[Date Prev][Date Next] [Chronological] [Thread] [Top]

RE: OpenLDAP in Production; what is "big" ? also, core-dumps.



Yes, this is "big." OpenLDAP 2.1 does as much parsing as possible on the
stack. The default stack size per thread is 4MB. Parsing a 7MB group
obviously isn't going to work. I suggest recompiling libldap_r
with -DLDAP_PVT_THREAD_STACK_SIZE=12582912 or something (12MB).

Also, as a matter of design, I believe a 130,000 member group is the wrong
way to manage access to a resource. It's already a problem at its current
size, and certainly doesn't scale to larger sizes. You should simply define a
boolean "canAuthenticate" attribute in each person's entry. This would be far
more efficient, especially since it can be indexed for very quick yes/no
determinations. The memberlist of a group is evaluated by linear search, and
this is going to kill you no matter what else you fix.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc
  Symas: Premier OpenSource Development and Support

> -----Original Message-----
> From: owner-openldap-software@OpenLDAP.org
> [mailto:owner-openldap-software@OpenLDAP.org]On Behalf Of asr@ufl.edu

> => On Wed, 26 Feb 2003 19:03:46 +0000, pod <pod@herald.ox.ac.uk> said:
> I cannot 'ldapadd' my group of 'people who might be able to
> authenticate' for
> love nor money.  It's about a 7MB group, ~130,000 members.
> It's not even
> indexed yet.