|
Security
Adding Person Entries to the Directory
Before you define person entries, you should have already
started the directory server, updated the schema, and
set the basic access control. (For details, see Setting up an LDAP Directory Server.)
Person entries are needed in order to make the directory useful to SAS applications.
For example, when you update access control, access decisions are based
on the Distinguished Name (DN) that the person binds to the directory.
SAS software also uses person entries to identify users and to obtain information
such as user ID and e-mail address. Some of the options for user data are
object class, directory structure, and DN.
To add person entries to the directory, follow these steps:
Select an object class to use for the entries.
A common choice is the inetOrgPerson class, which accepts many
useful attributes. If you need to add attributes
to your person entries and the attributes are not allowed by inetOrgPerson ,
you can create your own object class using inetOrgPerson as a parent class.
Enter the person entries in the directory. Follow these
guidelines to help your person data work better with SAS software:
Keep common names unique. Some SAS applications use the common
name when associating a person entry with other entries in the SAS
application entries.
Include the user ID and e-mail address in the person entry. Applications
need to look up the user ID.
When you load the directory with person entries for the first time, add a
default userpassword attribute. This attribute allows users
to bind to this DN when they use the directory.
Decide how the person data is laid out in the directory.
The two most popular options are as follows:
- Flat structure
puts all of the data in one place in
the directory. The benefit is that you do not have to move the entries
if users change organizations within the company.
- Organizational unit structure
places the
entries in a subtree according to the organizational unit within the company.
This structure can resemble the
company's organization, which allows you to visualize the
relationships between entries.
Decide on the structure of the distinguished names for your person entries.
Although your selection of the attribute for the relative distinguished
name is not critical, you must be consistent. Two
acceptable choices are common name and user ID. If you use a flat structure for
the person data, then use user ID for the DN, because common names
are duplicated more often than user IDs.
|