It's almost certain that this configuration is inappropriate. You're unlikely to get a decently tuned cache for a database that size in 32-bit process space.
As for your "crashes," you're likely OOM running into the DN cache issues previously discussed
http://www.openldap.org/lists/openldap-software/200708/msg00106.html
and you may find help with a 64-bit platform, OpenLDAP 2.4, or some combination thereof.
On Tue, 11 Sep 2007, Dieter Kluenter wrote:
"Sumith Narayanan" sumith.narayanan@gmail.com writes:
Hello,
We have a openldap 2.3.27 database with 42 GB of total data size . The server is running with 32 bit processor and it crashes once in 2-3 days.. As a work around for now , we are restarting the slapd process using cron job. But the restart takes 45 min to 1 hour time and sometimes more than that. Is there any way to bring up the server fast ? It is taking time to cache data before it start ?
what is your memory size? is your configuration appropriate? DB_CONFIG, slapd.conf?
-Dieter
-- Dieter Klünter | Systemberatung http://www.dkluenter.de GPG Key ID:8EF7B6C6