Dear all,
I have a question regarding my openldap DIT design. My design so far is
based on the model: ou=people,dc=example,dc=com. It is very possible that
I'll have to be able to find attributes of people belonging to some specific
group (eg, student, postgrad, etc). The easiest way to address this issue
for me would be to branch my DIT like this:
ou=undergrads,ou=people,dc=example,dc=com and
ou=postgrads,ou=people,dc=example,dc=com. On the other hand, I have several
classes that I would like to distinguish my users to apart from this (like
stuff, student, professors, etc.) but further sub-brunching shows to me that
there's something wrong with my design (since those classes may dynamically
change in the future).
As a second solution I thought that it would be very easy to make my users
in ou=people,dc=example,dc=com belong to some group located in
ou=groups,dc=example,dc=com. This way I feel much more flexible in making
such classifications, but my problem is how to formulate ldapsearch filters
so as to return an attribute of some user only if the specific user belongs
to one or more of my groups (for example to find all email accounts from my
people that belong to the undergrads group).