Unknown option: "-1" Unix manual page for dscl. (host=minya system=Darwin)
dscl(1)                   BSD General Commands Manual                  dscl(1)

NAME
     dscl -- Directory Service command line utility

SYNOPSIS
     dscl [options] [datasource [command]]

          options:
                -p           prompt for password
                -u user      authenticate as user
                -P password  authentication password
                -f filepath  targeted local node database file path
                -raw         don't strip off prefix from DirectoryService API
                             constants
                -plist       print out record(s) or attribute(s) in XML plist
                             format
                -url         print record attribute values in URL-style encod-
                             ing
                -q           quiet - no interactive prompt

          commands:
                -read [path [key ...]]
                -readall [path [key ...]]
                -readpl path key plist_path
                -readpli path key value_index plist_path
                -list path [key]
                -search path key val
                -create record_path [key [val ...]]
                -createpl record_path key plist_path val1 [val2 ...]
                -createpli record_path key value_index plist_path val1 [val2
                ...]
                -append record_path key val ...
                -merge record_path key val ...
                -delete path [key [val ...]]
                -deletepl record_path key plist_path [val ...]
                -deletepli record_path key value_index plist_path [val ...]
                -change record_path key old_val new_val
                -changei record_path key val_index new_val
                -diff path1 path2 [key ...]
                -passwd user_path [new_password | old_password new_password]

          available only in interactive mode:
                -cd dir
                -pushd [dir]
                -popd
                -auth [user [password]]
                -authonly [user [password]]
                -quit

DESCRIPTION
     dscl is a general-purpose utility for operating on Directory Service
     directory nodes.  Its commands allow one to create, read, and manage
     Directory Service data.  If invoked without any commands, dscl runs in an
     interactive mode, reading commands from standard input.  Interactive pro-
     cessing is terminated by the quit command.  Leading dashes ("-") are
     optional for all commands.

     dscl operates on a datasource specified on the command line.  This may be
     a node name or a Mac OS X Server (10.2 or later) host specified by DNS
     hostname or IP address.  Node names may be absolute paths beginning with
     a slash ("/"), or relative domain paths beginning with a dot (".") char-
     acter, which specifies the local domain, or "..", specifying the local
     domain's parent.  If the hostname or IP address form is used then the
     user must specify the -u option and either the -P of -p options to spec-
     ify an administrative user and password on the remote host to authenti-
     cate with to the remote host. The exception to this is if "localhost" is
     specified.  Passing passwords on the command line is inherently insecure
     and can cause password exposure.  For better security do not provide the
     password as part of the command and you will be securely prompted.

     The datasource may also be specified as "localonly" in which case a sepa-
     rate DirectoryService daemon process is activated which contains only the
     Local plugin for use by dscl.  If no file path is provided then access
     goes only to the registered local nodes on the system. However, if the -f
     option is specified then access is added to the local node "/Local/Tar-
     get" which points to the database located at the provided filepath. One
     example is to provide the filepath of "/Volumes/Build100/var/db/dslo-
     cal/nodes/Default" and then access to that database is provided via the
     nodename "/Local/Target".

PATH SPECIFICATION
     There are two modes of operation when specifying paths to operate on. The
     two modes correspond to whether the datasource is a node or a host. In
     the case of specifying a node, the top level of paths will be record
     types. Example top level paths would be:

           /Users/alice
           /Groups/admin

     In the case of specifying a host as a data source, the top level of paths
     correspond to Open Directory plug-ins and Search Paths. One can specify
     the plug-in to traverse to a node name, after which the paths are equiva-
     lent to the former usage. The following might be the equivalent paths as
     the above paths:

           /NetInfo/root/Users/alice
           /LDAPv3/10.0.1.42/Groups/admin

     If path components contain keys or values with embedded slash characters,
     the slash characters must be escaped with a leading backslash character.
     Since the shell also processes escape characters, an extra backslash is
     required to correctly specify an escape.  For example, to read a mount
     record with the name "ldapserver:/Users" in the "/Mounts" path, the fol-
     lowing path would be used:

           dscl  . -read /Mounts/ldaphost:\/Users

     All pathnames are case-sensitive.

