On Mon, Oct 3, 2011 at 11:37 AM, Crawford, Scott <[email protected]> wrote:
> RE: This is a fairly common model and there’s not really anything wrong with
> it.
> Why is it such a common model? I see it a lot too, and it always frustrates
> me that they don’t simply use the credentials supplied by the user to
> attempt to bind to AD.

  I'm confident laziness is the most common reason, but for a minority
of cases, some scenarios have a need to identify things in the
directory independently of a user doing something against that
directory.  In other words, the server needs to look something up
without having a user's credentials handy to do so.

  There may also be a performance/security thing.  If you're dealing
with a LDAP->Unix gateway (making your LDAP accounts look like
traditional Unix accounts), authenticating each user doing something
would require that many different LDAP connections, and arguably the
host should cache them separately as well (since if you care to do
this you presumably have users with different accesses to your
directory).  As having one account for the host that has read-only
access, the host can just use one connection and cache for everybody.

  But whether either of these possibilities apply depends entirely on
the scenario.

-- Ben

~ Finally, powerful endpoint security that ISN'T a resource hog! ~
~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/>  ~

---
To manage subscriptions click here: 
http://lyris.sunbelt-software.com/read/my_forums/
or send an email to [email protected]
with the body: unsubscribe ntsysadmin

Reply via email to