auto_master             master map for AutoFS
     /etc/config/autofs.options   options and arguments for the autofs command
     /etc/config/autofsd.options  options and arguments for the autofsd daemon

SEE ALSO
     autofs(1M), chkconfig(1M), mount(1M), sched_setscheduler(2).

CAVEATS
     For security reasons, the user should be extremely wary using the -E
     option. It's use is deprecated. Support for executable autofs maps may go
     away in the future.


                                                                        Page 3