COMMANDS
     The action of each command is described below.  Some commands have
     aliases.  For example, "cat" and "." are aliases for "read".  Command
     aliases are listed in parentheses.

   read (cat .)
     Usage: read [path [key ...]]

     Prints a directory.  The property key is followed by colon, then a space-
     separated list of the values for that property. If any value contains
     embedded spaces, the list will instead be displayed one entry per line,
     starting on the line after the key.

     If The -raw flag for raw output has been given, then read prints the full
     DirectoryService API constant for record and attribute types.

     If the -url flag has been specified then printed record path attribute
     values are encoded in the style of URLs. This is useful if a script or
     program is trying to process the output since values will not have any
     spaces or other control characters.

   readall
     Usage: readall [path [key ...]]

     readall prints all the records of a given type.  The output of readall is
     formatted in the same way as read with a "-" on a line as a delimeter
     between records.

   readpl
     Usage: readpl path key plist_path

     Prints the contents of plist_path.  The plist_path is followed by a
     colon, then a whitespace, and then the value for the path.

     If the plist_path is the key for a dictionary or array, the contents of
     it are displayed in plist form after the plist_path.  If plist_path is
     the key for a string, number, bool, date, or data object, only the value
     is printed out after the plist_path.

   readpli
     Usage: readpli path key value_index plist_path

     Prints the contents of plist_path for the plist at value_index of the
     key.  The plist_path is followed by a colon, then a whitespace, and then
     the value for the path.

     If the plist_path is the key for a dictionary or array, the contents of
     it are displayed in plist form after the plist_path.  If plist_path is
     the key for a string, number, bool, date, or data object, only the value
     is printed out after the plist_path.

   list (ls)
     Usage: list path

     Lists the subdirectories of the given directory.  Subdirectories are
     listed one per line.  In the case of listing a search path, the names are
     preceded by an index number that can act as a shortcut and used in place
     of the name when specifying a path.

     When used in interactive mode, the path is optional.  With no path given,
     the current directory will be used.

   search
     path key val

     Searches for records that match a pattern.  The search is rooted at the
     given path.  The path may be a node path or a record type path.  Valid
     keys are Directory Service record attribute types.

   create (mk)
     Usage: create record_path [key [val ...]]

     Creates a record, property, or value.  If only a record path is given,
     the create command will create the record if it does not exist.  If a key
     is given, then a property with that key will be created.

     WARNING - If a property with the given key already exists, it will be
     destroyed and a new property will be created in its place.  To add values
     to an existing property, use the append or merge commands.

     If values are included in the command, these values will be set for the
     given key.

     NOTE - Not all directory nodes support a property without a value. An
     error will be given if you attempt to create a property with no value in
     such a directory node.

   createpl
     Usage: createpl record_path key plist_path val1 [val2 ...]

     Creates a string, or array of strings at plist_path.

     If you are creating a value at the root of a plist that is an array, sim-
     ply use "0" as the plist_path.

     If only val1 is specified, a string will be created at plist_path.  If
     val1 val2 ... are specified, an array of strings will be created at
     plist_path.

     WARNING - If a value with the given plist_path already exists, it will be
     destroyed and a new value will be created in its place.

   createpli
     Usage: createpli record_path key value_index plist_path val1 [val2 ...]

     Creates a string, or array of strings at plist_path for the plist at
     value_index of the key.

     If you are creating a value at the root of a plist that is an array, sim-
     ply use "0" as the plist_path.

     If only val1 is specified, a string will be created at plist_path.  If
     val1 val2 ... are specified, an array of strings will be created at
     plist_path.

     WARNING - If a value with the given plist_path already exists, it will be
     destroyed and a new value will be created in its place.

   append
     Usage: append record_path key val ...

     Appends one or more values to a property in a given record.  The property
     is created if it does not exist.

   merge
     Usage: merge record_path key val ...

