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

Re: Not able to handle more than a few thousand users?

Hello Matthew,

This looks like the problem I reported in ITS #2368. It is currently fixed in CVS HEAD, and will also be fixed in the upcoming 2.1.16 release (thanks Howard!).


--On Thursday, March 13, 2003 8:07 AM -0900 Matthew Schumacher <matt.s@aptalaska.net> wrote:


I have found that openldap 2.1.12 cannot handle my 3000 user database
with any amount of change for any length of time without some form of
index corruption.  I am using perl to query the directory but I only
change it from a php web interface.

Because I don't have the programming knowledge and time to try and
duplicate the error in a lab environment I resorted to using the ldbm
backend instead of bdb against db-4.1.25.

Hope that helps,


Mudry, Robert (N-aerotek) wrote:
I'm getting strange behavior on 2.1.12 running on HPUX 11.11 and using
Berkeley DB 4.1.25. I have a script which is adding users, one at a time,
using Perl LDAP, from another source. A thousand or so users, and it is
fine. But at around 3000 or so, LDAP suddenly becomes painfully slow and
finally gives up the ghost with an LDAP_OTHER error. At this point, the
slapd daemon froze. I restarted it, and could do searches for individual
users. But trying to dump the entire database is miserable slow--on the
order of longer than an hour.

Any ideas? Any other details which would be useful to know? I'm really at
quite a loss...

Any help appreciated greatly!


-- Quanah Gibson-Mount Senior Systems Administrator ITSS/TSS/Computing Systems Stanford University GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html