--On Tuesday, August 31, 2004 5:23 AM -0600 katarn
<katarn@racsa.co.cr> wrote:
:-)
Ok. I've been reading:
I created a DB_CONFIG file for each database ( 35 ), then db_recover
for
each and check that it was using the new configuration using db_stat -m.
DB_CONFIG:
set_cachesize 0 1000000 0
set_lg_regionmax 262144
set_lg_bsize 2097152
set_flags DB_TXN_NOSYNC
I'm not sure you want DB_TXN_NOSYNC. There are perils in setting that
flag when you are doing things other than loading the database.
Using 2 browsers, went to the page that generates more searches , hit
refresh several consecutive times for each one, and then trying to
access
some of the suffixes with an ldap browser and I kept getting this:
=> bdb_search
bdb_locker_id: err Cannot allocate memory(12)
send_ldap_result: conn=807 op=24 p=3
send_ldap_response: msgid=25 tag=101 err=80
ber_flush: 28 bytes to sd 771
I'm wondering if this is being caused by the shear number of
database's you've established. I only run with one (all rooted at
dc=stanford,dc=edu). Is it possible to see if you have the same issue
when you have all your data under a single root?
--Quanah
--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html