Program Field; Table A-2: User Id Mapping Options - HP LCS60 Installation Manual

Network interface for ethernet r3.0
Table of Contents

Advertisement

srvtab

Table A-2: User ID Mapping Options

Option
*n, *o
Use the numeric user ID supplied in the call request information. The *o means
the user ID should be interpreted as an octal number. The *n indicates the user
ID should be treated as a self-determining number. For example, an initial 0x or
0X indicates hexadecimal, and an initial 0 indicates octal.
The *n and *o user ID formats will only provide a valid match if an /etc/passwd
file entry exists with the same numerical user ID and the password has not
expired. The group-id is determined from that password file entry.
&
Translate the supplied user ID and group-id names using the dkuidtab file. This
type of entry matches only those user IDs which have entries in the dkuidtab file.
This facility allows a user with a user ID on one host to be preauthorized as a
user with a different numerical user ID on another host. This preauthorization is
performed by the authorize command.
uid, uid
The previous two forms can be further restricted to a range of user IDs by
appending uid or uid to the field. This restricts the incoming user ID to be less
than (or greater than) the specified decimal number. For example, *n 0 prevents
root (user ID 0) from matching the line. Only one modifier may be appended to
an entry.
[login]
This type of entry provides a fixed login name that is used for all matching call
requests. The [login] user ID format will return a valid match as long as a valid
/etc/passwd entry exists for login. Fixed login name specifications are useful for
assigning a single uucp login to a group of originating sites. The facility is also
useful when invoking authorization type services that require a fixed set of per-
missions. The [login] user ID format can be thought of as a form of setuid facility,
since all call requests are mapped to the same login ID.

Program Field

The program field of a server table entry contains the pathname of the program to
be executed. The field may contain a %s which will be replaced by the pathname
of the user's shell as obtained from the /etc/passwd file.
The server uses the execv library function when invoking programs, so only paths
to binary executables may be specified in the program field of a server table entry.
The path should be fully specified (for example, /usr/lbin/program) for each entry
in the server table.
A-6
Description
Issue 3

Advertisement

Table of Contents
loading

Table of Contents