     Appends one or more values to a property in a given directory if the
     property does not already have those values.  The property is created if
     it does not exist.

   change
     Usage: change record_path key old_val new_val

     Replaces the given old value in the list of values of the given key with
     the new value in the specified record.

   changei
     Usage: changei path key index val

     Replaces the value at the given index in the list of values of the given
     key with the new value in the specified record.  index is an integer
     value.  An index of 1 specifies the first value.  An index greater than
     the number of values in the list will result in an error.

   diff
     Usage: diff path1 path2 key ...

     Compares the data from path1 and path2 looking at the specified keys (or
     all if no keys are specified).

   delete (rm)
     Usage: delete path [key [val ...]]

     Delete a directory, property, or value.  If a directory path is given,
     the delete command will delete the directory.  This can only be used on
     record type and record paths.  If a key is given, then a property with
     that key will be deleted.  If one or more values are given, those values
     will be removed from the property with the given key.

   deletepl
     Usage: deletepl record_path key plist_path [val ...]

     Deletes a value in a plist.  If no values are given deletepl deletes the
     plist_path.  If one or more values are given, deletepl deletes the values
     within plist_path.

   deletepli
     Usage: deletepli record_path key value_index plist_path [val ...]

     Deletes a value for the plist at value_index of the key.  If no values
     are given deletepli deletes the plist_path.  If one or more values are
     given, deletepli deletes the values within plist_path.

   passwd
     Usage: passwd user_path [new_pasword | old_password new_pasword]

     Changes a password for a user. The user must be specified by full path,
     not just a username.  If you are authenticated to the node (either by
     specifying the -u and -P flags or by using the auth command when in
     interactive node) then you can simply specify a new password.  If you are
     not authenticated or if FileVault is enabled then the user's old password
     must be specified.  If passwords are not specified while in interactive
     mode, you will be prompted for them.  Passing these passwords on the com-
     mand line is inherently insecure and can cause password exposure.  For
     better security do not provide the password as part of the command and
     you will be securely prompted.

INTERACTIVE COMMANDS
   cd
     Usage: cd dir

     Sets the current directory.  Path names for other dscl commands may be
     relative to the current directory.

   pushd (pd)
     Usage: pushd path

     Similar to the pushd command commonly found in Unix shells.  When a path
     is specified it sets the current directory while pushing the previous
     directory on to the directory stack.  If no path is specified it
     exchanges the top two elements of the directory stack.  It will also
     print the final directory stack.

   popd
     Usage: popd

     Pops the directory stack and returns to the new top directory.  It will
     also print the final directory stack.

   auth (su)
     Usage: auth [user [password]]

     Authenticate as the named user, or as "root" if no user is specified.  If
     a password is supplied, then that password is used for authentication,
     otherwise the command prompts for a password.

     If dscl is run in host mode, then when this command is run the current
     directory must be in the subdirectories of a node.

   authonly
     Usage: authonly [user [password]]

     Used to verify the password of a named user, or of "root" if no user is
     specified.  If a password is supplied, then that password is used for
     authentication, otherwise the command prompts for a password.

     If dscl is run in host mode, then when this command is run the current
     directory must be in the subdirectories of a node.

   quit (q)
     Usage: quit

     Ends processing of interactive commands and terminates the program.

   command history
     The up and down arrow keys will scan through the command history.

   tab completion
     When pathnames are being typed, pressing the tab key will result in a
     search to auto-complete the typed partial subdirectory name. It will also
     attempt to correct capitilization in the process.

EXAMPLES
     -view a record in the local directory node
              dscl . -read /Users/www

     -create or replace the UserShell attribute value for the www user record
              dscl . -create /Users/www UserShell /usr/bin/false

     -create or replace the test key of the mcx_application_data:loginwindow
              plist value for the MCXSettings attribute of the user1 user
              record
              dscl . -createpl /Users/user1 MCXSettings mcx_applica-
              tion_data:loginwindow:test value

     -list the uniqueID values for all user records on a given node
              dscl /LDAPv3/ldap.company.com -list /Users UniqueID

     -append a value that has spaces in it
              dscl . -append /Users/www Comment "This is a comment"

DIAGNOSTICS
     dscl will return -1(255) on error.

SEE ALSO
     DirectoryService(8), DirectoryServiceAttributes(7)

MacOSX                          August 25, 2003                         MacOSX